AI-Based Ad Evaluation and Methods for Ad Automation Using It

Abstract

An apparatus according to one embodiment collects advertising data and logistics data stored in an ERP system, extracts at least one of the number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement from the advertising data, extracts at least one of the shipment volume, inventory quantity, and expected time of inventory depletion per product from the logistics data, calculates an advertising performance indicator based on at least one of the extracted number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement, inputs at least one of the data extracted from the advertising performance indicator and logistics data into an artificial intelligence model trained based on past advertising data and logistics data stored in the ERP system, thereby generating advertising efficiency evaluation data including an efficiency level per advertisement and an advertising suitability degree per product, generates advertising planning data including advertising target products, advertising execution priorities, and advertising budget allocation information based on the advertising efficiency evaluation data, generates an advertising automation control value including at least one of an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value based on the advertising planning data, and can change at least one of the advertising exposure level, advertising budget, and advertising target products according to the advertising automation control value.

Description

Technology Field

[0001] The following embodiments relate to a technology that provides an artificial intelligence-based ad evaluation and an ad automation method using the same. Background Technology

[0003] With the recent expansion of the e-commerce market and the diversification of digital advertising platforms, companies are operating various forms of advertising, including search ads, social media ads, and content-based ads. Consequently, there is a growing demand to manage advertising efficiency based on diverse advertising data, such as impressions, clicks, conversions, advertising costs, and revenue, during the advertising process.

[0004] In particular, companies are integrating and managing product information, order information, inventory information, inbound and outbound information, and delivery information through ERP systems, and there is an increasing number of attempts to utilize advertising and logistics data stored in ERP systems together in the advertising operation process.

[0005] For example, if an increase in sales volume of a specific product leads to an inventory shortage, or conversely, if inventory accumulates for an extended period, it may be necessary to adjust the advertised products, budget, or exposure levels. Additionally, for products with a limited sales period, advertising operations may require consideration of inventory status and shipment flow.

[0006] However, conventional advertising management methods often remained at the level of individually analyzing advertising performance metrics provided by advertising platforms, and there were limitations in organically linking advertising operations with actual logistics data, such as product inventory status, shipment flow, and available sales periods.

[0007] Consequently, problems could arise where products with high advertising efficiency did not have sufficient stock, or where advertisements were not properly executed for products requiring stock depletion.

[0008] Furthermore, ad operators often adjusted advertising budgets, target products, and exposure levels based on empirical judgment after separately reviewing advertising and logistics data. In this process, it was difficult to comprehensively consider various data, and there was a possibility that the results of advertising operations could vary depending on the operator's proficiency or judgment criteria. In particular, when there were a large number of products or diverse advertising channels, the advertising operation process could become complex and management efficiency could decrease.

[0009] Meanwhile, with the recent rise of creator-based advertising utilizing social media platforms, the importance of advertising operations that consider both product and creator characteristics is increasing. However, traditionally, it has been difficult to systematically analyze content response characteristics, past advertising performance, and advertising costs by creator and incorporate them into advertising operations, and the prediction of advertising performance has often been limited.

[0010] In addition, advertising efficiency may be affected when advertisements for competing products in the same or similar product groups are executed simultaneously; however, conventional technology had limitations in adjusting advertising operation conditions by considering the timing of ad exposure, product shipment, and competitive advertising situations together.

[0011] Therefore, technology is required that can analyze advertising and logistics data together, evaluate advertising efficiency using artificial intelligence models, and automatically control advertising operation conditions based on this. Prior art literature

[0013] Republic of Korea Registered Patent No. 10-2312420 (Published Oct. 15, 2021) Republic of Korea Registered Patent No. 10-2574865 (Published Sep. 07, 2023) Republic of Korea Registered Patent No. 10-2497809 (Published Feb. 10, 2023) Republic of Korea Registered Patent No. 10-2091986 (Published Mar. 20, 2020) The problem to be solved

[0014] The embodiments aim to provide a method for evaluating advertising efficiency and performing advertising automation control based on advertising data and logistics data.

[0015] The embodiments aim to provide a method for generating advertising automation control values ​​based on inventory status and demand patterns.

[0016] The embodiments aim to provide a method for generating advertising plan data by reflecting advertising efficiency evaluation data and manager input.

[0017] The embodiments aim to provide a method for generating creator advertising execution data based on creator performance data and the characteristics of social media advertising platforms.

[0018] The embodiments aim to provide a method for calculating an advertising opportunity index based on the alignment between the time of ad exposure and the time of product shipment, and the impact of competitive advertising.

[0019] The objectives of the present invention are not limited to those mentioned above, and other unmentioned objectives will be clearly understood from the description below. means of solving the problem

[0021] According to one embodiment, in an artificial intelligence-based advertising evaluation and an advertising automation method using the same, the method comprises: collecting advertising data and logistics data stored in an ERP system; extracting at least one of the number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement from the advertising data, and extracting at least one of the shipment volume, inventory quantity, and expected time of inventory depletion per product from the logistics data; calculating an advertising performance indicator based on at least one of the extracted number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement; inputting at least one of the advertising performance indicator and the data extracted from the logistics data into an artificial intelligence model trained based on past advertising data and logistics data stored in the ERP system to generate advertising efficiency evaluation data including an efficiency level per advertisement and an advertising suitability degree per product; generating advertising planning data including advertising target products, advertising execution priorities, and advertising budget allocation information based on the advertising efficiency evaluation data; and generating an advertising automation control value including at least one of an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value based on the advertising planning data. and may include a step of changing at least one of the ad exposure level, ad budget, and ad target product according to the above ad automation control value.

[0022] The step of generating the above-mentioned advertising automation control value comprises: generating inventory status data from the above-mentioned logistics data including the current inventory quantity, shipment history, available sales period, and estimated time of inventory depletion for each product; classifying the shipment history by time units to generate product-specific shipment time series data; analyzing the shipment time series data to calculate demand pattern data including demand increase trends, demand decrease trends, and demand volatility for each product; aggregating the product-specific inventory status data and the demand pattern data by category to generate category-specific inventory levels and category-specific demand pattern data; calculating product-specific safety stock standards and product-specific overstock standards based on the above-mentioned inventory quantity, estimated time of inventory depletion, and demand volatility; using the above-mentioned inventory status data, the above-mentioned demand pattern data, the above-mentioned inventory levels, and the above-mentioned demand pattern data, and using the above-mentioned safety stock standards and overstock standards as comparison criteria to calculate inventory shortage risk and inventory overstock risk; generating conditions for increasing advertising exposure that induce priority shipment for products whose remaining available sales period is below a preset standard; and when the inventory shortage risk is above a preset standard and the advertisement included in the above-mentioned advertising efficiency evaluation data A step of generating a first ad control condition including a reduction in ad exposure and an ad budget when the efficiency level is above a preset standard; a step of generating a second ad control condition including an increase in ad exposure and an ad budget when the excess inventory risk is above a preset standard and the ad suitability for each product included in the ad efficiency evaluation data is above a preset standard; and selecting a condition according to a preset priority standard among the first ad control condition, the second ad control condition, and the ad exposure increase condition based on the remaining period of the sales period, an ad exposure level adjustment value,It may include a step of generating ad automation control values ​​including ad budget adjustment values ​​and ad target product change values.

[0023] The step of generating advertising plan data based on the above advertising efficiency evaluation data comprises: a step of calculating an advertising priority value for each target product using the advertising efficiency level and product-specific advertising suitability included in the above advertising efficiency evaluation data and the above logistics data; a step of displaying together the advertising priority value for each target product and adjustment interface elements for adjusting the advertising exposure level and advertising budget level corresponding to each product through the manager's terminal; a step of generating manager setting values ​​for the advertising exposure level and advertising budget level according to the manager's input to the adjustment interface elements; a step of calculating changes in advertising performance and inventory quantity corresponding to the manager setting values ​​using the above advertising efficiency evaluation data and the above logistics data; a step of providing the changes in advertising performance and inventory quantity through the manager's terminal so as to correspond to the manager setting values ​​for the advertising exposure level and advertising budget level; a step of setting constraints to ensure that the advertising exposure level and advertising budget are included within a preset allowable range using the inventory quantity and the expected time of inventory depletion included in the above logistics data; a step of generating a recommended control value by automatically adjusting the manager setting value according to the manager's input to satisfy the above constraints; a step of correcting the recommended control value using the manager input history and manager setting information collected from the manager's terminal; and the above It may include a step of selecting one of the recommended control value and the above-mentioned administrator setting value based on administrator selection input or preset selection criteria, and generating advertising plan data by reflecting the selected value.

[0024] Prior to the step of changing at least one of the ad exposure level, ad budget, and target product according to the ad automation control value, the method further includes a creator matching and performance prediction step based on the ad exposure level and ad budget for the target product specified by the ad automation control value, wherein the creator matching and performance prediction step utilizes creator performance data including content response data per creator collected from an SNS advertising platform and partnership performance data stored in an ERP system, and generates a candidate group of creators suitable for the target product based on the product category, content response, and past advertising performance of the target product; calculates an executable budget range and an expected exposure range, respectively, according to preset criteria using the ad exposure level and ad budget specified by the ad automation control value and the creator performance data, and generates an ad execution range per creator including these; calculates a matching score for each combination of creators and SNS advertising platforms using the candidate group of creators, the ad execution range, content exposure characteristics and content response characteristics per SNS advertising platform; and calculates an expected ad unit price for each combination of creators and SNS advertising platforms using the matching score and the ad budget included in the ad automation control value. A step of calculating estimated sales revenue and estimated profit margins for each combination of the creator and SNS advertising platform using the aforementioned advertising efficiency evaluation data and the aforementioned logistics data; a step of calculating a cost-performance index for each combination of the creator and SNS advertising platform using the aforementioned estimated advertising unit price and the aforementioned estimated sales revenue; and a step of providing an interface that displays the matching score, estimated advertising unit price, estimated sales revenue, and cost-performance index for each combination of the creator and SNS advertising platform through a manager's terminal so as to enable comparison.The method may include the steps of: providing a selection interface element for selecting a combination of a creator and an SNS advertising platform on the interface; selecting one of the combinations of a creator and an SNS advertising platform according to an administrator input or a preset selection criterion regarding the selection interface element; and combining the selected combination of a creator and an SNS advertising platform with the advertising automation control value to generate creator advertising execution data.

[0025] The step of generating the above-mentioned advertising automation control value includes the step of calculating an advertising opportunity index by combining the alignment between the advertising exposure time and the product shipment time and the advertising influence of competing products, and the step of calculating the advertising opportunity index includes the steps of extracting advertising exposure time data by advertisement from the advertising data and extracting product shipment time data by product from the logistics data; the step of generating time lag data representing the time difference from advertising exposure to shipment by comparing the advertising exposure time data by advertisement and the product shipment time data by product; the step of calculating a time alignment index representing the degree of temporal alignment between the advertising exposure time and the product shipment time based on the time lag data; the step of identifying advertising data of competing products that belong to the same category as the advertising target product specified by the above-mentioned advertising automation control value and whose advertising exposure period or sales period overlaps by more than a preset standard using the advertising data or advertising data collected from the SNS advertising platform; the step of calculating a competitive advertising intensity index representing the scale of advertising execution using at least one of the number of advertising exposures, clicks, and advertising costs from the advertising data of the competing products; and calculating a competitive influence index representing the degree of influence caused by the advertising of competing products by comparing the advertising performance indicator for the advertising target product with the competitive advertising intensity index. The method may include the steps of: calculating an advertising opportunity index defined as a value representing the potential for an expected increase in sales or an improvement in advertising efficiency resulting from the execution of an advertisement by combining the time alignment index and the competitive impact index; generating an advertising control condition that increases the level of advertising exposure or expands the advertising budget if the advertising opportunity index is greater than or equal to a preset first criterion; generating an advertising control condition that decreases the level of advertising exposure or reduces the advertising budget if the advertising opportunity index is less than or equal to a preset second criterion; and generating an advertising automation control value by reflecting the advertising control condition.

[0026] A device according to one embodiment may be combined with hardware and controlled by a computer program stored on a medium to execute the method of any one of the methods described above. Effects of the invention

[0028] The embodiments can provide a method to evaluate advertising efficiency and automatically control advertising operations based on advertising data and logistics data.

[0029] The embodiments can provide a method to automatically adjust ad exposure levels and ad budgets by reflecting inventory status and demand patterns.

[0030] The embodiments can provide a method for generating advertising plan data by reflecting advertising efficiency evaluation data and manager input together.

[0031] The embodiments can provide a method for generating advertising performance prediction and advertising execution data based on creator performance data and the characteristics of social media advertising platforms.

[0032] The embodiments can provide a method for calculating an advertising opportunity index by reflecting the alignment between the time of advertising exposure and the time of product shipment, as well as the impact of competitive advertising.

[0033] Meanwhile, the effects according to the embodiments are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description below. Brief explanation of the drawing

[0035] FIG. 1 is a schematic diagram showing the configuration of a system according to one embodiment. FIG. 2 is a flowchart illustrating an artificial intelligence-based ad evaluation and an ad automation process using the same according to an embodiment. FIG. 3 is a flowchart illustrating the process of generating advertising automation control values ​​according to one embodiment. FIG. 4 is a flowchart illustrating the process of generating advertising plan data based on advertising efficiency evaluation data according to one embodiment. FIG. 5 is a flowchart illustrating the creator matching and performance prediction process for an advertising target product according to one embodiment. FIG. 6 is a flowchart illustrating the process of calculating an advertising opportunity index according to one embodiment. FIG. 7 is an example diagram of the configuration of a device according to one embodiment. Specific details for implementing the invention

[0036] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights.

[0037] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept.

[0038] Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0039] When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or coupled with that other component, or that there may be other components in between.

[0040] The terms used in the embodiments are for illustrative purposes only and should not be interpreted as intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0041] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments pertain. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0042] In addition, when describing with reference to the attached drawings, identical components are assigned the same reference numeral regardless of drawing symbols, and redundant descriptions thereof are omitted. In describing the embodiments, if it is determined that a detailed description of related prior art could unnecessarily obscure the essence of the embodiments, such detailed description is omitted.

[0043] The embodiments can be implemented in various forms of products, such as personal computers, laptop computers, tablet computers, smartphones, smart home appliances, intelligent automobiles, kiosks, and wearable devices.

[0044] FIG. 1 is a schematic diagram showing the configuration of a system according to one embodiment.

[0045] Referring to FIG. 1, a system according to one embodiment may include a manager's terminal (100) and a device (200) that can communicate with each other through a communication network.

[0046] First, the communication network can be configured regardless of the mode of communication, such as wired or wireless, and can be implemented in various forms to enable communication between servers and between servers and terminals.

[0047] The administrator's terminal (100) may be a terminal used by an administrator who performs tasks related to advertising operations, product management, inventory management, or logistics operations. For example, the administrator's terminal (100) may be a terminal used by an advertising operations manager, a product operations manager, an inventory management manager, a logistics operations manager, or a creator collaboration manager, but this is merely an example and may be configured differently depending on the embodiment.

[0048] The administrator's terminal (100) can be connected to the device (200) via a wired or wireless communication network and can receive advertising efficiency evaluation data, advertising plan data, advertising automation control values, creator advertising execution data, and advertising opportunity index from the device (200). In addition, the administrator's terminal (100) can transmit administrator input regarding the advertising exposure level, advertising budget level, and advertising target products to the device (200).

[0049] The administrator's terminal (100) may be implemented in the form of a desktop computer, a laptop computer, a tablet terminal, a smartphone, or other information processing device. Additionally, the administrator's terminal (100) may be linked with the device (200) in a web browser-based environment or a dedicated application environment, but is not limited thereto.

[0050] The manager's terminal (100) can visually receive various information linked to advertising data and logistics data. For example, the manager's terminal (100) can display advertising priority values ​​by advertising target product, changes in advertising performance, changes in inventory quantity, advertising exposure levels, advertising budget levels, estimated advertising unit prices by creator, estimated sales revenue, and performance index relative to costs. Through this, the manager can check or adjust advertising operation strategies by considering the advertising operation status and product operation status together.

[0051] The administrator's terminal (100) may be configured to perform all or part of the computational functions, storage / reference functions, input / output functions, and control functions that a normal computer has, and the administrator's terminal (100) may be configured to communicate with the device (200) via wired or wireless communication.

[0052] The administrator's terminal (100) may be connected to a website operated by a person or organization providing a service using the device (200), or may have an application developed and distributed by a person or organization providing a service using the device (200) installed. The administrator's terminal (100) may be linked with the device (200) through a website or application. Additionally, the administrator's terminal (100) may access the device (200) through a web page or application provided by the device (200).

[0053] The device (200) may be a proprietary server owned by the entity or organization providing the service using the device (200), or it may be implemented in the form of a cloud server or a distributed processing environment. Additionally, the device (200) may be configured to perform all or part of the computational functions, storage / reference functions, input / output functions, and control functions that a conventional computer possesses.

[0054] The device (200) can be configured to communicate wirelessly or via wired connection with the manager's terminal (100).

[0055] The device (200) may be an information processing device for collecting advertising data and logistics data stored in an ERP system and generating advertising efficiency evaluation data, advertising planning data, and advertising automation control values ​​based thereon. The device (200) may be connected to communicate with the manager's terminal (100) and may perform data analysis related to advertising operations, advertising automation control, and advertising performance prediction functions.

[0056] The device (200) can extract at least one of the number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement from advertising data, and can extract at least one of the shipment volume, inventory quantity, shipment history, and estimated time of inventory depletion per product from logistics data. Additionally, the device (200) can calculate advertising performance indicators based on the extracted data and generate advertising efficiency evaluation data using an artificial intelligence model trained on past advertising data and logistics data stored in an ERP system.

[0057] The device (200) can generate advertising plan data including advertising target products, advertising execution priorities, and advertising budget allocation information based on advertising efficiency evaluation data. Additionally, the device (200) can generate advertising automation control values ​​including advertising exposure level adjustment values, advertising budget adjustment values, and advertising target product change values ​​by analyzing inventory status data, shipment time series data, and demand pattern data.

[0058] The device (200) can provide advertising priority values, changes in advertising performance, changes in inventory quantity, and recommendation control values ​​for each advertising target product through the manager's terminal (100), and can correct advertising plan data or adjust advertising automation control values ​​according to the manager's input. In addition, the device (200) can generate creator advertising execution data by analyzing creator performance data and SNS advertising platform characteristics, and can calculate an advertising opportunity index by reflecting the consistency between the advertising exposure time and the product shipment time and the impact of competing advertisements.

[0059] The device (200) may include one or more processors and memory to store and execute programs or data for performing each step of the present invention. The process of artificial intelligence-based ad evaluation and ad automation using the same may be performed through a processor included in the device (200), and in some cases, may be processed in an artificial neural network or machine learning-based manner.

[0060] Additionally, the device (200) can communicate wirelessly or via wired connection with an SNS advertising platform or website and can collect information related to advertising data, content response data, or creator performance data. For example, the SNS advertising platform may include, but is not limited to, a content sharing-based platform, a video-based platform, a community-based platform, or an advertising provision platform.

[0061] Meanwhile, for convenience of explanation, only one administrator terminal (100) is shown in FIG. 1 and the following description, but the number of terminals may vary depending on the embodiment. In addition, multiple administrator terminals (100) may be simultaneously connected to the device (200) within the range allowed by the processing capacity of the device (200).

[0062] Through this, the device (200) can organically link advertising data and logistics data to analyze advertising efficiency and perform advertising automation control that comprehensively reflects inventory status, outbound flow, advertising performance and creator performance.

[0063] In the present invention, Artificial Intelligence (AI) refers to a technology that imitates human learning ability, reasoning ability, and perceptual ability, and implements them on a computer, and may include concepts such as machine learning and symbolic logic. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input data on its own. AI technology can analyze input data as a machine learning algorithm, learn from the results of the analysis, and make judgments or predictions based on the results of the learning. Furthermore, technologies that mimic the functions of the human brain, such as cognition and judgment, by utilizing machine learning algorithms can also be understood as falling within the category of AI. For example, technological fields such as linguistic understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control may be included.

[0064] Machine learning can refer to the process of training neural network models using experience in processing data. It implies that through machine learning, computer software improves its own data processing capabilities. A neural network model is constructed by modeling the correlations between data, and these correlations can be expressed by multiple parameters. A neural network model extracts and analyzes features from given data to derive correlations between them; machine learning can be defined as the process of optimizing the model's parameters by repeating this process. For example, a neural network model can learn the mapping (correlation) between inputs and outputs for data given as input-output pairs. Alternatively, even when only input data is provided, a neural network model can derive regularities between the given data and learn those relationships.

[0065] An artificial intelligence learning model or neural network model can be designed to implement the structure of the human brain on a computer and may include multiple network nodes that have weights and simulate neurons of a human neural network. The multiple network nodes may have interconnected relationships by simulating the synaptic activity of neurons, where neurons exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes may be located in layers of different depths and exchange data according to convolutional connections. The artificial intelligence learning model may be, for example, an Artificial Neural Network (ANN) or a Convolutional Neural Network (CNN). As an embodiment, the artificial intelligence learning model may be machine learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning may include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Ada-boost, Perceptron, Genetic Programming, and Clustering.

[0066] Among these, CNNs are a type of multilayer perceptron designed to use minimal preprocessing. CNNs consist of one or more convolutional layers and standard artificial neural network layers stacked on top, additionally utilizing weights and pooling layers. Thanks to this structure, CNNs can fully utilize two-dimensional input data. Compared to other deep learning architectures, CNNs demonstrate good performance in both image and audio fields. CNNs can also be trained using standard backpropagation. CNNs have the advantage of being easier to train than other feedforward artificial neural network techniques and using a small number of parameters.

[0067] Convolutional networks are neural networks comprising sets of nodes with bounded parameters. Many computer vision tasks have been significantly improved, driven by the increased size of available training data and the availability of computational power, combined with algorithmic advancements such as discriminative linear units and dropout training. In the case of massive datasets, such as those available for many tasks today, outfitting is not critical, and increasing the network size improves test accuracy. Optimal utilization of computing resources becomes a limiting factor. To address this, distributed, scalable implementations of deep neural networks can be employed.

[0068] FIG. 2 is a flowchart illustrating an artificial intelligence-based ad evaluation and an ad automation process using the same according to an embodiment.

[0069] Referring to FIG. 2, first, in step S201, the device (200) can collect advertising data and logistics data stored in an ERP system.

[0070] In the present invention, an ERP system may refer to a system that stores and manages data related to a company's products, orders, inventory, inbound and outbound, delivery, and advertising operations.

[0071] In this case, the ERP system can be implemented as a single integrated system, and the advertising management module, logistics management module, inventory management module, and delivery management module can be implemented in a form where they are interconnected.

[0072] In the present invention, advertising data is data generated or stored during the advertising operation process and may include at least one of the number of impressions, clicks, advertising costs, conversions, sales revenue, advertising channel information, advertising target product information, advertising execution period information, and advertising campaign identification information.

[0073] In the present invention, logistics data is data related to the shipment, inventory, receipt, delivery, or availability of goods, and may include at least one of the shipment quantity per product, inventory quantity, estimated time of inventory depletion, receipt and shipment history, product category information, product identification information, and availability period information.

[0074] That is, the device (200) does not simply collect advertising data and logistics data stored in the ERP system separately, but can collect necessary data items together so that the products targeted for advertising operations and the actual inventory and shipment flow can correspond to each other.

[0075] For example, the device (200) can collect advertising data and logistics data based on at least one of product identification information, product name, product category, brand information, advertising campaign identification information, advertising channel information and period information.

[0076] Additionally, the device (200) can collect daily ad impressions, clicks, ad costs, conversions, and sales revenue as advertising data for a specific product from an ERP system, and can also collect daily shipment volume, current inventory quantity, inflow and outflow history, and estimated time of inventory depletion as logistics data for the same product.

[0077] Even if advertising data and logistics data exist in different storage areas, the device (200) can collect the two data in a corresponding form based on product identification information, category information, or period information.

[0078] For example, if advertising data is stored in units of advertising campaigns and logistics data is stored in units of products, the device (200) can match the advertising data and the logistics data with each other based on advertising target product information connected to the advertising campaign.

[0079] The device (200) can unify the standard period of advertising data and logistics data during the collection process.

[0080] For example, if advertising data is stored on a daily basis and logistics data is stored on a weekly basis, the device (200) can aggregate the daily values ​​of the advertising data on a weekly basis or convert the weekly values ​​of the logistics data into reference values ​​that can be used for daily analysis.

[0081] At this time, the reference period may be set on a daily, weekly, monthly, or advertising campaign operation period basis, but is not limited thereto and may be set differently depending on the embodiment.

[0082] Through this, the device (200) can obtain basic data that can reflect both the results of advertising operations and the actual status of product operations. The device (200) can then organically utilize advertising data and logistics data in the process of calculating advertising performance indicators, generating advertising efficiency evaluation data, generating advertising plan data, and generating advertising automation control values.

[0083] In step S202, the device (200) can extract at least one of the number of impressions, clicks, advertising costs, conversions, and sales per advertisement from the advertising data, and at least one of the number of shipments, inventory quantity, and estimated time of inventory depletion per product from the logistics data.

[0084] That is, the device (200) can selectively extract key data items that are actually used for advertising efficiency evaluation and advertising automation control from among the advertising data and logistics data collected from the ERP system.

[0085] In this context, ad impressions may refer to the number of times a specific ad is displayed on a user's device, and clicks may refer to the number of times a user selects an ad. Additionally, advertising costs may refer to the expenses incurred in executing a specific ad, and conversions may refer to the number of times a target action—such as placing an order, making a purchase, signing up, saving to a cart, or entering a product detail page—occurred through the ad. Furthermore, revenue may refer to the amount of product sales generated after the ad execution.

[0086] Additionally, the shipment volume may refer to the actual quantity of goods shipped during a specific period, and the inventory quantity may refer to the quantity of goods currently held in stock. The estimated time of inventory depletion may refer to the point in time when the inventory is expected to be completely exhausted based on the current inventory quantity and recent shipment trends.

[0087] That is, the device (200) can form a data structure capable of analyzing advertising performance and product operation status together by extracting items necessary for analyzing advertising operation efficiency from advertising data representing advertising operation status and logistics data representing actual product operation status.

[0088] For example, the device (200) can extract daily ad impressions, ad clicks, ad costs, conversions, and sales for a specific product from ad data stored in an ERP system. At this time, the device (200) can classify and extract ad data according to criteria by ad channel, ad campaign, product, or period.

[0089] For example, the device (200) can separate and extract advertising data generated from search advertising channels and advertising data generated from social media advertising platforms for the same product, and can form a data structure that can analyze differences in advertising performance by advertising campaign by extracting advertising costs and sales revenue for each advertising campaign for the same product.

[0090] Additionally, the device (200) can extract the shipment volume of a specific product, the current inventory quantity, and the expected time of inventory depletion together from logistics data. For example, the device (200) can extract daily shipment volume data for each product over the past 30 days and can also extract current inventory quantity data together.

[0091] The device (200) can extract data to calculate the expected time of inventory depletion based on recent shipment flow. For example, if the average shipment volume over the past 7 days is 100 units per day and the current inventory quantity is 300 units, the device (200) can calculate the expected time of inventory depletion as about 3 days later.

[0092] Conversely, if the average outflow over the past 30 days is very low and the current inventory quantity is excessive, the device (200) can extract data for analysis to determine the likelihood of maintaining inventory for a long period.

[0093] In this case, the estimated time of inventory depletion may be calculated using only the simple average value, or it may be calculated by considering the recent shipment growth rate, daily sales deviation, seasonal information, whether advertising was executed, or promotion period information together.

[0094] For example, the device (200) can adjust the expected time of stock depletion by reflecting the recent increase in shipments of the same season when a specific product is a summer season product.

[0095] Additionally, the device (200) can perform data normalization to match the reference period between the advertising data and the logistics data. For example, if the advertising data is stored in hourly units and the logistics data is stored in daily units, the device (200) can aggregate the hourly advertising data based on daily units for use.

[0096] Conversely, if logistics data is stored on a weekly basis, the device (200) can convert the weekly data into a daily analysis standard and use it.

[0097] In addition, the device (200) can remove or correct missing data or abnormal data during the extraction process. For example, if there are ad impressions but the number of clicks is recorded as abnormally excessively high, the device (200) can exclude or correct the data using a preset abnormal data standard.

[0098] At this time, the pre-set abnormal data criteria may include, but are not limited to, a deviation range relative to the average, a maximum allowable growth rate per period, a normal range per advertising channel, or a normal range per product category, and may be set differently depending on the embodiment.

[0099] Through this, the device (200) can obtain reliable core data that can be used for advertising efficiency evaluation and advertising automation control. By forming a data base that can simultaneously reflect the advertising operation status and the actual product operation status, the device (200) can improve the accuracy of calculating advertising performance indicators and generating advertising efficiency evaluation data thereafter.

[0100] In step S203, the device (200) can calculate an advertising performance indicator based on at least one of the extracted ad impressions, clicks, ad costs, conversions, and sales revenue.

[0101] That is, the device (200) can analyze the results of advertising operations included in the advertising data and calculate an advertising performance indicator that quantitatively represents the level of performance for each advertisement.

[0102] In the present invention, advertising performance indicators can be defined as data for evaluating advertising operation results based on numerical values, and may refer to evaluation values ​​calculated using the relationships between the number of impressions, clicks, advertising costs, conversions, and sales revenue for each advertisement.

[0103] For example, advertising performance indicators may include at least one of revenue relative to advertising costs, click-through rate, conversion rate, conversion performance relative to advertising costs, conversion performance relative to impressions, ROAS (Return On Ad Spend), CPC (Cost Per Click), CPM (Cost Per Mille), CTR (Click Through Rate), CVR (Conversion Rate), and CPS (Cost Per Sale).

[0104] The device (200) can calculate the click-through rate using the number of impressions and clicks per advertisement included in the advertisement data. In this case, the click-through rate may be a value calculated by dividing the number of ad clicks by the number of ad impressions, and may be used as an indicator of how much actual user interest a specific advertisement has induced.

[0105] For example, if the number of ad impressions for a specific ad is 100,000 and the number of ad clicks is 5,000, the device (200) can calculate the click rate as 5%.

[0106] Additionally, the device (200) can calculate a conversion rate using the number of ad clicks and conversions. In this case, the conversion rate may refer to the ratio of actual purchases or target actions that resulted from an ad click.

[0107] For example, if the number of ad clicks for a specific ad is 5,000 and the number of conversions is 250, the device (200) can calculate the conversion rate as 5%.

[0108] The device (200) can calculate a sales indicator relative to advertising costs using advertising costs and sales revenue. For example, if the advertising cost used for a specific advertisement is 1 million won and the sales revenue generated through the advertisement is 5 million won, the device (200) can calculate the ROAS as 500%.

[0109] Additionally, the device (200) can calculate CPC using advertising costs and click counts, and can calculate CPM using advertising costs and ad impression counts.

[0110] For example, if the advertising cost is 500,000 won and the number of ad clicks is 10,000, the device (200) can calculate the CPC representing the cost per click. In addition, if the number of ad impressions is 200,000, the device can also calculate the CPM representing the cost per 1,000 ad impressions.

[0111] The device (200) can calculate advertising performance indicators based on advertising channels, advertising campaigns, products, or periods, respectively.

[0112] For example, the device (200) can calculate advertising performance indicators in search advertising channels and advertising performance indicators in social media advertising platforms separately for the same product. Additionally, the device (200) may calculate advertising performance indicators from the start time to the end time of a specific advertising campaign, and may calculate advertising performance indicators on a daily or weekly basis.

[0113] The device (200) can apply weights according to specific criteria in the process of calculating advertising performance indicators.

[0114] For example, the device (200) may apply a higher weight to sales-based metrics or to conversion-rate-based metrics. Additionally, the device (200) may set different reflection ratios for advertising performance metrics depending on the characteristics of each product category.

[0115] For example, for product categories with high repeat purchase rates, a higher weight can be applied to the conversion rate, while for product categories where brand awareness is important, a higher weight can be applied to ad impressions or click-through rates.

[0116] In this case, the weights may be set based on advertising channel characteristics, product category characteristics, advertising objectives, sales strategies, or past advertising performance data, but are not limited thereto and may be set differently depending on the embodiment.

[0117] Additionally, the device (200) can exclude or correct abnormal advertising data during the process of calculating advertising performance indicators.

[0118] For example, if the number of ad clicks increases abnormally during a specific time period but the actual number of conversions is almost non-existent, the device (200) can exclude or correct the data in the process of calculating ad performance indicators by comparing it with a reference value for determining the data as abnormal ad data.

[0119] For example, if excessive repeated clicks occur within a short period of time on a specific advertising channel and the conversion rate is below a preset standard, the device (200) can classify the advertising data as abnormal data and reduce the reflection rate when calculating advertising performance indicators.

[0120] At this time, the pre-set criteria may include, but are not limited to, an average click growth rate, a normal click range per ad channel, a conversion rate threshold value, a standard for the number of repeated clicks by the same user, or a normal range per ad type.

[0121] Through this, the device (200) can generate advertising performance indicators that reflect actual advertising efficiency, going beyond the level of simply collecting advertising operation results. The device (200) can analyze advertising performance more precisely by reflecting the characteristics of each advertising channel, the characteristics of each product, and the purpose of advertising operation together, and can provide reliable input data in the subsequent process of generating advertising efficiency evaluation data.

[0122] In step S204, the device (200) can generate advertising efficiency evaluation data including an efficiency level per advertisement and an advertising suitability per product by inputting at least one of the advertising performance indicators and data extracted from logistics data into an artificial intelligence model trained based on past advertising data and logistics data stored in an ERP system.

[0123] That is, the device (200) can generate advertising efficiency evaluation data to evaluate advertising operation efficiency and product-specific advertising execution suitability by analyzing together advertising performance indicators representing the current advertising operation status and logistics data representing the actual product operation status.

[0124] In the present invention, advertising efficiency evaluation data can be defined as data for analyzing advertising operation efficiency by reflecting both the results of advertising operations and the status of product operations, and may include the efficiency level for each advertisement and the suitability of the advertisement for each product.

[0125] In this context, the efficiency level per advertisement may be an evaluation value indicating the extent to which a specific advertisement generates advertising performance relative to its advertising cost. Additionally, the advertising suitability per product may be an evaluation value indicating the degree to which a specific product is suitable as an advertising target, considering its current inventory status, shipment flow, available sales period, and advertising performance status.

[0126] The device (200) can train an artificial intelligence model using past advertising data and logistics data stored in an ERP system.

[0127] For example, the device (200) can use data on the number of ad impressions, clicks, ad costs, conversions, and sales revenue for each product during a past period as training data, and at the same time, information on the shipment volume, inventory quantity, estimated time of inventory depletion, and sales period for the same product can also be used as training data.

[0128] In addition, the device (200) can use advertising operation cases with excellent advertising performance and product operation cases where inventory shortage or inventory accumulation occurred together as training data.

[0129] For example, the device (200) can use as training data product cases where advertising performance was excellent but additional sales were difficult due to a shortage of stock, and conversely, can use as training data product cases where advertising performance was low but the stock accumulation continued for a long period.

[0130] Through this, the device (200) can configure an artificial intelligence model that reflects not only simple advertising performance but also the actual product operation status.

[0131] The device (200) can input at least one of the data extracted from advertising performance indicators and logistics data into a learned artificial intelligence model.

[0132] For example, the device (200) can input click-through rate, conversion rate, sales relative to advertising costs, and advertising performance data by advertising channel as advertising performance indicators. In addition, the device (200) can input information on the shipment volume by product, current inventory quantity, estimated time of inventory depletion, and available sales period as logistics data.

[0133] The device (200) can calculate the efficiency level for each advertisement based on input advertising performance indicators. For example, the device (200) can compare whether at least one of the click-through rate, conversion rate, and sales-to-advertising cost indicators is greater than or equal to a preset standard, and calculate the efficiency level for each advertisement based on the comparison result.

[0134] For example, if the click-through rate of a specific advertisement is higher than the average click-through rate of the same advertising channel, the conversion rate is higher than the average conversion rate by product category, and the sales indicator relative to advertising costs is higher than a preset standard, the device (200) can calculate a high level of efficiency for that advertisement. Conversely, if the number of impressions of a specific advertisement is high but the click-through rate or conversion rate is low and the sales indicator relative to advertising costs is lower than a preset standard, the device (200) can calculate a low level of efficiency for that advertisement.

[0135] At this time, the pre-set criteria may be based on average performance criteria by advertising channel, average conversion rate criteria by product category, sales criteria relative to advertising costs, or past advertising operation results, but are not limited thereto and may be set differently depending on the embodiment.

[0136] Additionally, the device (200) can calculate the advertising suitability for each product by reflecting the advertising performance indicators and data extracted from logistics data together. For example, the device (200) inputs the advertising performance indicators, inventory quantity, shipment quantity, and estimated time of inventory depletion together into an artificial intelligence model, and can calculate the degree to which a specific product is suitable as an advertising target using the output value of the artificial intelligence model.

[0137] For example, if the advertising performance indicator of a specific product is above a preset standard, the current inventory quantity is within a range capable of handling the expected shipment volume resulting from the advertising execution, and the expected time of inventory depletion is not excessively imminent, the device (200) can calculate a high product-specific advertising suitability for the product.

[0138] Conversely, even if the advertising performance indicators of a specific product are excellent, if the current inventory quantity is insufficient or the expected time of inventory depletion is imminent compared to the recent trend of increasing shipment volume, the device (200) can calculate the product-specific advertising suitability of the product as low.

[0139] In addition, even if the advertising performance indicator of a specific product is at an average level, if the current inventory quantity is excessive and the shipment quantity remains low for a long period, the device (200) can adjust the advertising suitability for each product by reflecting the need to deplete the inventory.

[0140] In addition, for products with a limited sales period, the advertising suitability for each product can be adjusted by considering the remaining sales period.

[0141] For example, if the remaining period of the sales period is short and the current inventory quantity is large, the device (200) can calculate a high advertising suitability for each product to induce inventory depletion.

[0142] Conversely, even if the remaining sales period is sufficient, if the inventory quantity is very insufficient, the advertising suitability for each product may be calculated as low.

[0143] In this way, the efficiency level for each advertisement is calculated primarily based on advertising performance indicators, and the suitability of the advertisement for each product can be calculated by reflecting the inventory quantity, shipment volume, and estimated time of inventory depletion extracted from the advertising performance indicators and logistics data. Accordingly, the device (200) can evaluate the efficiency of the advertisement itself and the operational feasibility of the advertised product separately.

[0144] The device (200) may also generate advertising efficiency evaluation data by analyzing the flow of changes in advertising performance indicators and logistics data together.

[0145] For example, if the recent ad click-through rate is increasing but the shipment volume growth rate remains low, the device (200) can analyze that the actual purchase conversion efficiency is low compared to the ad inflow. Conversely, if the ad click-through rate and the shipment volume growth rate increase together, the ad operation status and the product sales flow can be analyzed as being linked to each other.

[0146] Through this, the device (200) can generate advertising efficiency evaluation data that reflects both advertising performance and the actual product operation status. Since the device (200) can distinguish products with a high risk of stock shortage even if advertising performance is excellent, or identify products with a high potential for improving advertising operation efficiency despite being in a state of stock accumulation, it can derive a more sophisticated advertising operation direction during the subsequent process of generating advertising plan data.

[0147] In step S205, the device (200) can generate advertising plan data including advertising target products, advertising execution priorities, and advertising budget allocation information based on advertising efficiency evaluation data.

[0148] That is, the device (200) can generate advertising plan data to determine the advertising target product and the direction of advertising operation to be applied to actual advertising operations using advertising efficiency evaluation data.

[0149] In the present invention, advertising plan data is data for defining an advertising operation plan and may include information on the target product for advertising, the priority of advertising execution, and the allocation of the advertising budget.

[0150] In this case, the product targeted for advertising may refer to the product set as the target for advertising execution, and the advertising execution priority may be a value indicating which product among multiple products will be prioritized for advertising execution.

[0151] In addition, advertising budget allocation information may be information indicating how the total advertising budget used for advertising operations will be distributed by product, advertising channel, or advertising campaign.

[0152] That is, the advertising plan data may be planning data for defining direction and operational standards for advertising operations, and subsequently, the device (200) may generate advertising automation control values ​​to change actual advertising operation conditions based on the advertising plan data.

[0153] The device (200) can select an advertising target product using the advertising efficiency level and product-specific advertising suitability included in the advertising efficiency evaluation data.

[0154] For example, the device (200) can select a product as an advertising target product in which the efficiency level per advertisement is above a preset standard and the advertising suitability level per product is above a preset standard. Even if the efficiency level per advertisement is high, if the current stock quantity is very low or the expected time of stock depletion is imminent, the product may be excluded from the advertising target products or have its priority set low.

[0155] Additionally, the device (200) may include products that are in a state of inventory accumulation for a long period of time, even if the advertising performance indicators are relatively low, as advertising target products.

[0156] For example, if the recent shipment volume of a specific product is very low and the current inventory quantity is excessive, the device (200) may include the product as an advertising target product to induce the depletion of inventory.

[0157] At this time, the pre-set criteria may include, but are not limited to, average performance criteria by advertising channel, average advertising efficiency criteria by product category, inventory quantity criteria, criteria for the expected time of inventory depletion, or criteria for the sales period.

[0158] The device (200) can calculate the priority of advertising execution for each advertising target product.

[0159] For example, the device (200) can calculate the priority of ad execution by considering the efficiency level per ad, the suitability of the ad per product, the recent shipment growth rate, the inventory quantity, and the sales period information together.

[0160] In addition, the device (200) can adjust the priority of ad execution by reflecting the difference in ad performance by ad channel.

[0161] For example, if a specific product shows a high conversion rate in a search advertising channel but a low conversion rate in an SNS advertising platform, the device (200) can set a higher priority for advertising execution in the search advertising channel. If a specific product shows a high advertising exposure diffusion effect in an SNS advertising platform, the device (200) can set a higher priority for advertising operation based on the SNS advertising platform.

[0162] In addition, the device (200) may apply different criteria for calculating the priority of ad execution depending on the purpose of ad operation.

[0163] For example, if the goal is short-term sales growth, a higher weighting can be applied to conversion rates or sales-to-advertising cost metrics. If the goal is to induce inventory depletion, a higher weighting can be applied to current inventory levels, recent shipment trends, and the expected time of inventory depletion.

[0164] The device (200) can generate advertising budget allocation information based on advertising execution priority.

[0165] For example, the device (200) can allocate a higher advertising budget to products with a higher priority for advertising execution, and can set a lower advertising budget allocation ratio to products with a lower priority for advertising execution.

[0166] In addition, the device (200) can adjust the advertising budget allocation ratio by reflecting the difference in advertising performance by advertising channel.

[0167] For example, if the sales indicator relative to advertising costs in a specific advertising channel is consistently maintained at a high level, the device (200) can increase the advertising budget allocation ratio for that advertising channel. Conversely, in the case of an advertising channel where the number of ad clicks is high but the actual conversion rate is maintained at a low level, the advertising budget allocation ratio can be decreased.

[0168] In addition, the device (200) can also consider the total advertising operation budget limit during the advertising budget allocation process.

[0169] For example, the device (200) can adjust the advertising budget allocation ratio by product within the total advertising operation budget range and can apply an advertising budget upper limit standard so that the advertising budget is not excessively concentrated on a specific product or a specific advertising channel.

[0170] At this time, the advertising budget limit criteria may include, but are not limited to, the maximum advertising budget ratio by product category, the maximum advertising execution ratio by advertising channel, or the maximum advertising execution amount by advertising campaign.

[0171] The device (200) can also consider advertising operation stability during the process of generating advertising plan data.

[0172] For example, for products where advertising performance metrics have temporarily increased sharply, the priority of advertising execution can be adjusted by analyzing the volatility of advertising performance over a recent period.

[0173] In addition, if the fluctuation range of advertising performance indicators has been very large during a recent specific period, the increase in the advertising budget may be limited.

[0174] For example, if the sales indicator relative to advertising costs has increased rapidly over the past three days but the advertising performance during the previous period was unstable, the device (200) can limit the advertising budget growth rate to below a preset limit range.

[0175] At this time, the limit range may be set according to the allowable variation range by product category, the allowable budget increase rate by advertising channel, or the advertising operation policy, but is not limited thereto and may be set differently depending on the embodiment.

[0176] Additionally, the device (200) can provide advertising plan data through the manager's terminal (100).

[0177] For example, the device (200) can display a list of products to be advertised, advertising execution priority, advertising budget allocation information by product, and advertising operation plan information by advertising channel on the manager's terminal (100).

[0178] In addition, the device (200) provides advertising plan data along with advertising efficiency evaluation data, thereby enabling the manager to check the direction of advertising operations and the status of advertising budget allocation together.

[0179] In this way, the advertising plan data can be used as planning information to define the direction of advertising operations, and the device (200) can generate advertising automation control values ​​to subsequently change the actual advertising exposure level, advertising budget, and advertising target products based on the advertising target products, advertising execution priority, and advertising budget allocation information included in the advertising plan data.

[0180] Through this, the device (200) can generate an advertising operation plan that takes into account both the advertising performance status and the actual product operation status. Since the device (200) can focus advertising on products with high advertising efficiency while also taking into account inventory shortages or inventory accumulation, it can simultaneously improve advertising operation efficiency and product operation efficiency.

[0181] For a detailed explanation regarding this, please refer to Fig. 4.

[0182] In step S206, the device (200) can generate an advertising automation control value including at least one of an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value based on advertising plan data.

[0183] That is, the device (200) can generate advertising automation control values ​​to reflect the advertising target products, advertising execution priorities, and advertising budget allocation information included in the advertising plan data into actual advertising operation conditions.

[0184] In the present invention, the advertising automation control value may be defined as control data for automatically changing or adjusting the advertising operation status, and may include at least one of an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value.

[0185] In this case, the ad exposure level adjustment value may refer to a value for adjusting the ad exposure frequency, the number of ad deliveries, the ad exposure priority, or the ad exposure range.

[0186] Additionally, the advertising budget adjustment value may refer to a value for increasing or decreasing the budget used for advertising execution, and the advertising target product change value may refer to a value for adding, excluding, or replacing products set as advertising targets.

[0187] In other words, if advertising plan data is planning data intended to define the direction and standards of advertising operations, advertising automation control values ​​may be execution control data applied to the actual advertising platform or operating environment to change ad impression levels, advertising budgets, and target products.

[0188] For example, the advertising plan data may include operational direction information to maintain a high priority for advertising execution of a specific product and to increase the advertising budget allocation ratio, and the device (200) may generate an advertising automation control value including an advertising exposure level adjustment value to increase the actual advertising exposure frequency or an advertising budget adjustment value to increase the actual advertising budget based thereon.

[0189] The device (200) can generate an ad automation control value based on the ad execution priority included in the ad plan data.

[0190] For example, the device (200) can generate an ad automation control value that increases the ad exposure level adjustment value or increases the ad budget adjustment value for products with a high ad execution priority. Conversely, for products with a low ad execution priority, it can generate an ad automation control value that decreases the ad exposure level adjustment value or decreases the ad budget adjustment value.

[0191] Additionally, the device (200) can generate an advertising automation control value by considering the efficiency level per advertisement and the suitability of the advertisement per product included in the advertising efficiency evaluation data.

[0192] For example, for products with high ad efficiency and high ad suitability, ad automation control values ​​including increased ad impressions and increased ad budgets can be generated. Conversely, even if ad efficiency is high, if the current inventory is very low or the expected time of inventory depletion is imminent, ad automation control values ​​including decreased ad impressions or decreased ad budgets can be generated.

[0193] Additionally, the device (200) can generate an advertising target product change value for products that are in a stock accumulation state for a long period of time.

[0194] For example, if the shipment volume is maintained below a preset standard for a recent period and the current inventory quantity continues to increase, the device (200) can generate an advertising target product change value to include the product in the advertising target products.

[0195] At this time, the pre-set criteria may include, but are not limited to, criteria for average shipment volume by product category, criteria for inventory growth rate, criteria for sales period, or criteria for advertising operation policy.

[0196] The device (200) may also generate advertising automation control values ​​by reflecting the advertising operation characteristics of each advertising channel.

[0197] For example, in the case of search advertising channels, an increase in the advertising budget can be generated for products with high conversion-rate-based advertising performance, and in the case of social media advertising platforms, an increase in the advertising exposure level can be generated for products with a high advertising exposure diffusion effect.

[0198] Additionally, the device (200) can generate an ad budget reduction value when the number of ad clicks on a specific ad channel has increased but the actual increase in conversions is minimal.

[0199] For example, if the ad click-through rate is above a preset standard or the sales indicator relative to the ad cost is below a preset standard, the device (200) can generate an ad budget reduction value by comparing it with a standard for determining a state of reduced ad operation efficiency.

[0200] The device (200) can also consider advertising operation stability during the process of generating advertising automation control values.

[0201] For example, even if advertising performance indicators have increased rapidly in the short term, the increase in the advertising budget can be limited by analyzing the fluctuations in advertising performance over a recent period.

[0202] Additionally, the device (200) may limit the maximum range of change of the ad exposure level adjustment value or the ad budget adjustment value to prevent the ad operation status from changing rapidly.

[0203] For example, you can create ad automation controls to ensure that the ad budget growth rate does not exceed a preset maximum daily growth rate, even if the ad performance of a specific product increases rapidly.

[0204] At this time, the maximum increase ratio may be set according to the allowable increase range per product category, the operation policy per advertising channel, or the advertising operation stability standard, but is not limited thereto and may be set differently depending on the embodiment.

[0205] Additionally, the device (200) can generate advertising automation control values ​​while adjusting priority conflicts between advertising target products.

[0206] For example, if multiple products all have a high priority for advertising execution, the device (200) can adjust the advertising budget allocation ratio by comparing advertising efficiency evaluation data, current inventory status, available sales period, and advertising operation performance by advertising channel together.

[0207] Additionally, the device (200) may use execution unit standards applicable to an actual advertising operation environment when generating advertising automation control values. For example, the device (200) may adjust the frequency of ad impressions on an hourly, daily, or ad campaign basis, and the ad budget may be adjusted on a daily budget per ad channel, a weekly budget, or a total budget per ad campaign basis.

[0208] Additionally, the change value for advertised products can be generated in the form of adding new advertised products, excluding existing advertised products, or restricting advertised products to specific advertising channels only.

[0209] Additionally, the device (200) can provide the generated result through the manager's terminal (100) after generating the advertising automation control value.

[0210] For example, the device (200) can display information on the planned increase or decrease in the level of ad exposure, information on the planned adjustment of the ad budget, and information on the planned change of the target product for the ad on the manager's terminal (100).

[0211] In addition, the device (200) can also provide the cause of the generation of the advertising automation control value.

[0212] For example, the device (200) may display information such as an increase in the efficiency level per advertisement, maintenance of a stable inventory state, and an increase in the recent shipment volume as reasons for an increase in the advertising budget of a specific product. Conversely, it may also display a state of insufficient inventory or an imminent time of expected inventory depletion as reasons for a decrease in the advertising exposure of a specific product.

[0213] In this way, the advertising automation control value can be used as execution data to apply operational direction information included in the advertising plan data to the actual advertising operation environment, and the device (200) can automatically change the actual advertising exposure level, advertising budget, and advertising target products using the advertising automation control value.

[0214] Through this, the device (200) can automatically reflect advertising plan data into actual advertising operation conditions. Since the device (200) can automatically adjust advertising operation conditions by simultaneously considering the advertising performance status and the actual product operation status, it can improve both advertising operation efficiency and inventory operation efficiency.

[0215] For a detailed explanation regarding this, refer to Fig. 3.

[0216] In step S207, the device (200) can change at least one of the ad exposure level, ad budget, and ad target product according to the ad automation control value.

[0217] That is, the device (200) can automatically change the advertising operation conditions by reflecting the generated advertising automation control value in the actual advertising operation environment.

[0218] In the present invention, the level of ad exposure may include the frequency of ad exposure, the number of ad transmissions, the priority of ad exposure, the range of ad exposure, or the range of target users for ad exposure.

[0219] In addition, the advertising budget may refer to the cost limit used for advertising execution, and the advertised product may refer to individual products or product groups set as targets for advertising execution.

[0220] The device (200) can change the ad exposure level using an ad exposure level adjustment value included in the ad automation control value.

[0221] For example, for products with a high priority for advertising execution, the number of ad broadcasts can be increased or the frequency of ad exposure can be raised.

[0222] Additionally, the device (200) can adjust the ad exposure priority to place an advertisement for a specific product in the main ad area or the priority exposure area.

[0223] For example, in the case of a product where the efficiency level per advertisement is above a preset standard and the product-specific advertisement suitability is high, the device (200) can increase the frequency of advertisement exposure or expand the range of advertisement exposure. Conversely, in the case of a product where the current stock quantity is insufficient or the expected time of stock depletion is imminent, the frequency of advertisement exposure may be reduced or the range of advertisement exposure may be limited.

[0224] At this time, the pre-set criteria may include, but are not limited to, average advertising performance criteria by advertising channel, advertising efficiency criteria by product category, inventory quantity criteria, or sales availability period criteria.

[0225] The device (200) can change the advertising budget using the advertising budget adjustment value included in the advertising automation control value.

[0226] For example, in the case of a product where advertising performance indicators are continuously improving and recent shipment flow is maintained stably, the device (200) can increase the advertising budget. Conversely, if the number of ad clicks has increased but the actual number of conversions or sales revenue increase is minimal, the advertising budget can be reduced.

[0227] In addition, the device (200) can adjust the advertising budget for each advertising channel differently by reflecting the difference in advertising performance for each advertising channel.

[0228] For example, if a specific product shows a high conversion rate in search advertising channels and a high exposure diffusion effect in social media advertising platforms, the device (200) can allocate a conversion-oriented advertising budget to search advertising channels and an exposure diffusion-oriented advertising budget to social media advertising platforms.

[0229] In addition, the device (200) can also consider the total advertising operation budget limit during the advertising budget change process.

[0230] For example, when increasing the advertising budget for a specific product, the device (200) can adjust the advertising budgets of other products or other advertising channels together to maintain the overall advertising operation budget range.

[0231] The device (200) can change the advertising target product using the advertising target product change value included in the advertising automation control value.

[0232] For example, products that have been in a state of inventory backlog for a recent period can be newly added as advertising targets. Conversely, products with very low current stock levels or those whose sales period is about to end can be excluded from advertising targets.

[0233] In addition, the device (200) can also consider the balance of advertising operations by product category during the process of changing the advertising target product.

[0234] For example, if the advertising budget is excessively concentrated in a specific category, the device (200) can adjust the balance of advertising operations by adding products from other categories as advertising target products.

[0235] In addition, the device (200) can check whether the ad automation control value has been properly reflected in the actual ad operation environment.

[0236] For example, the device (200) can determine whether to apply advertising automation control values ​​by comparing the results of changes in advertising exposure levels, changes in advertising budgets, or changes in advertising target products with data received from an advertising platform or ERP system.

[0237] Additionally, the device (200) can provide the result of changing the advertising operation conditions through the manager's terminal (100).

[0238] For example, the device (200) can display changes to the ad exposure level, changes to the ad budget, and changes to the ad target product on the manager's terminal (100).

[0239] Additionally, the device (200) can provide information on advertising operation conditions changed according to advertising automation control values ​​through the manager's terminal (100).

[0240] For example, the device (200) can display to the manager's terminal (100) whether the frequency of ad exposure for a specific product has increased, whether the ad budget has been increased or decreased, and whether the product targeted for advertising has been changed.

[0241] At this time, verifying the result of changing the ad operation conditions may be an additional example to verify whether the ad automation control value has been applied normally, and depending on the example, it may be omitted or implemented in a different way.

[0242] Through this, the device (200) can automatically change the advertising operation conditions by reflecting both the advertising performance status and the actual product operation status. Since the device (200) can consider not only the advertising operation efficiency but also the inventory operation status and product sales flow, it can improve the connectivity between advertising operation and product operation.

[0243] Thus, the device (200) can analyze advertising data and logistics data stored in the ERP system together to sequentially generate advertising efficiency evaluation data, advertising plan data and advertising automation control values, and perform advertising automation control that reflects advertising performance and product operation status together.

[0244] FIG. 3 is a flowchart illustrating the process of generating advertising automation control values ​​according to one embodiment.

[0245] Referring to FIG. 3, first, in step S301, the device (200) can generate inventory status data including the current inventory quantity per product, shipment history, sales period, and expected time of inventory depletion from logistics data.

[0246] In the present invention, inventory status data may be defined as data for indicating the current inventory operation status of a product, and may include the current inventory quantity per product, shipment history, available sales period, and estimated time of inventory depletion.

[0247] In this context, the current inventory quantity may refer to the quantity of product inventory held as of a specific point in time. Additionally, the shipment history may refer to data storing past product shipment records in chronological order, and may include shipment date and time, shipment quantity, shipment type, and shipment destination information.

[0248] In addition, the sales availability period may refer to the period during which the product can be sold normally. For example, the sales availability period may include, but is not limited to, the expiration date, the best-before date, the seasonal sales availability period, the promotion operation period, or the contractually permitted sales period. It may be set differently depending on the embodiment.

[0249] The estimated time of inventory depletion may refer to the point in time when all inventory is expected to be exhausted based on the current inventory quantity and recent shipment trends.

[0250] That is, the device (200) can generate inventory status data by organizing the inventory and outbound information included in the logistics data into a structure that allows for analysis of the inventory operation status by product, rather than maintaining it in the form of simple stored data.

[0251] For example, the device (200) can collect the current inventory quantity of a specific product, the shipment history over a recent period, information on the period during which the product can be sold, and the estimated time of inventory depletion from logistics data stored in an ERP system, and generate inventory status data by linking these together on a product-by-product basis.

[0252] For example, if the current stock quantity of a specific product is 1,000 units and the average daily shipment quantity over the past 14 days is 100 units, the device (200) can calculate the expected time of stock depletion as about 10 days later.

[0253] In addition, the device (200) can adjust the expected time of inventory depletion by taking into account the degree of change in the recent outflow flow.

[0254] For example, if the rate of increase in shipment volume over the past three days has risen sharply, the device (200) can calculate an earlier time as the expected time for inventory depletion than when using only the simple average shipment volume. Conversely, if the recent shipment volume shows a decreasing trend, the expected time for inventory depletion can be adjusted to a later time.

[0255] At this time, the estimated time of inventory depletion may be calculated based on the average value of recent shipments, the growth rate of recent shipments, sales deviation by day of the week, seasonal information, whether a promotion is being operated, or the status of advertising execution, but is not limited thereto and may be calculated differently depending on the embodiment.

[0256] In addition, the device (200) can analyze the inventory management status of a product by comparing the sales period and the expected time of inventory depletion together.

[0257] For example, if it is expected that all inventory will be depleted before the end of the sales period, the device (200) can analyze the product as being in a normal inventory operation state. Conversely, if it is expected that a significant amount of inventory will remain even after the end of the sales period, the device (200) can analyze it as being in a state with a high probability of inventory accumulation.

[0258] Additionally, the device (200) can generate inventory status data by sorting the outgoing history in chronological order or grouping it by period.

[0259] For example, the device (200) can store shipment history based on the last 7 days, last 30 days, or last 90 days, and can use this to analyze short-term shipment flow and long-term shipment flow together.

[0260] In addition, the device (200) can generate inventory status data by reflecting the characteristics of each product category.

[0261] For example, in the case of product categories with high seasonality, a significant increase in shipment volume may occur during a specific period, so the device (200) can generate inventory status data by reflecting the shipment history of the previous same season. Conversely, in the case of household consumer goods with a high repeat purchase rate, inventory status data can be generated by applying a higher reflection rate to the average shipment volume over a recent period.

[0262] Additionally, the device (200) can exclude or correct abnormal outgoing data during the process of generating inventory status data.

[0263] For example, if a temporarily excessive volume of shipments occurs at a specific point in time but does not match actual sales data, the device (200) can classify the shipment data as abnormal shipment data and exclude or correct it.

[0264] At this time, the criteria for determining abnormal shipment data may include, but are not limited to, a deviation range relative to the average shipment volume, an allowable fluctuation range by product category, a maximum allowable growth rate within a specific period, or a range of difference from actual sales data, and may be set differently depending on the embodiment.

[0265] Through this, the device (200) can generate inventory status data that reflects both the current inventory status and the actual outbound flow for each product. The device (200) can analyze the possibility of inventory shortage or inventory accumulation more precisely by considering both the sales period and the expected time of inventory depletion, and can subsequently secure basic data that can be used in the process of generating demand pattern data for each product and generating advertising automation control values.

[0266] In step S302, the device (200) can classify the shipment history by time unit to generate shipment time series data by product.

[0267] In the present invention, shipment time series data can be defined as data indicating how the shipment flow of products changes over time, and may include shipment quantity information sorted in chronological order.

[0268] In this case, the time unit may refer to a standard unit for distinguishing shipment history, and the time unit may be set by hour, day, week, month, season, or advertising operation period, but is not limited thereto and may be set differently depending on the embodiment.

[0269] That is, the device (200) can generate product-specific shipment time series data by converting the shipment history included in the inventory status data into a time series structure so as to analyze the shipment change pattern over time, rather than maintaining it in the form of simple record data.

[0270] For example, the device (200) can generate product-specific shipment time series data by sorting the shipment history of a specific product in chronological order based on the time of shipment occurrence and matching the shipment volume for each time interval with each other.

[0271] For example, if the shipment volume of a specific product is recorded as 100 on Monday, 120 on Tuesday, 140 on Wednesday, and 160 on Thursday, the device (200) can generate product-specific shipment time series data in which the daily shipment flow is connected in chronological order.

[0272] Additionally, the device (200) can set different time units depending on the characteristics of the product or the purpose of advertising operation.

[0273] For example, for products with rapid fluctuations in shipment volume, time-series shipment data can be generated on an hourly or daily basis. Conversely, for products with relatively gradual changes in shipment volume, time-series shipment data can be generated on a weekly or monthly basis.

[0274] In addition, for products with strong seasonality, shipment time-series data can also be generated on a seasonal basis.

[0275] For example, in the case of summer seasonal products, the device (200) can generate seasonal shipment time series data by linking the shipment history of the same season from the previous year together.

[0276] The device (200) can aggregate the shipment volume for each time interval during the shipment time series data generation process.

[0277] For example, if there are multiple shipment records on the same date, the device (200) can generate daily shipment data by summing the total shipment volume for that date.

[0278] Additionally, the device (200) can maintain information about the corresponding time interval even if no shipment occurs during a specific time interval.

[0279] For example, if the daily shipment volume of a specific product is 100 on Monday, 0 on Tuesday, and 130 on Wednesday, the device (200) can maintain a continuous time series structure including the Tuesday shipment volume as 0.

[0280] Through this, the device (200) can maintain the continuity of the shipment flow, including periods during which actual shipments do not occur, and can more accurately reflect the characteristics of changes over time when analyzing demand patterns thereafter.

[0281] In addition, the device (200) can generate outgoing time series data by reflecting the characteristics of each product category.

[0282] For example, for everyday consumer goods with short repurchase cycles, shipment time-series data can be generated by applying a higher weighting to recent short-term shipment flows. Conversely, for products with strong seasonality, shipment time-series data can be generated by reflecting long-term shipment flows or shipment flows from the previous same season.

[0283] In addition, the device (200) can link advertising execution period information with shipment time series data so as to analyze the correlation between the advertising operation status and the shipment flow.

[0284] For example, if the volume of shipments increases rapidly during a specific advertising campaign operation period, the device (200) can generate shipment time series data in a way that corresponds the advertising operation status and the shipment flow during that period.

[0285] Additionally, the device (200) can correct abnormal shipment flow during the process of generating shipment time series data.

[0286] For example, if a temporarily excessive volume of shipments occurs at a specific point in time but does not match the actual sales flow or advertising operation status, the device (200) can classify the data as abnormal shipment data and correct or exclude it.

[0287] At this time, the criteria for determining abnormal shipment data may include, but are not limited to, a deviation range relative to the average shipment volume, an allowable growth rate per time unit, an allowable fluctuation range per product category, or a range of difference from actual sales data.

[0288] Additionally, the device (200) can provide the status of product-specific shipment changes over time through the manager's terminal (100) after generating shipment time series data.

[0289] For example, the device (200) can display the daily shipment volume change trend of a specific product, the weekly average shipment volume change status, or the seasonal shipment flow change in the form of a graph or a table on the manager's terminal (100).

[0290] Through this, the device (200) can generate product-specific shipment time series data that reflects the characteristics of actual shipment changes over time. Since the device (200) can secure base data for analyzing the increasing trend, decreasing trend, or fluctuation characteristics of the shipment flow for each product, it can subsequently improve the accuracy of calculating demand pattern data and generating advertising automation control values.

[0291] In step S303, the device (200) can analyze outgoing time series data to produce demand pattern data including demand increase trends, demand decrease trends, and demand variability by product.

[0292] In the present invention, demand pattern data can be defined as data for representing the characteristics of changes in demand for a product over time, and may include a demand increase trend, a demand decrease trend, and demand volatility for each product.

[0293] In this case, an increasing demand trend may refer to a tendency for the volume of product shipments or sales flow to increase during a specific period, while a decreasing demand trend may refer to a tendency for the volume of product shipments or sales flow to decrease during a specific period.

[0294] In addition, demand volatility can be a value indicating how significantly the range of change in product shipment volume varies over time.

[0295] That is, the device (200) can not use the outgoing time series data as simple storage data, but can analyze the pattern of change in outgoing volume over time to produce demand pattern data representing the demand flow characteristics for each product.

[0296] For example, the device (200) can calculate the demand growth trend for each product by analyzing the increase rate of shipments over a recent period.

[0297] For example, if the average shipment volume of a specific product over the past 7 days continuously increases compared to the average shipment volume over the previous 7 days, the device (200) can analyze that a trend of increasing demand for the product has been formed. Conversely, if the average shipment volume over a recent period continuously decreases, the device can analyze that a trend of decreasing demand has been formed.

[0298] For example, if the average shipment volume of a specific product over the past 14 days continues to decrease compared to the average shipment volume over the previous 14 days, the device (200) can calculate the trend of decreasing demand for the product.

[0299] In addition, the device (200) can calculate demand variability using the range of change in the shipment volume per hour.

[0300] For example, if the daily shipment volume of a specific product is very large and rapid changes in shipment volume occur repeatedly at specific intervals, the device (200) can calculate the demand variability of the product as high. Conversely, if the shipment volume is maintained stably for a certain period, the demand variability can be calculated as low.

[0301] At this time, demand volatility may be calculated based on the fluctuation range relative to the average shipment volume, the standard deviation of shipment volume by period, the range of change in the shipment volume growth rate, or the deviation of shipment volume by time unit, but is not limited thereto and may be calculated differently depending on the embodiment.

[0302] In addition, the device (200) can calculate demand pattern data by comparing short-term demand flow and long-term demand flow together.

[0303] For example, if the shipment volume over the past 7 days is on an increasing trend but the shipment volume over the past 90 days is on a decreasing trend, the device (200) can be analyzed as a product in which a short-term demand increase state and a long-term demand decrease state exist simultaneously. Conversely, if both short-term and long-term shipment flows are increasing, it can be analyzed as a continuous demand increase state.

[0304] In addition, the device (200) can generate demand pattern data by reflecting the characteristics of each product category.

[0305] For example, in the case of a product category with strong seasonality, the device (200) can calculate a demand increase trend or a demand decrease trend by comparing the shipment flow of the previous same season with the current shipment flow. Conversely, in the case of household consumer goods with a high repeat purchase rate, demand pattern data can be calculated by applying a higher reflection rate to the recent short-term shipment flow.

[0306] In addition, the device (200) can calculate demand pattern data by considering the correlation between the advertising operation status and the actual shipment flow.

[0307] For example, if the increase rate of shipment volume continues to rise after the operation of a specific advertising campaign, the device (200) can analyze that the trend of demand increase is strengthened by the advertising operation. Conversely, if the number of ad impressions or ad clicks has increased but there has been almost no increase in shipment volume, the device can analyze that the effect of actual demand increase is low compared to the inflow of ads.

[0308] The device (200) can correct temporary abnormal outflows during the process of calculating demand pattern data.

[0309] For example, if an excessive volume of shipments occurs temporarily on a specific date but the shipment flow subsequently returns to a normal level, the device (200) can classify the data as temporary abnormal shipment data and reduce the reflection ratio when calculating the demand increase trend.

[0310] In addition, the device (200) can analyze the phenomenon of a surge in shipment volume during a specific promotion period by distinguishing it from the general trend of increasing demand.

[0311] For example, if the shipment volume increases only during a short-term discount event period, the device (200) can classify the data for that period as promotion impact data and process it separately in the process of calculating the long-term demand increase trend.

[0312] At this time, the criteria for determining abnormal shipment flow may include, but are not limited to, a deviation range relative to the average shipment volume, a maximum allowable growth rate per hour, an allowable fluctuation range per product category, or a range of increase in shipment volume relative to the advertising operation status.

[0313] Additionally, the device (200) can provide the calculated demand pattern data through the manager's terminal (100).

[0314] For example, the device (200) can display information on the demand increase trend, demand decrease trend, and demand volatility by product in the form of a graph, an indicator, or comparison information by period on the manager's terminal (100).

[0315] Additionally, the device (200) can visually distinguish whether a specific product is in a state of increased demand or decreased demand.

[0316] For example, the device (200) may prioritize displaying products with a high demand growth trend or provide a warning indicator for products with high demand volatility.

[0317] Through this, the device (200) can produce demand pattern data that reflects changes in the actual demand flow for each product. Since the device (200) can analyze product demand characteristics over time, it can subsequently improve the accuracy of analyzing inventory levels by category, calculating inventory shortage risk, calculating inventory excess risk, and generating advertising automation control values.

[0318] In step S304, the device (200) can aggregate product-specific inventory status data and demand pattern data by category to generate category-specific inventory levels and category-specific demand pattern data.

[0319] In the present invention, the inventory level by category can be defined as data representing the aggregated inventory status of products belonging to the same category, and may include the current inventory quantity, average inventory quantity, distribution of expected inventory depletion times, or inventory accumulation status information by category.

[0320] In addition, demand pattern data by category can be defined as data representing the aggregated demand flow characteristics of products belonging to the same category, and may include demand growth trends, demand decline trends, average demand volatility, or shipment flow information by period.

[0321] In this case, the category may refer to a standard unit for classifying products and may include, but is not limited to, food, household goods, clothing, seasonal products, electronic products, or product groups by brand. It may be set differently depending on the embodiment.

[0322] That is, the device (200) can analyze the inventory status and demand flow characteristics of the entire category by aggregating products with similar characteristics at the category level, rather than using product unit inventory status data and demand pattern data individually.

[0323] For example, the device (200) can generate a current inventory level by category by summing the current inventory quantities of multiple products belonging to the same category.

[0324] In addition, the device (200) can analyze the inventory depletion flow at the category level by calculating the average expected time of depletion of products belonging to the same category.

[0325] For example, if the average expected time of depletion of inventory for products belonging to a specific category is continuously delayed, the device (200) can analyze that the overall possibility of inventory accumulation in that category has increased. Conversely, if the expected time of depletion of inventory for products in a specific category is generally brought forward, it can analyze that the demand at the category level has increased.

[0326] In addition, the device (200) can aggregate product-specific demand pattern data by category to generate category-specific demand pattern data.

[0327] For example, the device (200) can calculate the demand increase status by category by averaging or weighting the demand increase trends of products belonging to the same category.

[0328] In addition, the device (200) can analyze the demand stability of the entire category by analyzing the demand variability of products within the same category.

[0329] For example, if the demand volatility of products in a specific category remains generally high, the device (200) can analyze the category as a high-variability category with large demand changes. Conversely, if the outbound flow remains stable for a certain period, it can be analyzed as a stable demand category.

[0330] The device (200) can reflect the importance of each product in the process of generating category-specific inventory levels and category-specific demand pattern data.

[0331] For example, the device (200) can generate category-unit data by applying a higher reflection rate to products with high recent shipment volume, products with high advertising performance indicators, or products with high sales contribution.

[0332] In addition, the device (200) can generate demand pattern data by category by reflecting the seasonal shipment flow in the case of categories with strong seasonality.

[0333] For example, in the case of a summer season product category, the device (200) can calculate the demand increase trend by category by comparing the shipment time series data of the previous same season with the current shipment time series data. Conversely, in the case of a household consumer goods category with a high repeat purchase rate, a higher reflection rate can be applied to the recent short-term shipment flow.

[0334] In addition, the device (200) can generate demand pattern data by category by taking into account the advertising operation status.

[0335] For example, if the number of ad impressions and ad clicks for a specific category is continuously increasing, and at the same time the growth rate of the total shipment volume of the category is also rising, the device (200) can analyze that the increase in demand at the category level is occurring due to the ad operation. Conversely, if the growth rate of the total shipment volume of the category remains low compared to the increase in the number of ad impressions, it can be analyzed as a category with a low actual demand conversion effect compared to the ad inflow.

[0336] Additionally, the device (200) can correct abnormal product data during the process of generating category-specific inventory levels and category-specific demand pattern data.

[0337] For example, if the shipment volume of a specific product temporarily increases rapidly but does not match the actual sales flow, the device (200) can classify the product data as abnormal data and reduce the reflection rate in the category-unit aggregation process.

[0338] At this time, the criteria for determining abnormal data may include, but are not limited to, a deviation range relative to the average shipment volume, an allowable fluctuation range by category, a maximum allowable growth rate within a specific period, or a range of increase in shipment volume relative to the advertising operation status.

[0339] Additionally, the device (200) can provide generated category-specific inventory levels and category-specific demand pattern data through the manager's terminal (100).

[0340] For example, the device (200) can display the average inventory level by category, the state of increased demand by category, the variability of demand by category, and the change in outgoing flow by category in the form of a graph, a table, or a comparison indicator on the manager's terminal (100).

[0341] Additionally, the device (200) can visually distinguish and provide the possibility of a stock shortage or stock accumulation in a specific category.

[0342] For example, the device (200) may prioritize displaying categories with a high probability of stock shortage, or display warning information for categories with high demand variability.

[0343] Through this, the device (200) can analyze the inventory status and demand flow characteristics of not only individual product units but also entire categories. Since the device (200) can reflect changes in demand and inventory flow at the category level, it can improve the accuracy of calculating safety stock standards and excess stock standards, as well as calculating stock shortage risk and stock excess risk.

[0344] In step S305, the device (200) can calculate a safety stock standard and an excess stock standard for each product based on the stock quantity, the expected time of stock depletion, and demand variability.

[0345] In the present invention, the safety stock standard can be defined as a reference value representing the minimum stock level required to maintain the normal sales flow of a specific product.

[0346] In addition, the excess inventory criterion can be defined as a threshold value for determining whether the inventory of a specific product has accumulated excessively.

[0347] In other words, the safety stock standard may be a criterion regarding the minimum maintenance inventory that must be maintained to prevent sales suspensions or shipment delays caused by stock shortages. On the other hand, the excess inventory standard may be a criterion regarding an upper inventory limit to determine whether to induce inventory depletion through increased ad impressions or advertising budgets, given that current inventory has accumulated excessively compared to normal sales flow.

[0348] In this case, the safety stock standard can be used as a criterion to prevent sales suspension or advertising operation restrictions due to stock shortages, and the excess stock standard can be used as a criterion to induce stock depletion by assessing the state of stock accumulation.

[0349] Accordingly, the device (200) can generate subsequent advertising control conditions centered on the possibility of a stock shortage when the current stock quantity is below the safety stock standard, and can generate subsequent advertising control conditions centered on the possibility of inducing shipment through advertising when the current stock quantity is above the excess stock standard.

[0350] That is, the device (200) can determine the inventory operation status based not only on the current inventory status, but also by considering the outbound flow and demand change characteristics for each product together to calculate the safety stock standard and the excess stock standard for each product.

[0351] For example, the device (200) can calculate a safety stock standard by analyzing the average shipment volume of a specific product, the recent trend of increasing demand, and the variability of demand together.

[0352] For example, if the average daily shipment volume of a specific product over the past 30 days is 100 units and the recent trend of increasing demand continues to rise and demand volatility is high, the device (200) can calculate a relatively high safety stock standard to respond to a situation of sudden increase in demand. Conversely, for a product where the shipment flow is maintained stably and demand volatility is low, the safety stock standard can be calculated relatively low.

[0353] In addition, the device (200) can calculate the safety stock standard by taking into account the expected time of stock depletion.

[0354] For example, if the expected time for the depletion of a specific product's inventory is very imminent, the device (200) can analyze that there is a high probability that the product will fall short of the safety stock standard, and accordingly, the safety stock standard can be adjusted. Conversely, if there is sufficient time before the expected time for the depletion of inventory, the current inventory status can be analyzed as stable, and the safety stock standard can be relatively relaxed.

[0355] Additionally, the device (200) can adjust the safety stock standard by utilizing the demand variability for each product.

[0356] For example, if the daily shipment volume of a specific product varies significantly and a rapid increase in shipment volume occurs repeatedly during specific periods, the device (200) can calculate a high safety stock standard to respond to the possibility of a surge in demand. Conversely, if the shipment flow remains constant and the shipment volume variation is small, a relatively low safety stock standard can be applied.

[0357] At this time, the safety stock standard may be calculated based on the average shipment volume, maximum shipment volume, demand growth rate, demand volatility, expected period for inventory depletion, or average shipment flow by product category, but is not limited thereto and may be calculated differently depending on the embodiment.

[0358] In addition, the device (200) can calculate the excess inventory criteria for each product by taking into account the inventory quantity, the expected time of inventory depletion, and the trend of decreasing demand.

[0359] For example, if the current inventory quantity of a specific product is higher than the average inventory level, the recent trend of decreasing shipment volume continues, and the estimated time of inventory depletion is calculated to be a long period later, the device (200) can analyze that there is a high possibility of excess inventory of the product.

[0360] In this case, the device (200) can generate basic data to determine whether inventory-draining ad controls, such as increasing ad exposure, increasing the ad budget, or including the product as an ad target, are necessary for the product by using whether the current stock quantity is above the excess stock standard.

[0361] In addition, if the demand volatility of a specific product is low and the state of reduced shipment volume is maintained for a long period, the device (200) can analyze the product as one for which the possibility of inventory accumulation is continuously maintained and strengthen the excess inventory criteria.

[0362] Conversely, if a recent trend of increasing demand is established and the expected time for inventory depletion is brought forward, the excess inventory criteria may be relaxed.

[0363] The device (200) may also calculate safety stock standards and excess stock standards by reflecting the characteristics of each product category.

[0364] For example, in the case of product categories with strong seasonality, the possibility of increased demand may be higher immediately before a specific season, so the device (200) can calculate a high safety stock standard during the period prior to the start of the season. Conversely, as the end of the season approaches, the excess stock standard can be relatively strengthened to enable operations focused on depleting inventory.

[0365] In addition, for household consumer goods with high repeat purchase rates, safety stock standards can be calculated by applying a higher reflection rate to recent short-term shipment flows.

[0366] The device (200) may also calculate safety stock standards and excess stock standards by taking into account the advertising operation status.

[0367] For example, if the rate of increase in shipment volume continues to rise after an increase in the number of ad impressions, the device (200) can analyze that there is a high possibility of additional demand increase due to the ad operation and calculate a high safety stock standard. Conversely, if there is almost no increase in actual shipment volume despite an increase in the number of ad impressions, the excess stock standard can be relatively strengthened.

[0368] Additionally, the device (200) can correct temporary abnormal outflow flow or abnormal inventory data during the process of calculating safety stock standards and excess stock standards.

[0369] For example, if an excessive volume of shipments occurs temporarily during a specific period but the shipment flow subsequently returns to a normal state, the device (200) can classify the data as temporary abnormal data and reduce the reflection ratio when calculating the safety stock standard.

[0370] At this time, the criteria for determining abnormal data may include, but are not limited to, a deviation range relative to the average shipment volume, an allowable fluctuation range by product category, and a maximum allowable growth rate or inventory fluctuation rate range within a specific period, and may be set differently depending on the embodiment.

[0371] Additionally, the device (200) can provide the calculated safety stock standard and excess stock standard through the manager's terminal (100).

[0372] For example, the device (200) can display safety stock standards by product, excess stock standards by product, whether standards are met relative to the current stock quantity, and stock risk status information on the manager's terminal (100) in the form of a table, a graph, or warning information.

[0373] Additionally, the device (200) can visually distinguish whether a specific product is below the safety stock standard or above the excess stock standard.

[0374] For example, the device (200) may provide a warning indication for products with a high probability of safety stock shortage, or display information to induce stock depletion for products with a high probability of excess stock.

[0375] Through this, the device (200) can calculate safety stock standards and excess stock standards that reflect the actual demand flow and inventory status for each product. Since the device (200) can analyze the risk of stock shortage through the safety stock standards and analyze the need to induce stock depletion through advertising through the excess stock standards, it can subsequently improve the accuracy of calculating the risk of stock shortage and the risk of stock excess, as well as the generation of automated advertising control values.

[0376] In step S306, the device (200) can use inventory status data, demand pattern data, inventory levels by category, and demand pattern data by category, and calculate the risk of inventory shortage and the risk of inventory excess based on the safety stock standard and the excess stock standard as comparison criteria.

[0377] In the present invention, the inventory shortage risk can be defined as an evaluation value indicating the possibility that the inventory of a specific product will reach a shortage state in the future.

[0378] In addition, the risk of excess inventory can be defined as an assessment value indicating the likelihood of excessive stockpiling of a specific product.

[0379] In this case, the inventory shortage risk can be used as a value to analyze the likelihood of sales suspension or shipment delays caused by inventory shortages during the advertising operation process, while the inventory excess risk can be used as a value to analyze the likelihood of long-term inventory accumulation or sluggish sales.

[0380] That is, the device (200) can calculate the risk of inventory shortage and the risk of inventory excess by considering the demand flow by product, the characteristics of demand change by category, and the inventory operation status together, rather than determining the risk state based only on the current inventory quantity.

[0381] For example, the device (200) can analyze the possibility of a stock shortage by comparing the current stock quantity of a specific product with a safety stock standard.

[0382] For example, if the current inventory quantity is below the safety stock standard and the recent trend of increasing demand continues to rise, the device (200) can calculate a high risk of inventory shortage for the product.

[0383] In addition, even if the current inventory quantity is above the safety stock standard, if the expected time of inventory depletion is very imminent and the recent increase rate of shipment volume is rapidly rising, the device (200) can analyze the situation as having a high probability of future inventory shortage and calculate a high risk of inventory shortage.

[0384] Conversely, if the current inventory level is sufficiently higher than the safety stock threshold and recent outflow trends remain stable, the risk of inventory shortage can be calculated as low.

[0385] In addition, the device (200) can adjust the risk of inventory shortage by taking into account the variability in demand for each product.

[0386] For example, if the demand volatility of a specific product is very high, there may be a possibility of a rapid increase in the volume of shipments within a short period, so the device (200) can calculate a higher risk of inventory shortage even at the same inventory level. Conversely, if the demand volatility is low and the shipment flow is maintained constant, the risk of inventory shortage can be calculated relatively low.

[0387] The device (200) can correct the risk of inventory shortage by using the inventory level by category and the demand pattern data by category together.

[0388] For example, even if the inventory status of a specific product itself is stable, if the trend of increasing demand for the entire same category rises rapidly, the device (200) can adjust the risk of inventory shortage upward by reflecting the impact of increased demand at the category level.

[0389] Conversely, if the downward trend in demand across the entire same category continues, the risk of inventory shortages can be mitigated.

[0390] In addition, the device (200) can calculate the risk of excess inventory by comparing the current inventory quantity with the excess inventory standard.

[0391] For example, if the current inventory quantity of a specific product is above the excess inventory threshold and a recent trend of decreasing demand continues, the device (200) can calculate a high risk of excess inventory for the product.

[0392] In addition, even if the estimated time for inventory depletion is calculated to be a long period in the future and the recent decline in shipment volume persists, the analysis indicates a high probability of inventory accumulation, allowing for a high calculation of the risk of excess inventory.

[0393] Conversely, if the current inventory quantity is below the excess inventory threshold and a recent trend of increasing demand is forming, the risk of excess inventory can be calculated as low.

[0394] Additionally, the device (200) can correct the risk of excess inventory by using category-specific inventory levels and category-specific demand pattern data.

[0395] For example, even if the inventory status of a specific product exceeds the excess inventory threshold, if an increasing demand trend is formed for the entire same category, the device (200) can reduce the risk of excess inventory by reflecting the possibility of future inventory depletion. Conversely, if a decreasing demand trend for the entire same category continues, the risk of excess inventory can be calculated as higher.

[0396] The device (200) can calculate the risk of stock shortage and the risk of stock excess in numerical or grade form.

[0397] For example, the device (200) can calculate the inventory shortage risk and inventory excess risk in the form of a score in the range of 0 to 100.

[0398] Additionally, the device (200) may classify the risk level into risk grades such as low, normal, high, or very high depending on the risk level.

[0399] At this time, the risk calculation criteria may include, but are not limited to, the ratio of safety stock to current stock quantity, the ratio of excess stock to current stock quantity, the expected period for stock depletion, the demand growth rate, the demand decrease rate, demand volatility, or the flow of demand changes by category.

[0400] Additionally, the device (200) can correct temporary abnormal outflow flow or abnormal inventory data during the process of calculating the inventory shortage risk and inventory excess risk.

[0401] For example, if an excessive volume of shipments occurs temporarily during a specific period but the shipment flow subsequently returns to a normal state, the device (200) can classify the data as temporary abnormal data and reduce the reflection rate in the risk calculation process.

[0402] Additionally, if the inventory data of a specific product increases or decreases abnormally due to a temporary system error, the device (200) can calculate the risk level after correcting or excluding the data.

[0403] At this time, the criteria for determining abnormal data may include, but are not limited to, a deviation range relative to the average shipment volume, an allowable fluctuation range by category, and a maximum allowable growth rate or inventory fluctuation rate range within a specific period, and may be set differently depending on the embodiment.

[0404] Additionally, the device (200) can provide the calculated inventory shortage risk and inventory excess risk through the manager's terminal (100).

[0405] For example, the device (200) can display product-specific inventory shortage risk status, product-specific inventory excess risk status, risk score change flow and risk grade information in the form of a graph, a table, or warning information on the manager's terminal (100).

[0406] Additionally, the device (200) may prioritize displaying a product or provide warning notification information when the risk level of a specific product increases rapidly.

[0407] Through this, the device (200) can calculate the risk of inventory shortage and the risk of inventory excess by reflecting the actual demand flow by product, the inventory status, and the change in demand by category. Since the device (200) can analyze the possibility of inventory shortage and the possibility of inventory accumulation more precisely, it can improve the accuracy of subsequently generating conditions for increasing ad exposure, generating first ad control conditions, generating second ad control conditions, and generating ad automation control values.

[0408] In step S307, the device (200) can generate an advertising exposure increase condition that induces priority shipment for products whose remaining sales period is below a preset standard.

[0409] In the present invention, the remaining period of the saleable period may refer to the remaining period during which a specific product can be sold normally.

[0410] For example, the remaining period of the sales period may include, but is not limited to, the remaining period until the expiration date, the remaining period until the best before date, the remaining period until the end of the season, the remaining period until the end of the promotion, or the remaining period until the end of the contractual sales permission.

[0411] Additionally, the conditions for increasing ad exposure can be defined as condition data for increasing the level of ad exposure for a specific product, and may include condition information for increasing ad exposure frequency, raising ad exposure priority, or expanding the scope of ad exposure.

[0412] That is, the device (200) can generate conditions for increasing ad exposure so that, rather than simply concentrating ads on products with high advertising efficiency, priority can be given to the shipment of products with a short remaining sales period.

[0413] For example, the device (200) can calculate the remaining period of the sales period by comparing the end point of the sales period of a specific product with the current point in time.

[0414] For example, if the remaining period until the end of the sales period for a specific product is 5 days and the preset standard period is 7 days, the device (200) may determine that the product is eligible for the creation of an increased ad exposure condition. Conversely, if the remaining period until the end of the sales period is sufficiently long and the current inventory management status is stable, it may be excluded from the creation of an increased ad exposure condition.

[0415] At this time, the pre-set criteria may include, but are not limited to, the average sales period per product category, expiration date criteria, season end criteria, promotion end criteria, or inventory management policy criteria.

[0416] Additionally, the device (200) can generate conditions for increasing ad exposure by taking into account the remaining period of the sales period as well as the current inventory status.

[0417] For example, if the remaining period of the sales period is short and the current inventory quantity is above the excess inventory standard, the device (200) can analyze the situation as having a high need for priority shipment and set the priority of the conditions for increasing ad exposure high.

[0418] Conversely, if the remaining sales period is short but the current inventory quantity is very low, the level of increase in ad impressions may be limited or the item may be excluded from the conditions for increasing ad impressions.

[0419] In addition, the device (200) can generate conditions for increasing ad exposure by considering product-specific demand pattern data together.

[0420] For example, if the remaining sales period of a specific product is short and a recent trend of declining demand continues, the device (200) can analyze that there is a high possibility of inventory accumulation and set the level of increased ad exposure higher. Conversely, if a recent trend of increasing demand is formed and a natural flow of increased shipments is maintained, the level of increased ad exposure can be relatively mitigated.

[0421] In addition, the device (200) can generate conditions for increasing ad exposure by reflecting the characteristics of each ad channel.

[0422] For example, if a specific advertising channel is analyzed as having a high short-term sales conversion effect, the device (200) may preferentially apply conditions to increase ad exposure to that advertising channel. Conversely, in the case of a brand awareness-centered advertising channel, the increase in ad exposure may be relatively limited.

[0423] In addition, the device (200) can generate conditions for increasing ad exposure by considering the characteristics of each product category.

[0424] For example, in the case of the fresh food category, the criteria for the remaining sales period can be set to a relatively short duration, and in the case of the seasonal product category, conditions for increasing ad impressions can be created by considering the end-of-season timeframe as well.

[0425] In addition, the device (200) can also consider advertising operation stability during the process of creating conditions for increasing advertising exposure.

[0426] For example, even if the remaining sales period for a specific product is short, criteria for limiting the increase rate of ad exposure can be applied to prevent the level of ad exposure from increasing excessively within a short period.

[0427] For example, the device (200) can generate an ad exposure increase condition based on the maximum increase rate of ad exposure frequency, the allowable increase range per ad channel, or the maximum increase amount per ad campaign.

[0428] At this time, the criteria for limiting the increase rate of ad impressions may include, but are not limited to, ad operation policies by product category, operation standards by ad channel, or inventory depletion target standards, and may be set differently depending on the embodiment.

[0429] Additionally, the device (200) can provide the generated ad exposure increase conditions through the manager's terminal (100).

[0430] For example, the device (200) can display on the manager's terminal (100) a list of products with an imminent remaining period of sale, information on the planned increase in ad exposure, information on the change in the expected time of stock depletion, and information on the status of priority shipment.

[0431] Additionally, the device (200) can also provide the reason why the conditions for increased ad exposure were created.

[0432] For example, the device (200) can display information on the remaining sales period shortage of a specific product, the current inventory quantity status, and the recent demand decrease trend.

[0433] Through this, the device (200) can generate conditions for increasing ad exposure that can induce priority shipment for products with a short remaining sales period. Since the device (200) can adjust ad operation conditions to efficiently deplete inventory before the end of the sales period, it can reduce the possibility of losses due to inventory accumulation and exceeding the sales period.

[0434] In step S308, the device (200) can generate a first ad control condition including a reduction in ad exposure and a reduction in ad budget if the risk of stock shortage is above a preset standard and the efficiency level per ad included in the ad efficiency evaluation data is above a preset standard.

[0435] In the present invention, the first advertising control condition can be defined as condition data for adjusting the advertising exposure level and advertising budget for products with a high probability of stock shortage.

[0436] At this time, the first ad control condition may include at least one of an ad exposure reduction condition, an ad budget reduction condition, an ad target product exclusion condition, or an ad operation restriction condition per ad channel.

[0437] In other words, the first advertising control condition can be utilized as an inventory shortage prevention suppression condition to prevent the inventory shortage from worsening due to the expansion of advertising operations.

[0438] That is, the device (200) can generate a first advertising control condition so that it can automatically adjust the advertising operation intensity even if the product has excellent advertising performance, in cases where there is a high risk of stock shortage.

[0439] For example, the device (200) can determine whether the risk of a stock shortage of a specific product is above a preset standard.

[0440] For example, if the current inventory quantity of a specific product is below the safety stock standard, the recent trend of increasing demand continues to rise, and the expected time of inventory depletion is very imminent, the device (200) can calculate a high risk of inventory shortage for the product.

[0441] In addition, the device (200) can determine whether the efficiency level of each advertisement included in the advertisement efficiency evaluation data is above a preset standard.

[0442] For example, if the ad click-through rate, conversion rate, and sales-to-advertisement ratio of a specific product are all above the average standard for each ad channel, the device (200) can analyze that the efficiency level of the product's ad is high.

[0443] That is, the device (200) can generate a first advertising control condition for a product that has excellent advertising performance, is likely to have its advertising operations continuously expanded, and at the same time has a high risk of stock shortage.

[0444] For example, if the advertising efficiency of a specific product is very high and the number of ad clicks and shipments continues to increase, but the current inventory quantity decreases rapidly and the estimated time for inventory depletion is calculated to be within a short period, the device (200) can generate a first ad control condition including a reduction in ad exposure and a reduction in the ad budget.

[0445] Conversely, if the efficiency level per advertisement is low or the risk of inventory shortage is below a preset standard, it may be excluded from the target for generating the first advertisement control condition.

[0446] At this time, the criteria for inventory shortage risk may include, but are not limited to, the ratio of safety stock to current inventory quantity, the estimated period for inventory depletion, the demand growth rate, or the demand volatility.

[0447] In addition, the efficiency level criteria for each advertisement may include, but are not limited to, the average click-through rate criteria, average conversion rate criteria, revenue-to-advertising cost criteria, or advertising operation policy criteria per advertising channel, and may be set differently depending on the embodiment.

[0448] In addition, the device (200) can generate a first advertising control condition by considering the advertising operation characteristics of each advertising channel.

[0449] For example, if a specific advertising channel is analyzed as a channel with a very high purchase conversion effect, the device (200) may preferentially apply an ad exposure reduction condition or an ad budget reduction condition to that advertising channel.

[0450] Conversely, in the case of brand awareness-focused advertising channels, the reduction in ad exposure or ad budget can be relatively mitigated.

[0451] In addition, the device (200) can generate a first advertising control condition by taking into account the characteristics of each product category.

[0452] For example, in the case of household consumer goods with a high repeat purchase rate, the extent of the reduction in ad exposure can be limited to maintain a certain level of advertising operations even if inventory shortages occur.

[0453] Conversely, for seasonal or limited-edition products with limited supply, the level of ad exposure reduction can be set more significantly when a stock shortage risk occurs.

[0454] Additionally, the device (200) can generate a first ad control condition by considering the stability of ad operation.

[0455] For example, even if advertising performance indicators rise sharply in the short term, criteria for limiting the advertising reduction rate can be applied to prevent the level of ad impressions or the advertising budget from decreasing excessively all at once.

[0456] For example, the device (200) can generate a first ad control condition based on an ad exposure reduction rate, an ad budget reduction rate, or a maximum allowable reduction range per ad channel.

[0457] At this time, the criteria for limiting the advertising reduction rate may include, but are not limited to, advertising operation policies by product category, operation standards by advertising channel, or inventory operation stability standards, and may be set differently depending on the embodiment.

[0458] Additionally, the device (200) can consider category-specific inventory levels and category-specific demand pattern data together during the process of generating the first ad control condition.

[0459] For example, even if the risk of a specific product's stock shortage is high, the level of reduction in ad impressions can be relatively mitigated if the overall demand for the same category continues to decline. Conversely, if the overall demand for the same category continues to increase, the likelihood of a stock shortage may expand further, so the level of reduction in ad impressions can be intensified.

[0460] Additionally, the device (200) can provide the generated first advertisement control condition through the manager's terminal (100).

[0461] For example, the device (200) can display information on products scheduled for reduced ad exposure, information on the planned reduction of the ad budget, information on the risk of stock shortage, and information on the expected time of stock depletion on the manager's terminal (100).

[0462] In addition, the device (200) can also provide a cause for generating a first ad control condition.

[0463] For example, the device (200) can display a high advertising efficiency level of a specific product, a high risk of stock shortage, and an imminent stock depletion time.

[0464] In addition, the first ad control condition may be applied separately from the second ad control condition that is subsequently created, and the first ad control condition may be utilized to suppress ad operations in order to prevent the worsening of the inventory shortage state.

[0465] Through this, the device (200) can perform advertising operation control that reflects the risk of stock shortage for products with excellent advertising performance and a high potential for increased demand. Since the device (200) can reduce the possibility of sales suspension or shipment delay due to stock shortage, it can improve both advertising operation efficiency and inventory operation stability.

[0466] In step S309, the device (200) can generate a second ad control condition including an increase in ad exposure and an increase in the ad budget if the excess inventory risk is above a preset standard and the ad suitability for each product included in the ad efficiency evaluation data is above a preset standard.

[0467] In the present invention, the second advertising control condition can be defined as condition data for increasing the level of advertising exposure and the advertising budget for products with a high probability of inventory accumulation.

[0468] At this time, the second ad control condition may include at least one of a condition to increase ad exposure, a condition to increase the ad budget, a condition to prioritize the exposure of the target product, or a condition to expand ad operations by ad channel.

[0469] In other words, the second advertising control condition can be utilized as an inventory-resolving expansion condition that expands advertising operations to prevent the state of excess inventory from persisting for a long period and to induce inventory depletion.

[0470] That is, the device (200) can generate a second advertising control condition to expand advertising operations by considering whether the product is suitable for actual advertising operations, rather than simply unconditionally increasing advertising for products with a high risk of excess inventory.

[0471] For example, the device (200) can determine whether the risk of excess inventory of a specific product is above a preset standard.

[0472] For example, if the current inventory quantity of a specific product is above the excess inventory threshold, the recent trend of decreasing demand continues, and the estimated time of inventory depletion is calculated to be a long period later, the device (200) can calculate the excess inventory risk of the product as high.

[0473] In addition, the device (200) can determine whether the product-specific ad suitability included in the ad efficiency evaluation data is above a preset standard.

[0474] For example, if the advertising performance indicators of a specific product are maintained stably, the current inventory quantity is at a level capable of responding to the expansion of advertising operations, and the recent shipment flow is maintained above a certain level, the device (200) can analyze that the product's product-specific advertising suitability is high.

[0475] That is, the device (200) can generate a second advertising control condition for products that are suitable for expanding advertising operations and have a high probability of inventory accumulation.

[0476] For example, if the current inventory quantity of a specific product is above the excess inventory standard and the recent shipment volume continues to decrease, but the product-specific advertising suitability is analyzed to be high, the device (200) can generate a second advertising control condition including an increase in advertising exposure and an increase in the advertising budget.

[0477] Conversely, even if the risk of excess inventory is high, if the ad suitability for each product is low, it can be excluded from the target for generating the second ad control condition.

[0478] For example, if the inventory of a specific product is excessively accumulated but the ad click-through rate, conversion rate, or sales indicator relative to the ad cost remains consistently low, the device (200) can analyze that the effect of expanding the ad operation is limited and exclude it from the target for creating the second ad control condition.

[0479] At this time, the criteria for the risk of excess inventory may include, but are not limited to, the ratio of excess inventory to the current inventory quantity, the criteria for the expected period of inventory depletion, the criteria for the rate of decrease in demand, or the criteria for demand volatility.

[0480] In addition, the criteria for ad suitability by product may include, but are not limited to, criteria for average ad efficiency by ad channel, criteria for average conversion rate by product category, or criteria for ad operation policy.

[0481] Additionally, the device (200) can generate a second advertising control condition by taking into account the advertising operation characteristics of each advertising channel.

[0482] For example, if a specific advertising channel is analyzed as a channel with a high advertising exposure diffusion effect, the device (200) may preferentially apply an advertising exposure increase condition to that advertising channel.

[0483] Additionally, if a specific advertising channel is analyzed as having a high purchase conversion rate, the device (200) may prioritize applying an advertising budget increase condition to that advertising channel.

[0484] Conversely, for advertising channels where advertising efficiency remains consistently low, the level of increase in ad impressions or the level of increase in the ad budget may be limited.

[0485] Additionally, the device (200) can generate a second advertising control condition by taking into account the characteristics of each product category.

[0486] For example, in the case of product categories with strong seasonality, it is necessary to first deplete inventory before the end of the season, so the device (200) can set the level of increased ad exposure and the level of increased ad budget relatively high.

[0487] Conversely, for everyday consumer goods with a high repeat purchase rate, the advertising growth rate can be set relatively moderately to ensure long-term advertising operational stability.

[0488] Additionally, the device (200) can generate a second ad control condition in consideration of ad operation stability.

[0489] For example, even if the risk of excess inventory for a specific product is very high, criteria to limit the advertising growth rate can be applied to prevent ad impression levels or the ad budget from increasing excessively in the short term.

[0490] For example, the device (200) can generate a second ad control condition based on an ad exposure increase rate, an ad budget increase rate, or a maximum allowable increase range per ad channel.

[0491] At this time, the criteria for limiting the advertising growth rate may include, but are not limited to, advertising operation policies by product category, operation standards by advertising channel, or inventory depletion target standards, and may be set differently depending on the embodiment.

[0492] Additionally, the device (200) can generate a second ad control condition by considering the inventory level by category and the demand pattern data by category together.

[0493] For example, if the risk of excess inventory of a specific product is high and the state of reduced demand for the entire same category continues, the device (200) can set a higher level of increased ad exposure or increased ad budget to reflect the possibility of increased risk of inventory accumulation.

[0494] Conversely, if a state of increased demand is established across the entire same category, the level of advertising increase can be relatively mitigated by considering the possibility of a natural increase in shipments.

[0495] Additionally, the device (200) can provide the generated second advertisement control condition through the manager's terminal (100).

[0496] For example, the device (200) can display information on products scheduled for increased ad exposure, information on an increased ad budget, information on a risk of excess inventory, and information on the expected time of inventory depletion on the manager's terminal (100).

[0497] In addition, the device (200) can also provide a second advertising control condition generation cause.

[0498] For example, the device (200) can display information on the high inventory overload risk of a specific product, high product-specific advertising suitability, and the possibility of long-term inventory accumulation.

[0499] In addition, the second ad control condition may be applied separately from the first ad control condition, and the second ad control condition may be utilized to expand ad operations in order to resolve the inventory backlog.

[0500] Through this, the device (200) can automatically expand advertising operations for products with a high likelihood of inventory accumulation. Since the device (200) can induce inventory depletion while preventing unnecessary expansion of advertising for products with low actual advertising operation efficiency, it can improve both advertising operation efficiency and inventory operation efficiency.

[0501] In step S310, the device (200) can generate an advertising automation control value including an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value by selecting a condition according to a preset priority criterion among a first advertising control condition, a second advertising control condition, and an advertising exposure increase condition based on the remaining period of the sales period.

[0502] In the present invention, the priority criterion can be defined as reference data for determining which condition to apply first when multiple ad control conditions are simultaneously satisfied.

[0503] In addition, ad automation control values ​​can be applied to the actual ad operation environment and defined as execution control data to automatically change ad impression levels, ad budgets, and target products.

[0504] That is, the device (200) can generate a final ad automation control value by comparing multiple ad control conditions that conflict with each other or are satisfied simultaneously, and determining the priority, rather than simply applying each ad control condition independently.

[0505] For example, if the first ad control condition and the ad exposure increase condition based on the remaining period of the sales period are simultaneously satisfied for a specific product, the device (200) can determine which condition to apply first using a preset priority criterion.

[0506] For example, if the risk of a stock shortage of a specific product is very high and the first advertising control condition is satisfied, but at the same time the remaining period of the sales period is very short and priority shipment inducement is required, the device (200) can determine the priority application condition by comparing the level of stock shortage risk, the current stock quantity, the remaining period of the sales period, and the advertising suitability for each product together.

[0507] Additionally, the device (200) can generate an ad exposure level adjustment value according to priority criteria.

[0508] For example, if the priority of the first ad control condition is set high, an ad automation control value including an ad exposure reduction value can be generated.

[0509] Conversely, if the priority of the second ad control condition or the ad exposure increase condition based on the remaining period of the sales period is set high, an ad automation control value including the ad exposure increase value can be generated.

[0510] Additionally, the device (200) can generate an ad budget adjustment value based on priority criteria.

[0511] For example, if the risk of an overstock of a specific product is very high and the ad suitability for that product is high, you can generate an ad automation control that includes an increase in the ad budget. Conversely, if the risk of an understock of a specific product is very high and the expected time of stock depletion is imminent, you can generate an ad automation control that includes a decrease in the ad budget.

[0512] Additionally, the device (200) can generate an advertising target product change value.

[0513] For example, products with a very high risk of stock shortages can be excluded from advertising targets or have their advertising priority lowered. Conversely, products with a high risk of excess inventory and a short remaining sales period can be newly added to the advertising target list or changed to priority exposure targets.

[0514] At this time, the priority criteria may include, but are not limited to, the level of risk of stock shortage, the level of risk of stock excess, the remaining period of the sales period, the suitability of advertisements by product, the level of efficiency by advertisement, the expected time of stock depletion, or the advertising operation policy by product category.

[0515] In addition, the device (200) can generate advertising automation control values ​​considering advertising operation stability.

[0516] For example, if multiple ad control conditions are repeatedly changed, ad change limit criteria may be applied to prevent excessive fluctuations in ad impression levels or ad budgets within a short period.

[0517] For example, the device (200) can generate ad automation control values ​​based on the maximum allowable range of ad exposure increase and decrease, the maximum allowable range of ad budget increase and decrease, or the allowable range of variation per ad channel.

[0518] At this time, the criteria for limiting changes in advertising may include, but are not limited to, advertising operation policies by product category, operation standards by advertising channel, inventory operation stability standards, or advertising performance maintenance standards, and may be set differently depending on the embodiment.

[0519] In addition, the device (200) can generate advertising automation control values ​​by taking into account the characteristics of each advertising channel.

[0520] For example, if a specific ad channel is focused on purchase conversion, a higher weighting can be applied to the ad budget adjustment value. Conversely, if a specific ad channel is focused on ad impression diffusion, a higher weighting can be applied to the ad impression level adjustment value.

[0521] In addition, the device (200) can generate advertising automation control values ​​by taking into account the characteristics of each product category.

[0522] For example, for product categories with strong seasonality, the priority for increasing ad impressions can be set high by prioritizing the depletion of inventory before the end of the season. Conversely, for product categories with limited supply, the priority for decreasing ad impressions can be set high by prioritizing the risk of stock shortages.

[0523] Additionally, the device (200) can adjust conflicting ad control conditions during the ad automation control value generation process.

[0524] For example, if the conditions for increasing ad exposure and decreasing ad exposure are simultaneously satisfied for the same product, the device (200) can determine the final ad exposure level adjustment value by comparing the risk of stock shortage and the risk of stock excess and considering the remaining period of the sales period together.

[0525] For example, if the risk of the sales period expiring is higher than the risk of stock shortage, the increase in ad impressions can be maintained; conversely, if the risk of stock shortage is higher, the decrease in ad impressions can be applied first.

[0526] Additionally, the device (200) can provide the generated advertising automation control value through the manager's terminal (100).

[0527] For example, the device (200) can display information on the planned change of the ad exposure level, information on the planned adjustment of the ad budget, information on the planned change of the ad target product, and information on the priority applied ad control conditions on the administrator's terminal (100).

[0528] In addition, the device (200) can also provide a cause for generating advertising automation control values.

[0529] For example, the device (200) can provide a reason for generating advertising automation control values ​​by displaying together the risk of stock shortage, the risk of stock excess, the remaining period of the sales period, and the results of the advertising efficiency evaluation of a specific product.

[0530] In this way, the device (200) can generate an ad automation control value that reflects both inventory status and ad efficiency, or an ad automation control value adjusted according to priority criteria, by comparing priorities between a first ad control condition, a second ad control condition, and an ad exposure increase condition based on the remaining period of the sales period.

[0531] Through this, the device (200) can generate an advertising automation control value adjusted according to priority criteria even when multiple advertising control conditions exist simultaneously. Since the device (200) can automatically adjust advertising operation conditions by considering the possibility of inventory shortage, the possibility of inventory accumulation, and the sales period status together, it can simultaneously improve advertising operation efficiency and inventory operation stability.

[0532] Thus, the device (200) can analyze the inventory status and demand flow for each product from logistics data, and control the level of advertising exposure, the advertising budget, and the target products for advertising in accordance with the inventory operation status by reflecting the risk of inventory shortage, the risk of inventory excess, and the remaining period of the sales period together with the advertising efficiency evaluation data.

[0533] FIG. 4 is a flowchart illustrating the process of generating advertising plan data based on advertising efficiency evaluation data according to one embodiment.

[0534] Referring to FIG. 4, first, in step S401, the device (200) can calculate an advertising priority value for each advertising target product using the advertising efficiency level and product-specific advertising suitability and logistics data included in the advertising efficiency evaluation data.

[0535] In the present invention, the advertising priority value can be defined as an evaluation value indicating which product among a plurality of advertising target products will be preferentially allocated the level of advertising exposure and the advertising budget.

[0536] In this case, the advertising priority value may be a value calculated by reflecting the necessity of expanding advertising operations, the potential for advertising performance, the status of inventory management, and product sales flow.

[0537] In addition, the advertised product may refer to a product that is set as an advertising target or for which the expansion of advertising operations is being considered.

[0538] That is, the device (200) can calculate an advertising priority value for each advertising target product by considering logistics data including actual inventory operation status and outbound flow, rather than calculating an advertising priority based solely on advertising efficiency evaluation data.

[0539] For example, the device (200) can collect the efficiency level per advertisement and the suitability of the advertisement per product included in the advertising efficiency evaluation data.

[0540] In this case, the efficiency level per advertisement may be a value reflecting ad click-through rates, conversion rates, sales relative to advertising costs, or ad performance by advertising channel.

[0541] In addition, the advertising suitability for each product may be a value calculated by reflecting the current inventory quantity, shipment flow, estimated time of inventory depletion, and the feasibility of advertising operations.

[0542] In addition, the device (200) can use information including the current inventory quantity, shipment quantity, estimated time of inventory depletion, and sales period included in the logistics data.

[0543] For example, if the advertising efficiency level of a specific product is high, the advertising suitability for the product is high, and the current inventory quantity is sufficient, the device (200) can calculate a high advertising priority value for the product. Conversely, even if the advertising efficiency is high, if the current inventory quantity is insufficient or the expected time of inventory depletion is very imminent, the advertising priority value can be calculated low.

[0544] In addition, even if the advertising efficiency is at an average level, if the current inventory quantity is excessive and there is a high possibility of inventory accumulation, the advertising priority value can be calculated high to induce inventory depletion.

[0545] That is, the device (200) can calculate an advertising priority value by reflecting both the potential for advertising performance and the need for inventory management.

[0546] In addition, the device (200) can calculate an advertising priority value by taking into account the demand flow for each product.

[0547] For example, if the recent trend of increasing shipment volume of a specific product continues and demand volatility remains stable, the device (200) can calculate a high advertising priority value by analyzing that there is a high possibility of additional sales when expanding advertising operations. Conversely, if the recent trend of decreasing demand continues for a long period and the effect of expanding advertising operations is expected to be low, the advertising priority value can be calculated low.

[0548] Additionally, the device (200) can adjust the advertising priority value by using the category-specific inventory level and category-specific demand pattern data together.

[0549] For example, even if the advertising efficiency of a specific product itself is high, the ad priority value can be limited if there is a high probability of stock shortages across the entire category. Conversely, if there is a high probability of stock accumulation across the entire category, the ad priority value can be increased.

[0550] In addition, the device (200) can calculate an advertising priority value by considering the advertising operation characteristics of each advertising channel.

[0551] For example, if a specific product shows a high conversion rate in search advertising channels and a high exposure diffusion effect in social media advertising platforms, the device (200) can calculate an advertising priority value by reflecting the advertising performance characteristics of each advertising channel together.

[0552] Additionally, the device (200) may apply different criteria for calculating advertising priority values ​​depending on the purpose of advertising operation.

[0553] For example, if the primary goal is short-term sales growth, a higher weighting can be applied to the efficiency level of each advertisement. Conversely, if the primary goal is inventory depletion, a higher weighting can be applied to the current inventory quantity, the estimated time of depletion, and the risk of excess inventory.

[0554] At this time, the criteria for calculating the advertising priority value may include, but are not limited to, the reflection ratio of efficiency levels by advertisement, the reflection ratio of advertising suitability by product, inventory quantity criteria, the estimated time of inventory depletion criteria, the demand growth rate criteria, or advertising operation policy criteria.

[0555] Additionally, the device (200) can calculate an ad priority value in the form of a numerical value or a grade.

[0556] For example, the device (200) can calculate an ad priority value in the form of a score in the range of 0 to 100.

[0557] Additionally, the device (200) may classify ad priority values ​​into grades such as very high, high, normal, or low according to the ad priority level.

[0558] Additionally, the device (200) can correct temporary abnormal advertising performance data or abnormal inventory data during the process of calculating advertising priority values.

[0559] For example, if the number of ad clicks temporarily surged during a specific period but there was almost no increase in actual shipment volume, the device (200) can classify the data as temporary abnormal ad data and lower the reflection ratio when calculating the ad priority value.

[0560] Additionally, if the inventory data of a specific product temporarily increases or decreases abnormally due to a system error, the device (200) can calculate the advertising priority value after correcting or excluding the data.

[0561] At this time, the criteria for determining abnormal data may include, but are not limited to, a deviation range relative to average advertising performance, a deviation range relative to average shipment volume, an allowable variation range per advertising channel, or an inventory fluctuation rate range, and may be set differently depending on the embodiment.

[0562] Additionally, the device (200) can provide the calculated ad priority value through the manager's terminal (100).

[0563] For example, the device (200) can display advertising priority values, advertising efficiency status, inventory status, and demand flow status for each advertising target product in the form of a table, a graph, or a priority list on the manager's terminal (100).

[0564] Additionally, the device (200) can prioritize displaying products with high advertising priority values ​​or provide highlighting for products with a high need for expanded advertising operations.

[0565] Through this, the device (200) can calculate an advertising priority value for each advertising target product that reflects both the results of the advertising efficiency evaluation and the actual inventory operation status. Since the device (200) can consider both the need to expand advertising operations and the stability of inventory operations, it can improve the accuracy of subsequent adjustments to advertising exposure levels, adjustments to advertising budgets, and generation of advertising plan data.

[0566] In step S402, the device (200) can display, through the manager's terminal, an adjustment interface element for adjusting the advertising priority value for each advertising target product, the advertising exposure level corresponding to each product, and the advertising budget level.

[0567] In the present invention, the adjustment interface element may be defined as a user interface component for changing or adjusting the ad exposure level and the ad budget level.

[0568] At this time, the adjustment interface elements may include a slide bar, an input window, an increase / decrease button, a selection menu, a graph-based adjustment area, or a step-by-step selection interface, but are not limited thereto and may be configured differently depending on the embodiment.

[0569] Additionally, the level of ad exposure may include the frequency of ad exposure, the number of ad broadcasts, the priority of ad exposure, or the scope of ad exposure, and the level of ad budget may refer to the level of ad execution amount per product or per ad channel.

[0570] That is, the device (200) not only simply displays the ad priority value, but can also provide adjustment interface elements so that an administrator can directly adjust the ad operation status for each ad target product.

[0571] Additionally, the device (200) may provide adjustment interface elements so that the ad priority value, the ad exposure level, and the ad budget level can be adjusted in a linked state, thereby enabling an administrator to intuitively adjust the ad exposure level and the ad budget level according to changes in the ad operation priority.

[0572] For example, the device (200) can display the advertising priority value of each product along with the list of products to be advertised on the manager's terminal (100).

[0573] Additionally, the device (200) can display the current ad exposure level and the current ad budget level corresponding to each product.

[0574] For example, the device (200) can display the information together in the same screen area when the advertising priority value of a specific product is 90 points, the current advertising budget level is 500,000 won based on the daily budget, and the advertising exposure level is in a high exposure state.

[0575] Additionally, the device (200) may provide an adjustment interface element for adjusting the level of ad exposure.

[0576] For example, the device (200) may display an adjustment interface element in the form of a slide bar to increase or decrease the frequency of ad exposure.

[0577] In addition, adjustment interface elements in the form of a selection menu to raise or lower the priority of ad exposure may also be displayed.

[0578] Additionally, the device (200) may provide an adjustment interface element for adjusting the advertising budget level.

[0579] For example, the device (200) can display an adjustment interface element in the form of an input window for directly entering an advertising budget amount.

[0580] In addition, adjustment interface elements in the form of a step-by-step selection interface or increase / decrease buttons may be displayed to select the ratio of increase or decrease in the advertising budget.

[0581] Additionally, the device (200) can configure the display form of the adjustment interface element differently according to the advertisement priority value.

[0582] For example, for products with a high ad priority value, you can display them preferentially in the top area or provide accent colors, enlarged displays, or recommendations for priority adjustments. Conversely, for products with a low ad priority value, you can provide them in a default display format or display a lower level of adjustment recommendation.

[0583] In addition, the device (200) can display advertising efficiency evaluation data and logistics data together.

[0584] For example, the device (200) can display the efficiency level per advertisement, the suitability of the advertisement per product, the current stock quantity, the expected time of stock depletion, and the risk of stock shortage or stock excess.

[0585] Through this, managers can adjust ad impression levels and ad budget levels while simultaneously checking the ad operation status and inventory operation status.

[0586] In addition, the device (200) can display adjustment interface elements by reflecting the characteristics of each product category.

[0587] For example, for product categories with strong seasonality, information regarding the end of the season can be displayed. Conversely, for household consumer goods with high repeat purchase rates, information on recent changes in shipment trends can be displayed.

[0588] Additionally, the device (200) can display the advertising operation status by advertising channel separately. For example, the level of ad exposure and the level of ad budget for a search advertising channel and the level of ad exposure and the level of ad budget for an SNS advertising platform can be displayed in a form that is distinct from each other.

[0589] Additionally, the device (200) can display differences in advertising performance by advertising channel. For example, if a specific product shows a high conversion rate in a search advertising channel and a high exposure diffusion effect in an SNS advertising platform, the device (200) can display the corresponding advertising performance characteristics.

[0590] Additionally, the device (200) can consider advertising operation stability during the process of displaying adjustment interface elements.

[0591] For example, an acceptable adjustment range can be displayed to prevent excessive changes in ad impression levels or ad budget levels.

[0592] For example, the device (200) may display the range of possible increases in ad impressions, the maximum allowable increase rate of the ad budget, or the allowable adjustment range per ad channel together around the adjustment interface element.

[0593] At this time, the acceptable adjustment range may include, but is not limited to, advertising operation policies by product category, operation standards by advertising channel, inventory operation stability standards, or advertising performance maintenance standards, and may be set differently depending on the embodiment.

[0594] Additionally, the device (200) can display advertising operation result prediction information together with adjustment interface elements.

[0595] For example, changes in estimated ad clicks, estimated shipment volume, or estimated inventory depletion time can be displayed together with an increase in ad impression levels.

[0596] In addition, when the advertising budget level is changed, changes in sales or estimated inventory relative to the estimated advertising cost can also be displayed.

[0597] Through this, the device (200) can support an administrator in adjusting the level of ad exposure and the level of the ad budget while considering the results of the ad efficiency evaluation and the inventory operation status together. Since the device (200) can provide an intuitive adjustment interface element based on the ad priority value, it can improve the efficiency and accuracy of the ad planning data generation process.

[0598] In step S403, the device (200) can generate administrator setting values ​​for the ad exposure level and ad budget level based on the administrator input to the adjustment interface element.

[0599] In the present invention, the administrator setting value can be defined as a value set by the administrator by directly adjusting the ad exposure level and the ad budget level.

[0600] In this case, the administrator settings may include ad impression increase values, ad impression decrease values, ad budget increase values, ad budget decrease values, ad operation settings by ad channel, or ad operation settings by ad target product.

[0601] That is, the device (200) is not limited to simply displaying adjustment interface elements, but can generate administrator setting values ​​applicable to actual advertising operation plans by reflecting the administrator's input.

[0602] For example, the device (200) can receive an input for adjusting the ad exposure level entered through the manager's terminal (100).

[0603] For example, if an administrator adjusts a slider bar to increase the frequency of ad exposure for a specific product by 20% compared to the existing one, the device (200) can generate the corresponding input as an administrator setting value in the form of an ad exposure increase value.

[0604] Additionally, if an administrator inputs a selection menu to increase the advertising exposure priority of a specific product, the device (200) can generate the input as an administrator setting value in the form of an advertising priority exposure setting value.

[0605] Additionally, the device (200) can receive an input for adjusting the advertising budget level.

[0606] For example, if an administrator changes the daily advertising budget of a specific product from 500,000 won to 700,000 won, the device (200) can generate the changed advertising budget value as an administrator setting value in the form of an advertising budget increase value. Conversely, if an input for a decrease in the advertising budget is received considering the state of reduced advertising operation efficiency of a specific product, an administrator setting value in the form of an advertising budget decrease value can be generated.

[0607] In addition, the device (200) can generate administrator setting values ​​by distinguishing administrator inputs by advertising channel.

[0608] For example, if an administrator performs an input to increase the advertising budget for a search advertising channel and an input to decrease the advertising exposure level for a social media advertising platform, the device (200) can generate different administrator setting values ​​for each advertising channel.

[0609] In addition, the device (200) can independently generate input values ​​for each advertising target product.

[0610] For example, you can create a setting to increase ad impressions for a specific product and a setting to decrease the ad budget for another product.

[0611] In addition, the device (200) can generate administrator setting values ​​by considering the ad priority value and the ad efficiency evaluation data together.

[0612] For example, if an administrator performs an input to increase ad exposure for a product with a high ad priority value, the device (200) can apply a high reflection rate of the said administrator input. Conversely, if an excessive input to increase ad exposure is received for a product with a low ad priority value or a high risk of stock shortage, adjustment limit criteria can be applied together during the administrator setting value generation process.

[0613] For example, if an input is received that excessively increases the advertising budget while the risk of a specific product's stock shortage is very high, the device (200) can generate an administrator setting value by limiting the portion that exceeds a preset allowable range.

[0614] At this time, the allowable range may include, but is not limited to, the maximum increase rate of the advertising budget, the maximum increase range of the advertising exposure level, the allowable criteria for inventory shortage risk, or the criteria for advertising operation stability.

[0615] In addition, the device (200) can generate administrator setting values ​​by taking into account the administrator input history.

[0616] For example, if a specific manager repeatedly performs high ad budget increase inputs for a specific ad channel, the device (200) can adjust the criteria for generating manager settings per ad channel by reflecting the manager input pattern.

[0617] In addition, the device (200) can analyze the correlation between past advertising operation results and administrator input and reflect it in the process of generating administrator settings.

[0618] For example, if there is a history of a specific manager input pattern leading to an improvement in actual advertising performance, the device (200) can apply a relatively high reflection rate of the input pattern.

[0619] Additionally, the device (200) can correct abnormal administrator input during the process of generating administrator settings.

[0620] For example, if an input for an increase in the advertising budget that significantly exceeds the scope of the advertising operation policy is received, the device (200) may classify the input as an abnormal input and restrict it or generate warning information.

[0621] In addition, if conflicting ad exposure adjustment inputs occur repeatedly within a short period, the device (200) may limit the input reflection ratio for the stability of ad operations.

[0622] At this time, the criteria for determining abnormal input may include, but are not limited to, the maximum allowable growth rate per advertising channel, advertising operation policies per product category, inventory operation stability criteria, or advertising performance maintenance criteria, and may be set differently depending on the embodiment.

[0623] Additionally, the device (200) can provide the generated administrator setting value through the administrator's terminal (100).

[0624] For example, the device (200) can display information on the manager's terminal (100), including a value for a planned change in the ad exposure level, a value for a planned change in the ad budget level, an ad operation setting value per ad channel, and an adjustment result information per ad target product.

[0625] In addition, the device (200) can display the expected change in advertising performance or the expected change in inventory based on the application of the administrator setting value.

[0626] For example, when applying the setting to increase ad impressions, changes in the estimated increase in ad clicks, estimated increase in shipment volume, or estimated inventory depletion time can be displayed together.

[0627] Through this, the device (200) can generate administrator setting values ​​applicable to an advertising operation plan by reflecting administrator input to adjustment interface elements. Since the device (200) can generate administrator setting values ​​by considering the results of the advertising efficiency evaluation, the inventory operation status, and the stability of the advertising operation together, the accuracy of calculating changes in advertising performance and inventory quantity can be improved.

[0628] In step S404, the device (200) can calculate changes in advertising performance and inventory quantity corresponding to manager settings using advertising efficiency evaluation data and logistics data.

[0629] In the present invention, changes in advertising performance may be defined as data representing the state of changes in advertising operation performance before and after the application of administrator settings, and may include changes in the number of ad clicks, changes in the number of conversions, changes in sales relative to advertising costs, changes in sales revenue, or changes in performance by advertising channel.

[0630] Additionally, changes in inventory quantity can be defined as data representing changes in the product inventory status expected based on the application of administrator settings, and may include changes in expected shipment volume, expected inventory decrease, expected inventory increase, or expected time of inventory depletion.

[0631] That is, the device (200) does not simply store the manager's ad operation adjustment input, but can analyze the impact of said input on the actual ad operation and inventory operation status to calculate changes in ad performance and inventory quantity.

[0632] For example, the device (200) can calculate the expected change in advertising performance when applying administrator settings by using the efficiency level per advertisement and the suitability of the advertisement per product included in the advertising efficiency evaluation data.

[0633] For example, when a setting value for increasing the level of ad exposure for a specific product is entered, the device (200) can calculate the expected increase in ad clicks and the expected increase in conversions using past ad operation history and ad efficiency evaluation data.

[0634] In addition, if an advertising budget increase setting is entered, the change in sales relative to advertising costs or the expected increase in sales can be calculated.

[0635] Conversely, if a setting to reduce ad impressions or ad budget is entered, the expected decrease in ad performance can be calculated together.

[0636] In addition, the device (200) can calculate changes in advertising performance by reflecting the advertising operation characteristics of each advertising channel.

[0637] For example, in the case of search advertising channels, a higher weighting can be applied to expected changes in conversion rates or expected increases in purchases. Conversely, in the case of social media advertising platforms, a higher weighting can be applied to expected ad exposure diffusion effects or expected increases in ad clicks.

[0638] In addition, the device (200) can calculate the change in inventory quantity according to the application of manager setting values ​​using logistics data.

[0639] For example, when an increased ad exposure level setting is applied, the device (200) can calculate an expected increase in shipment volume and use this to calculate an expected decrease in inventory.

[0640] In addition, when the ad budget increase setting is applied, the change in the estimated time of inventory depletion can be calculated using the expected increase in ad traffic and the expected increase in purchase conversions.

[0641] For example, if the advertising budget level of a specific product increases by 30% compared to the existing level, the device (200) can calculate the expected increase in shipment volume and the expected decrease in inventory based on the next 14 days using past advertising operation results and shipment flow data. Conversely, if an advertising exposure reduction setting value or an advertising budget reduction setting value is applied, the expected decrease in shipment volume and the expected increase in inventory retention period can be calculated.

[0642] In addition, the device (200) can calculate changes in advertising performance and inventory quantity by taking into account demand pattern data.

[0643] For example, if the recent trend of increasing demand for a specific product continues, the device (200) can calculate a higher expected increase in advertising performance and an expected increase in shipment volume even with the same advertising exposure increase setting value. Conversely, if the recent trend of decreasing demand continues, the expected increase in advertising performance can be calculated relatively lower even with the same advertising budget increase setting value.

[0644] Additionally, the device (200) can correct changes in advertising performance and inventory quantity by using category-specific inventory levels and category-specific demand pattern data together.

[0645] For example, even if the advertising efficiency of a specific product itself is high, if the state of decreased demand for the entire same category continues, the device (200) can adjust the expected increase in advertising performance to a lower level. Conversely, if the state of increased demand for the entire same category is formed, the expected increase in advertising performance and the expected increase in shipment volume can be adjusted upward.

[0646] In addition, the device (200) can calculate changes in advertising performance and inventory quantity by taking into account the stability of advertising operations.

[0647] For example, if the level of ad exposure or the level of the ad budget increases excessively in a short period of time, the device (200) may limit the expected increase in ad performance by reflecting the possibility that the actual increase in ad performance may decrease non-linearly.

[0648] For example, if the advertising budget growth rate exceeds a certain level, the device (200) can calculate changes in advertising performance by gradually decreasing the growth rate of the number of ad clicks or the expected sales growth rate.

[0649] At this time, the criteria for calculating changes in advertising performance may include criteria for changes in ad click-through rates, criteria for changes in conversion rates, criteria for changes in sales relative to advertising costs, criteria for average performance by ad channel, or criteria for ad operation by product category, but are not limited thereto and may be set differently depending on the embodiment.

[0650] In addition, the criteria for calculating changes in inventory quantity may include, but are not limited to, the expected increase rate of shipments, criteria for changes in the expected period of inventory depletion, average shipment flow by product category, or criteria for demand volatility.

[0651] Additionally, the device (200) can correct temporary abnormal advertising data or abnormal inventory data during the process of calculating changes in advertising performance and inventory quantity.

[0652] For example, if the number of ad clicks temporarily surged during a specific period but there was almost no increase in actual shipment volume, the device (200) can classify the data as abnormal ad data and reduce the reflection rate in the process of calculating changes in ad performance.

[0653] Additionally, if the inventory data of a specific product temporarily increases or decreases abnormally due to a system error, the device (200) can calculate the change in inventory quantity after correcting or excluding the data.

[0654] Additionally, the device (200) can provide the calculated changes in advertising performance and inventory quantity through the manager's terminal (100).

[0655] For example, the device (200) can display changes in the expected number of ad clicks, expected sales, expected shipment volume, and expected inventory depletion time on the manager's terminal (100) in the form of a graph, a table, or comparison information.

[0656] Additionally, the device (200) can display the effect of increasing advertising performance and the possibility of a shortage of stock when a specific manager setting value is applied.

[0657] For example, if expected sales increase due to an increase in the advertising budget but the risk of inventory shortage increases at the same time, the device (200) can provide the risk status in the form of warning information.

[0658] Through this, the device (200) can predict in advance changes in advertising operations and inventory operations resulting from the application of administrator settings. Since the device (200) can calculate changes in advertising performance and inventory quantity by reflecting both the advertising efficiency evaluation results and the actual inventory operation status, it can improve the accuracy of subsequent advertising plan data generation and operational stability.

[0659] In step S405, the device (200) can provide changes in advertising performance and inventory quantity through the manager's terminal so as to correspond to the manager's settings for the advertising exposure level and advertising budget level.

[0660] In the present invention, providing a corresponding state may mean that expected changes in advertising performance and changes in inventory quantity according to specific manager setting values ​​are displayed in a connected state.

[0661] That is, the device (200) does not display changes in advertising performance and changes in inventory quantity independently, but can provide them in conjunction so that the manager can intuitively check which changes occur due to the manager's settings for the advertising exposure level and advertising budget level.

[0662] For example, the device (200) can display information on changes in the number of expected ad clicks, changes in the number of expected conversions, and changes in expected sales that correspond to a specific ad exposure level increase setting value.

[0663] In addition, information on changes in expected shipment volume, expected inventory reduction, and expected inventory depletion time corresponding to the same setting value can be displayed together.

[0664] For example, if an administrator setting value is entered to increase the level of ad exposure of a specific product by 20% compared to the existing level, the device (200) can provide information on an expected ad click rate increase rate of 15%, an expected shipment volume increase rate of 12%, and an expected time to shorten the expected inventory depletion date in a corresponding form.

[0665] In addition, if an administrator setting value that increases the advertising budget level of a specific product is entered, the device (200) can provide information on changes in sales relative to advertising costs and information on the expected reduction in inventory.

[0666] For example, if an administrator setting value is entered to increase the advertising budget level from 500,000 won to 700,000 won, the device (200) can display the expected sales increase, expected advertising cost increase, and expected inventory decrease in a linked state.

[0667] In addition, the device (200) can distinguish and provide changes in advertising performance and inventory quantity by advertising channel.

[0668] For example, the expected change in conversion rate corresponding to the increased ad budget setting for search ad channels and the expected change in ad clicks corresponding to the increased ad impression setting for social media ad platforms can be provided separately.

[0669] In addition, the device (200) can provide changes in expected shipment volume and expected inventory by advertising channel.

[0670] For example, if the effect of increasing ad exposure in a specific ad channel is high but the effect of increasing actual shipment volume is low, the device (200) can display the ad operation efficiency status of the said ad channel together.

[0671] In addition, the device (200) can provide changes in advertising performance and changes in inventory quantity in a visually comparable form.

[0672] For example, the device (200) can simultaneously display the expected sales change and the expected inventory reduction amount in the form of a graph as the level of ad exposure increases.

[0673] In addition, changes in sales relative to estimated advertising costs and changes in the estimated time of inventory depletion resulting from changes in the advertising budget level can be provided in the form of tables or comparative indicators.

[0674] Additionally, the device (200) can provide changes in advertising performance and inventory quantity to compare differences between administrator setting values.

[0675] For example, it is possible to provide a comparison of the expected changes in advertising performance and expected changes in inventory when increasing the level of ad exposure by 10% and when increasing it by 30%.

[0676] In addition, it is possible to compare and display the expected sales changes and expected inventory reductions corresponding to multiple manager settings that increase the advertising budget level in stages.

[0677] Through this, managers can determine the direction of advertising operations by comparing changes in advertising performance and inventory management based on the level of advertising operation expansion.

[0678] Additionally, the device (200) can provide changes in advertising performance and inventory quantity by reflecting both the inventory shortage risk state and the inventory excess risk state.

[0679] For example, if applying a specific manager setting value results in a high expected sales increase effect but a rapid increase in the risk of inventory shortage, the device (200) may display the risk state in the form of warning information. Conversely, if applying an ad exposure increase setting value results in a high expected inventory reduction effect while the risk of excess inventory of a specific product is high, the device (200) may display a state with a high possibility of inducing inventory depletion in the form of highlight information.

[0680] In addition, the device (200) can provide changes in advertising performance and inventory quantity by taking into account the characteristics of each product category.

[0681] For example, for product categories with strong seasonality, the likelihood of expected inventory depletion before the end of the season can be displayed. Conversely, for household consumer goods with high repeat purchase rates, changes in the expected inventory retention period can be displayed during long-term advertising campaigns.

[0682] In addition, the device (200) can provide changes in advertising performance and inventory quantity in consideration of advertising operation stability.

[0683] For example, when a specific manager setting value is applied, if the range of change in the ad exposure level or ad budget level is excessive, the device (200) may also indicate the possibility of increased ad operation variability.

[0684] In addition, if the rate of inventory reduction is excessively fast compared to the expected increase in advertising performance, information on the risk of inventory shortage can be provided.

[0685] At this time, the advertising operation stability criteria may include, but are not limited to, the allowable fluctuation range per advertising channel, the maximum increase rate of the advertising budget, the allowable inventory shortage criteria, or the advertising operation policy criteria, and may be set differently depending on the embodiment.

[0686] Additionally, the device (200) can store changes in advertising performance and inventory quantity along with manager settings.

[0687] For example, the device (200) can store expected advertising performance changes and expected inventory changes corresponding to specific manager settings in the form of historical data.

[0688] In addition, the corresponding historical data can be utilized in the subsequent process of analyzing administrator input patterns or adjusting recommended control values.

[0689] Through this, the device (200) can intuitively provide changes in advertising operations and inventory operations based on the administrator's settings. Since the device (200) can provide changes in advertising performance and inventory status that can be compared together, it can support the administrator in setting advertising plan data by considering both advertising operation efficiency and inventory operation stability.

[0690] In step S406, the device (200) can set constraints to ensure that the ad exposure level and ad budget are included within a preset allowable range by using the inventory quantity and the expected time of inventory depletion included in the logistics data.

[0691] In the present invention, constraints can be defined as condition data to limit the level of ad exposure and the level of the ad budget so that they do not exceed the inventory operation state.

[0692] At this time, the constraints may include, but are not limited to, conditions for limiting the level of ad exposure, conditions for limiting the ad budget, conditions for limiting ad operation by ad channel, conditions for preventing stock shortage or stock excess, and may be set differently depending on the embodiment.

[0693] In addition, the allowable range may refer to the standard range set to ensure that ad impression levels and ad budget levels can maintain normal ad and inventory operations.

[0694] That is, the device (200) does not apply the manager's ad operation settings as they are, but can set constraints to prevent the ad operation from being excessively expanded or reduced by taking into account the current inventory status and the expected time of inventory depletion.

[0695] For example, the device (200) can analyze the current stock quantity of a specific product and the expected time of stock depletion.

[0696] For example, if the current stock quantity of a specific product is close to the safety stock standard and the expected time of stock depletion is very imminent, the device (200) can set restrictions on increasing the level of ad exposure and increasing the ad budget.

[0697] For example, the device (200) may set constraints to limit the increase rate of ad exposure to a maximum of 10% or less, or to limit the increase rate of the ad budget to a specific allowable range or less.

[0698] Conversely, if the current inventory quantity of a specific product is above the excess inventory threshold and the expected time of inventory depletion is calculated to be after a long period, the device (200) can set constraints that expand the allowable range for increased ad exposure and increased ad budget.

[0699] In addition, the device (200) can set constraints by considering both the risk of stock shortage and the risk of stock excess.

[0700] For example, for products with a very high risk of stock shortage, restrictions on increasing ad impression levels or ad budgets may be tightened. Conversely, for products with a very high risk of stock oversupply, the permissible range for increasing ad impressions or ad budgets may be expanded.

[0701] In addition, the device (200) can set constraints by considering the characteristics of each product category.

[0702] For example, for product categories with strong seasonality, the allowable range for increasing ad impressions can be set relatively high to prioritize depleting inventory before the end of the season. Conversely, for product categories with limited supply, the allowable range for increasing the ad budget can be set relatively low to prevent stock shortages.

[0703] In addition, the device (200) can set constraints by considering the advertising operation characteristics of each advertising channel.

[0704] For example, in the case of an advertising channel with a high purchase conversion effect, since an increase in the level of ad exposure is likely to quickly lead to an increase in actual shipment volume, the device (200) can set stricter ad increase limit conditions for the said advertising channel.

[0705] Conversely, in the case of brand awareness-focused advertising channels, the impact on actual shipment volume increases may be relatively low, so the allowable range for increased ad exposure can be set relatively wide.

[0706] In addition, the device (200) can set constraints considering the stability of advertising operations.

[0707] For example, if the level of ad exposure or the level of the ad budget fluctuates rapidly within a short period, the stability of ad operation may be reduced, so the device (200) can set a limit condition on the ad change rate.

[0708] For example, the device (200) can set constraints based on the maximum increase rate of the ad exposure level, the maximum increase rate of the ad budget, the maximum allowable range of variation per ad channel, or the amount of change in ad operation that is allowable within a certain period.

[0709] In addition, the device (200) can set constraints by taking into account the rate of change of the expected time of stock depletion.

[0710] For example, if it is expected that the time of expected inventory depletion will be excessively shortened due to an increase in the level of ad exposure, the device (200) may automatically reduce the allowable range for increasing ad exposure. Conversely, if there is little change in the time of expected inventory depletion even after the expansion of ad operations, the allowable range for increasing the ad budget may be expanded.

[0711] At this time, the criteria for setting the allowable range may include, but are not limited to, the safety stock standard ratio relative to the current stock quantity, the excess stock standard ratio, the expected stock depletion period standard, the average conversion rate standard per advertising channel, the advertising operation policy standard, or the operation standard per product category.

[0712] Additionally, the device (200) can correct abnormal inventory data or abnormal advertising operation data during the constraint setting process.

[0713] For example, if the inventory quantity is abnormally reduced due to a system error during a specific period, the device (200) can set constraints after correcting or excluding the data.

[0714] In addition, if the number of ad clicks has increased rapidly but there has been almost no increase in the actual shipment volume, the device (200) can classify the ad data as abnormal ad data and adjust the reflection ratio during the process of setting the ad increase limit criteria.

[0715] Additionally, the device (200) can provide the set constraints through the manager's terminal (100).

[0716] For example, the device (200) can display information on the manager's terminal (100), including an allowable range for ad exposure levels, an allowable range for ad budgets, an allowable adjustment range per ad channel, and an inventory operation restriction status.

[0717] Additionally, the device (200) may display the reason for restriction or the risk of stock shortage when a specific manager setting value exceeds the allowable range.

[0718] For example, if the advertising budget increase setting value of a specific product exceeds the allowable range, the device (200) may provide information in the form of warning information, such as an increased possibility of expected stock shortage or a shortened expected time for stock depletion.

[0719] Through this, the device (200) can control the allowable range of the ad exposure level and the ad budget level by reflecting the current inventory status and the expected time of inventory depletion. Since the device (200) can consider the risk of inventory shortage due to the expansion of ad operations or the risk of inventory accumulation due to the reduction of ad operations, it can simultaneously improve the stability of ad operations and inventory operations.

[0720] In step S407, the device (200) can generate a recommended control value by automatically adjusting the administrator setting value according to the administrator input to satisfy the constraint.

[0721] In the present invention, the recommended control value may be defined as an ad operation control value that is automatically corrected to satisfy constraints, based on the administrator setting value entered by the administrator.

[0722] In this case, the recommendation control values ​​may include ad impression level adjustment values, ad budget adjustment values, ad operation adjustment values ​​by ad channel, or ad operation adjustment values ​​by ad target product.

[0723] Additionally, the administrator setting value may be an ad operation setting value that directly reflects the administrator's input, and the recommended control value may be an ad operation control value that is automatically corrected for the administrator setting value by reflecting the inventory quantity, expected time of inventory depletion, inventory shortage risk status, inventory excess risk status, and constraints.

[0724] That is, the device (200) can generate recommended control values ​​applicable to actual ad operations by considering the current inventory status, the expected time of inventory depletion, and the stability of ad operations, rather than simply applying the administrator input.

[0725] For example, the device (200) can determine whether the administrator setting value satisfies the constraint.

[0726] For example, if the setting value for increasing the ad exposure level of a specific product exceeds the allowable range for increasing the ad exposure, the device (200) can automatically adjust the manager setting value to generate a recommendation control value.

[0727] For example, if an administrator performs an input to increase the ad exposure level of a specific product by 50% compared to the existing level, but the allowable ad exposure increase range is set to 20% or less based on the current inventory quantity and the expected time of inventory depletion, the device (200) can automatically adjust the ad exposure level increase value to 20% or less to generate a recommendation control value.

[0728] That is, in this case, the administrator setting value may be a 50% increase value entered by the administrator, and the recommended control value may be a 20% increase value automatically adjusted to satisfy the constraint.

[0729] Additionally, if an administrator performs an input that excessively increases the advertising budget level of a specific product, the device (200) can automatically adjust the advertising budget adjustment value based on the allowable range for increasing the advertising budget.

[0730] Conversely, if the current inventory quantity exceeds the excess inventory threshold and there is a high possibility of inventory accumulation, recommendation control values ​​can be generated by expanding the allowable range for increasing ad impressions.

[0731] In addition, the device (200) can generate a recommendation control value by considering advertising efficiency evaluation data and logistics data together.

[0732] For example, if the level of advertising efficiency for a specific product is high and the suitability of the product for advertising is high, the device (200) may apply a relatively small automatic adjustment range for the administrator setting value. Conversely, if the advertising efficiency is low but the risk of stock shortage is high, the setting value for increasing advertising exposure or the setting value for increasing the advertising budget may be restricted more significantly.

[0733] In addition, the device (200) can generate recommendation control values ​​by considering the advertising operation characteristics of each advertising channel.

[0734] For example, in the case of an advertising channel with a high purchase conversion effect, the increase in the level of ad exposure has a significant impact on the increase in actual shipment volume, so the device (200) can relatively strengthen the limit on the increase in ad exposure. Conversely, in the case of a brand awareness-centered advertising channel, the limit on the increase in ad exposure can be relatively relaxed.

[0735] In addition, the device (200) can generate recommendation control values ​​by considering the characteristics of each product category.

[0736] For example, for product categories with strong seasonality where there is a high need to clear inventory before the end of the season, recommendation controls can be generated by expanding the allowable range for increased ad impressions. Conversely, for product categories with limited supply, the limit on ad budget increases can be tightened by prioritizing the risk of stock shortages.

[0737] In addition, the device (200) can generate recommendation control values ​​by considering the stability of advertising operations.

[0738] For example, since the variability of ad operations may increase when the administrator setting value is repeatedly and rapidly changed, the device (200) can generate a recommendation control value based on the maximum allowable change amount of the ad exposure level or ad budget level within a certain period.

[0739] For example, if the level of ad exposure repeatedly increases and decreases within a short period, the device (200) can generate a recommendation control value in a direction that mitigates the amount of ad change in order to maintain ad operation stability.

[0740] In addition, the device (200) can generate a recommendation control value by taking into account the results of changes in advertising performance and changes in inventory quantity.

[0741] For example, if the expected sales increase effect is high when a specific manager setting value is applied but the risk of inventory shortage increases rapidly, the device (200) can generate a recommended control value that automatically reduces the level of increase in ad impressions or the level of increase in ad budget. Conversely, if it is analyzed that the expected inventory accumulation state will continue, it can generate a recommended control value that automatically increases the level of increase in ad impressions or the level of increase in ad budget.

[0742] At this time, the criteria for generating recommendation control values ​​may include, but are not limited to, an allowable range for increasing ad impressions, an allowable range for increasing ad budget, criteria for inventory shortage risk, criteria for inventory excess risk, criteria for ad operation stability, or criteria for operation per ad channel.

[0743] Additionally, the device (200) can correct abnormal manager input or abnormal ad operation data during the recommendation control value generation process.

[0744] For example, if an input for an increase in the advertising budget that significantly exceeds the scope of the advertising operation policy is received, the device (200) can classify the input as an abnormal manager input and strengthen the automatic adjustment rate.

[0745] In addition, if the number of ad clicks on a specific ad channel has increased rapidly but there has been almost no increase in the actual shipment volume, the device (200) can classify the ad data as abnormal ad data and adjust the reflection ratio during the recommendation control value generation process.

[0746] Additionally, the device (200) can provide the generated recommended control value through the manager's terminal (100).

[0747] For example, the device (200) can display information on the difference between the administrator setting value and the recommendation control value, information on the automatically adjusted ad exposure level, information on the automatically adjusted ad budget level, and information on the result of applying constraints on the administrator's terminal (100).

[0748] In addition, the device (200) can also provide a cause for generating a recommended control value.

[0749] For example, the device (200) may provide a reason for generating a recommendation control value by displaying together a stock shortage risk state, a stock excess risk state, an ad operation stability limitation state, or a change in the expected time of stock depletion.

[0750] Through this, the device (200) can generate recommended control values ​​based on administrator input while satisfying constraints. Since the device (200) can automatically adjust the ad operation control values ​​by considering both ad operation efficiency and inventory operation stability, the stability and reliability of ad plan data generation can be improved.

[0751] In step S408, the device (200) can correct the recommended control value using the administrator input history and administrator setting information collected from the administrator's terminal.

[0752] In the present invention, the manager input history may be defined as historical data regarding inputs for adjusting ad exposure levels, adjusting ad budget levels, changing advertised products, or adjusting ad operations by ad channel that were performed by the manager in the past.

[0753] In addition, administrator setting information may include ad operation standards, ad budget operation standards, ad channel operation standards, or ad operation policy information by product category that the administrator prefers or uses repeatedly.

[0754] That is, the device (200) can correct the recommendation control value by reflecting the actual manager's operational tendencies and past advertising operation patterns together, rather than generating the recommendation control value based only on constraints.

[0755] For example, the device (200) can analyze the range of increased ad exposure levels and the range of increased ad budgets that a specific manager has repeatedly selected in the past.

[0756] For example, if a specific manager maintains an input pattern that repeatedly increases the level of ad exposure to a level higher than the average for products with a high risk of excess inventory, the device (200) can adjust the level of increase in ad exposure of the recommendation control value upward by reflecting the manager's input pattern.

[0757] Conversely, if a specific manager repeatedly performs an input pattern that limits the increase in the advertising budget while in a state of stock shortage risk, the device (200) can correct the level of increase in the advertising budget during the recommendation control value generation process.

[0758] Additionally, the device (200) can correct the recommended control value by analyzing the manager input pattern for each advertising channel.

[0759] For example, if a specific manager frequently performs an increase in the advertising budget for a search advertising channel and repeatedly performs adjustments centered on the level of ad exposure for an SNS advertising platform, the device (200) can correct the recommendation control value by reflecting the operational tendency of each advertising channel.

[0760] Additionally, if there is a manager input pattern that leads to an improvement in actual advertising performance in a specific advertising channel, the device (200) can adjust the recommendation control value reflection ratio for the advertising channel to a high level.

[0761] In addition, the device (200) can correct the recommended control value by analyzing the manager input pattern for each product category.

[0762] For example, if a specific manager repeatedly selects to increase the level of ad exposure before the end of the season for a product category with strong seasonality, the device (200) can adjust the level of increase in ad exposure upward when generating recommendation control values ​​for products of the same category.

[0763] Conversely, if a specific manager maintains an input pattern that limits the increase in the advertising budget for a product category with a limited supply, the device (200) can adjust the recommendation control value in a direction that limits the increase in the advertising budget for that category.

[0764] Additionally, the device (200) can correct recommendation control values ​​by analyzing the correlation between past administrator settings and actual ad operation results.

[0765] For example, if there is a history of improvement in ad click-through rates, conversion rates, or sales metrics relative to advertising costs after applying specific administrator settings, the device (200) can apply a high reflection rate of the corresponding administrator setting pattern.

[0766] Conversely, if a decrease in advertising performance or an increase in the risk of inventory shortage occurs repeatedly compared to expectations after applying a specific manager setting value, the device (200) may apply a low reflection rate of the corresponding manager setting pattern.

[0767] Additionally, the device (200) can correct the recommendation control value by taking into account the results of changes in advertising performance and changes in inventory quantity.

[0768] For example, if a specific manager input pattern shows a high increase in advertising performance while having a low increase in the risk of stock shortage, the device (200) can correct the recommendation control value by reflecting the input pattern first.

[0769] Conversely, for input patterns that have a low effect on increasing advertising performance but a large increase in the risk of stock shortages, the reflection ratio of recommendation control values ​​can be lowered.

[0770] In addition, the device (200) can correct the recommended control value by reflecting the difference in operational tendencies of each manager.

[0771] For example, if a specific manager has an operational tendency to prioritize stable inventory management, the device (200) can adjust the level of increase in ad exposure or the level of increase in ad budget relatively conservatively. Conversely, if a specific manager has an operational tendency to prioritize short-term sales growth, the device (200) can adjust the level of increase in ad exposure or the level of increase in ad budget relatively actively.

[0772] At this time, the administrator's operational tendencies may be analyzed based on past administrator input history, history of changes in advertising operation results, setting patterns by advertising channel, or setting patterns by product category, but are not limited thereto and may be analyzed differently depending on the embodiment.

[0773] Additionally, the device (200) can correct abnormal manager input patterns during the recommended control value correction process.

[0774] For example, if a specific manager repeatedly performs an ad budget increase input that exceeds the scope of the ad operation policy, the device (200) can limit the reflection rate of the input pattern.

[0775] In addition, if an input to increase the advertising budget is repeatedly performed on a specific advertising channel regardless of the improvement of advertising performance, the device (200) can classify the input pattern as an abnormal operation pattern and reduce the reflection rate in the recommendation control value correction process.

[0776] At this time, the criteria for determining abnormal operation patterns may include, but are not limited to, advertising operation policy criteria, allowable increase range per advertising channel, inventory operation stability criteria, or advertising performance maintenance criteria.

[0777] Additionally, the device (200) can provide a corrected recommended control value through the manager's terminal (100).

[0778] For example, the device (200) can display information on comparison before and after correction of recommendation control values, information on reflection of administrator input patterns, information on reflection ratios by advertising channel, and information on correction results by product category on the administrator's terminal (100).

[0779] In addition, the device (200) can also provide a cause for correcting the recommended control value.

[0780] For example, the device (200) can provide a reason for correcting the recommended control value by displaying together the history of past ad operation performance improvement, the pattern of reflecting manager input, the inventory shortage risk status, or the ad operation stability status.

[0781] Through this, the device (200) can generate recommended control values ​​suitable for the actual operating environment by reflecting the administrator input history and administrator setting information. Since the device (200) can correct the recommended control values ​​by considering the ad operation tendencies of each administrator and the actual ad operation results together, it can improve the accuracy of ad plan data generation and operational suitability.

[0782] In step S409, the device (200) can select one of the recommended control value and the administrator setting value according to the administrator selection input or preset selection criteria, and generate advertising plan data by reflecting the selected value.

[0783] In the present invention, the selection criterion can be defined as reference data for determining which value to apply first among the recommended control value or the administrator setting value.

[0784] At this time, the selection criteria may include criteria for inventory shortage risk, criteria for inventory excess risk, criteria for advertising operation stability, criteria for advertising efficiency, criteria for administrator approval, or criteria for operation per advertising channel, but are not limited thereto and may be set differently depending on the embodiment.

[0785] In addition, advertising plan data can be defined as advertising operation plan data including advertising target products, advertising exposure levels, advertising budget levels, advertising operation information by advertising channel, or advertising priority information.

[0786] Additionally, the advertising plan data can be used as reference data or intermediate planning data to generate advertising automation control values ​​thereafter, and the device (200) can generate advertising automation control values ​​including an advertising exposure level adjustment value, an advertising budget adjustment value, or an advertising target product change value based on the advertising plan data.

[0787] That is, the device (200) can generate advertising plan data by selecting a final advertising operation value to be applied to actual advertising operations, rather than simply storing recommendation control values ​​or administrator setting values.

[0788] For example, the device (200) can receive an administrator selection input for a recommended control value or an administrator setting value through the administrator's terminal (100).

[0789] For example, if the administrator selects a recommendation control value, the device (200) can generate advertising plan data by applying the recommendation control value as the final advertising operation value. Conversely, if the administrator selects an administrator setting value, the device (200) can generate advertising plan data by applying the administrator setting value as the final advertising operation value.

[0790] Additionally, the device (200) can automatically select one of the recommended control value or the administrator setting value according to preset selection criteria without administrator selection input.

[0791] For example, if the risk of a specific product being out of stock is very high, the device (200) can automatically select a recommendation control value by prioritizing the stability of advertising operations.

[0792] Conversely, if the inventory operation status is stable and the advertising operation restriction conditions are low, the administrator settings can be applied first.

[0793] In addition, the device (200) can apply selection criteria by considering advertising efficiency evaluation data and logistics data together.

[0794] For example, if the efficiency level per advertisement of a specific product is high but the risk of stock shortage is very high, the device (200) may select a recommendation control value that limits the level of increase in ad exposure by prioritizing the stability of inventory management.

[0795] Conversely, if the risk of excess inventory is high and the ad fit for each product is high, you can select recommendation control values ​​that reflect the level of increased ad impressions and increased ad budget.

[0796] In addition, the device (200) can select a recommended control value or an administrator setting value by considering the operational characteristics of each advertising channel.

[0797] For example, for advertising channels with high purchase conversion rates, recommendation control values ​​can be selected with priority given to the risk of stock shortages. Conversely, for brand awareness-focused advertising channels, the reflection ratio of administrator settings can be applied relatively higher.

[0798] In addition, the device (200) can select a recommended control value or an administrator setting value by considering the characteristics of each product category.

[0799] For example, for product categories with strong seasonality where there is a high need to clear inventory before the end of the season, you can prioritize recommendation controls that reflect an increase in ad impressions. Conversely, for product categories with limited supply, you can prioritize recommendation controls that reflect restrictions on ad budget increases.

[0800] Additionally, the device (200) can select a recommended control value or an administrator setting value in consideration of the stability of the advertisement operation.

[0801] For example, if a specific administrator setting value exceeds the scope of the advertising operation policy or if the level of ad exposure or the level of the ad budget changes excessively within a short period, the device (200) may prioritize selecting a recommended control value for the stability of advertising operations. Conversely, if the administrator setting value satisfies the constraints and does not impede inventory operation stability, the administrator setting value may be applied as is.

[0802] Additionally, the device (200) can generate advertising plan data based on the selected value.

[0803] For example, the device (200) can generate advertising operation plan data per advertising channel using selected advertising exposure level adjustment value, advertising budget adjustment value, and advertising target product setting value.

[0804] In addition, it is possible to generate advertising plan data that includes advertising priority values ​​for each target product, advertising exposure level information, advertising budget level information, and expected advertising performance information.

[0805] In addition, the device (200) can reflect the results of changes in advertising performance and changes in inventory quantity together during the process of generating advertising plan data.

[0806] For example, if the expected risk of a shortage of stock increases rapidly when a specific setting value is applied, the device (200) can generate advertising plan data after automatically adjusting the level of ad exposure or the level of the ad budget.

[0807] Conversely, if the risk of excess inventory is expected to persist for a long period, advertising plan data can be generated by reflecting the level of increase in ad impressions or the level of increase in the ad budget.

[0808] Additionally, the device (200) can provide advertising plan data through the manager's terminal (100).

[0809] For example, the device (200) can display the final selected ad exposure level information, ad budget level information, ad target product information, and ad operation plan information by ad channel on the manager's terminal (100).

[0810] Additionally, the device (200) can provide information on the difference between the recommended control value and the manager setting value, along with the reason for the final selection.

[0811] For example, the device (200) may provide a reason for the final selection by displaying together the inventory shortage risk state, inventory excess risk state, advertising operation stability state, or advertising efficiency evaluation results.

[0812] Through this, the device (200) can generate advertising plan data by selecting a value suitable for the actual advertising operation environment among the recommended control value and the manager setting value. Since the device (200) can generate advertising plan data that reflects the advertising efficiency evaluation result, inventory operation status, and advertising operation stability together, it can simultaneously improve advertising operation efficiency and inventory operation stability.

[0813] Thus, the device (200) can calculate an advertising priority value based on advertising efficiency evaluation data and logistics data, and generate advertising planning data that considers advertising operation efficiency and inventory operation stability by reflecting adjustment input, constraints, and recommendation control values ​​through the manager's terminal (100).

[0814] FIG. 5 is a flowchart illustrating the creator matching and performance prediction process for an advertising target product according to one embodiment.

[0815] Specifically, the device (200) can perform creator matching and performance prediction based on the ad exposure level and ad budget for the ad target product specified by the ad automation control value, prior to the step of changing at least one of the ad exposure level, ad budget and ad target product according to the ad automation control value.

[0816] Referring to FIG. 5, first, in step S501, the device (200) can use creator performance data including creator-specific content response data collected from an SNS advertising platform and partnership performance data stored in an ERP system, and can generate a group of creator candidates suitable for the advertising target product based on the product category, content response, and past advertising performance of the advertising target product.

[0817] In the present invention, creator performance data can be defined as data representing content operation performance, advertising response performance, and past collaboration performance by creator.

[0818] At this time, creator performance data may include content views, likes, comments, shares, content saves, click-through rate, conversion rate, sales relative to advertising costs, collaboration history, or past advertising execution performance data, but is not limited thereto and may be configured differently depending on the embodiment.

[0819] In addition, content response data can be defined as data representing the results of users' reactions to content posted by a specific creator.

[0820] For example, content response data may include content views, average watch time, like rate, comment response rate, share rate, link click-through rate, or purchase conversion rate.

[0821] In addition, partnership performance data can be defined as data representing advertising performance generated following advertising collaboration with a specific creator.

[0822] For example, partnership performance data may include the increase in sales after advertising execution, changes in ad conversion rates, changes in sales relative to advertising costs, the influx of new customers, or changes in repurchase rates.

[0823] That is, the device (200) can generate a group of candidate creators suitable for the advertised product by reflecting actual advertising performance and product suitability together, rather than selecting creators based solely on the number of followers or content views.

[0824] For example, the device (200) can collect content response data by creator from an SNS advertising platform.

[0825] For example, you can collect data on a specific creator's content views, average number of likes, comment response rate, and link click-through rate.

[0826] In addition, the device (200) can collect past partnership performance data stored in the ERP system.

[0827] For example, data on ad click-through rates, conversion rates, revenue relative to advertising costs, and actual revenue growth can be collected regarding the results of past advertising campaigns in collaboration with specific creators.

[0828] Additionally, the device (200) can analyze the product category of the product to be advertised. In the present invention, the product category may refer to the product type or product classification information of the product to be advertised.

[0829] For example, product categories may include, but are not limited to, food categories, beauty categories, fashion categories, household goods categories, or electronic product categories, and may be classified differently depending on the embodiment.

[0830] In addition, the device (200) can analyze the content responsiveness.

[0831] In the present invention, content responsiveness may be defined as an evaluation value representing the degree to which users respond to the content of a specific creator. For example, content responsiveness may be calculated based on the ratio of likes to views, comment engagement rate, sharing rate, content save rate, link click-through rate, or purchase conversion rate.

[0832] Additionally, the device (200) can analyze past advertising performance. For example, if a specific creator has recorded a high advertising conversion rate or high sales performance relative to advertising costs for the same product category in the past, the device (200) can evaluate the creator's advertising performance level as high. Conversely, if the content views are high but the actual advertising conversion performance is low, the advertising performance level can be evaluated as low.

[0833] Additionally, the device (200) can determine whether to include a creator in the candidate group by analyzing the level of product category match, the level of content responsiveness, and the level of past advertising performance, respectively. For example, a creator who consistently posts content in a product category identical or similar to the advertised product and exhibits high content responsiveness and high advertising conversion performance may have their priority for inclusion in the candidate group set high. Conversely, if the product category relevance is low or the content responsiveness or past advertising performance is below a preset standard, the priority for inclusion in the candidate group may be lowered or the creator may be excluded from the candidate group.

[0834] Additionally, the device (200) can generate a group of creator candidates by analyzing the association between product categories and creator content characteristics.

[0835] For example, if the product to be advertised belongs to the beauty category and a specific creator consistently shows high content responsiveness in beauty-related content, the device (200) may include the creator in the pool of creator candidates suitable for the product to be advertised. Conversely, if the creator mainly posts content that is not relevant to the product to be advertised or has low past advertising performance, the creator may be excluded from the pool of creator candidates.

[0836] Additionally, the device (200) can generate a pool of creator candidates by taking into account the level of ad exposure and the level of the ad budget, which are determined by the ad automation control value.

[0837] For example, if the advertising budget is relatively low, creators with low estimated ad rates can be included in the priority pool. Conversely, if expanding ad exposure is necessary, creators with high content diffusion potential can be included in the priority pool.

[0838] In addition, the device (200) can generate a group of creator candidates by considering the content exposure characteristics of each SNS advertising platform.

[0839] For example, if a specific social media advertising platform is a platform centered on short video content, the device (200) may include creators who show high content responsiveness in short video content in the priority candidate group. Conversely, in the case of an image-based content-centered platform, creators who show high image content responsiveness may be included in the priority candidate group.

[0840] Additionally, the device (200) can generate a group of creator candidates by considering the inventory status and logistics data for each product category together.

[0841] For example, if there is a high risk of excess inventory for a specific product, creators with a high content diffusion effect within a short period can be included in the priority candidate pool. Conversely, if there is a high risk of shortage inventory for a specific product, creators with excessively high ad conversion rates can be excluded from the candidate pool or have their priority lowered.

[0842] Additionally, the device (200) can correct abnormal reaction data during the process of generating a creator candidate group.

[0843] For example, if the number of views of a specific creator's content is very high but the comment response rate or purchase conversion rate is significantly low, the device (200) can adjust the content response rate by reflecting the possibility of abnormal response data.

[0844] Additionally, if an abnormal increase in ad clicks or a repetitive false response pattern is detected, the device (200) may exclude the creator in question from the candidate pool generation target.

[0845] At this time, the criteria for determining adverse reaction data may include, but are not limited to, the average comment participation rate, the average purchase conversion rate, the content reaction pattern, or the allowable reaction range for each SNS advertising platform.

[0846] Additionally, the device (200) can provide the generated creator candidate group through the manager's terminal (100).

[0847] For example, the device (200) can display content responsiveness by creator, past advertising performance, estimated advertising price range, and product category suitability together.

[0848] Additionally, the device (200) may provide a group of creator candidates sorted in order of priority or provide recommended creator display information together.

[0849] Through this, the device (200) can generate a group of candidate creators suitable for the advertised product by considering the product characteristics, advertising performance characteristics, and logistics status of the advertised product together. Since the device (200) can select creators with a high potential for actual advertising performance while also reflecting the inventory management status and advertising operation purpose, it can improve the accuracy of subsequent creator matching and advertising performance prediction.

[0850] In step S502, the device (200) can calculate the executable budget range and the expected exposure range, respectively, according to preset criteria using the ad exposure level and ad budget and creator performance data specified by the ad automation control value, and generate an ad execution range per creator that includes these.

[0851] In the present invention, the scope of advertising execution can be defined as advertising operation range data including an applicable advertising budget range and an expected advertising exposure range when executing advertising using a specific creator.

[0852] In this context, the executable budget range may refer to the range of the actual advertising budget available for a specific creator, and the estimated exposure range may refer to the expected scale of content exposure or user reach at a specific advertising budget level.

[0853] That is, the device (200) can generate an actual range for ad execution by considering the content response characteristics and past ad performance of each creator, rather than simply applying the ad exposure level and ad budget determined by the ad automation control value.

[0854] For example, the device (200) can collect ad exposure level and ad budget information included in the ad automation control value.

[0855] For example, if conditions for increased ad exposure and increased ad budget are applied to a specific ad target product, the device (200) can calculate the ad execution range based on the corresponding ad budget level and ad exposure level.

[0856] In addition, the device (200) can analyze the characteristics of advertising operations for each creator using creator performance data.

[0857] For example, you can analyze data on a specific creator's average content views, average content engagement, past ad click-through rates, and ad conversion rates.

[0858] In addition, it is possible to analyze the average advertising expenditure and actual ad impression performance during a specific creator's past advertising collaborations.

[0859] In addition, the device (200) can calculate the range of an executable budget.

[0860] For example, if a specific creator has shown stable advertising performance in an advertising budget range of 1,000,000 won to 3,000,000 won on average when executing advertisements in the past, the device (200) can calculate that range as an executionable budget range.

[0861] Conversely, if a specific creator exhibits characteristics where the advertising exposure effect is limited at a low advertising budget level and the advertising performance increases when an advertising budget above a certain level is invested, the device (200) can also calculate a minimum advertising execution budget standard.

[0862] Additionally, the device (200) can calculate the expected exposure range. For example, the expected exposure range by advertising budget level can be calculated using the average content views of a specific creator, the content diffusion rate, and exposure characteristics by social media advertising platform.

[0863] For example, as the advertising budget level increases, the estimated number of content impressions, estimated ad clicks, and estimated user reach can be calculated in stages.

[0864] In addition, the device (200) can generate an advertising execution range by considering the content exposure characteristics of each SNS advertising platform.

[0865] For example, in the case of social media advertising platforms centered on short video content, the expected exposure range can be calculated by reflecting the content diffusion speed and the initial exposure growth rate. Conversely, in the case of social media advertising platforms centered on image-based content, the expected exposure range can be calculated by reflecting the content retention rate, the repeat exposure rate, or the characteristics of long-term exposure retention.

[0866] In addition, the device (200) can generate an advertising execution range by considering the product category characteristics of the product to be advertised.

[0867] For example, for product categories with strong seasonality, the expected exposure range can be adjusted upward to reflect the possibility that the effect of increased ad exposure will be concentrated during a specific period. Conversely, for everyday consumer goods with high repeat purchase rates, the feasible duration of ad execution and the expected exposure maintenance range can be calculated by considering the long-term effects of ad exposure.

[0868] In addition, the device (200) can generate an advertising execution range by taking logistics data into account.

[0869] For example, if the stock quantity of a specific product targeted for advertising is insufficient or the expected time of stock depletion is imminent, the device (200) may limit the upper limit of the executable budget range. Conversely, if the risk of excess stock is high, the upper limit of the executable budget range and the expected exposure range may be expanded to induce increased advertising exposure.

[0870] In addition, the device (200) can generate an advertising execution range by considering the stability of advertising operations.

[0871] For example, since there is a possibility of reduced advertising efficiency when the level of the advertising budget increases rapidly in a short period of time, the device (200) may limit the allowable range of increase in the advertising budget.

[0872] For example, if the advertising budget growth rate exceeds a certain level, the device (200) can generate an advertising execution range by gradually decreasing the growth rate of the expected exposure growth rate or the expected ad click growth rate.

[0873] At this time, the criteria for calculating the executable budget range may include, but are not limited to, criteria for past advertising execution amounts, criteria for average advertising unit prices per advertising channel, criteria for advertising performance per creator, or criteria for advertising operation policies.

[0874] In addition, the criteria for calculating the expected exposure range may include, but are not limited to, criteria for content views, criteria for content diffusion ratios, criteria for ad click-through rates, criteria for average exposure characteristics by social media advertising platform, or criteria for ad operation by product category.

[0875] Additionally, the device (200) can correct abnormal response data during the process of generating the advertising execution range.

[0876] For example, if the number of views of a specific creator's content is very high but the actual ad conversion rate is significantly low, the device (200) can reduce the reflection rate of the corresponding view count data in the process of calculating the expected exposure range.

[0877] Additionally, if an abnormal increase in ad clicks pattern or a repetitive false response pattern is detected, the device (200) may limit the result of calculating the ad execution range of the creator or exclude it from the candidate group.

[0878] Additionally, the device (200) can provide the scope of advertising execution for each generated creator through the manager's terminal (100).

[0879] For example, the device (200) can display the execution budget range per creator, the estimated exposure range, the estimated ad click range, and the estimated user reach range together.

[0880] Additionally, the device (200) can display the range of possible advertising execution in the form of a graph, a comparison table, or a step-by-step advertising budget range.

[0881] Through this, the device (200) can generate an actual range for ad execution based on ad automation control values ​​and creator performance data. Since the device (200) can generate an ad execution range by considering the purpose of ad operation, creator characteristics, and inventory operation status together, it can subsequently improve the accuracy of calculating matching scores and predicting ad performance for each combination of creator and SNS ad platform.

[0882] In step S503, the device (200) can calculate a matching score for each combination of creators and social media advertising platforms using a creator candidate group, an advertising execution range, content exposure characteristics and content response characteristics for each social media advertising platform.

[0883] In the present invention, the matching score can be defined as an evaluation value indicating the degree to which a combination of a specific creator and a specific social media advertising platform is suitable for an advertised product.

[0884] In this case, the matching score can be calculated by including the suitability of content response characteristics, ad impression characteristics, product category, ad budget, ad performance, or inventory management.

[0885] In addition, content exposure characteristics may refer to the way content is exposed to users on social media advertising platforms or the characteristics of exposure diffusion.

[0886] For example, content exposure characteristics may include the initial exposure diffusion rate, recommendation algorithm-based exposure rate, repeat exposure rate, content retention exposure period, or time zones for concentrated ad exposure.

[0887] In addition, content response characteristics can refer to the tendency or response patterns of users to react to content on a specific social media advertising platform.

[0888] For example, content response characteristics may include like response rate, comment engagement rate, content save rate, share rate, link click-through rate, or purchase conversion rate.

[0889] That is, the device (200) can calculate a matching score for each combination of creator and SNS advertising platform by considering the characteristics of the product to be advertised, the purpose of the advertisement operation, the characteristics of the SNS advertising platform, and the scope of the advertisement execution, rather than simply calculating a matching score based only on the popularity level of the creator itself.

[0890] For example, the device (200) can analyze the content response characteristics of each creator included in the creator candidate group.

[0891] For example, if a specific creator consistently shows a high increase in views and a high sharing rate in short video content, the device (200) may evaluate the creator's content diffusion characteristics highly. Conversely, if the content views are high but the actual ad click-through rate or purchase conversion rate is low, the ad performance suitability may be evaluated low.

[0892] In addition, the device (200) can analyze the content exposure characteristics of each SNS advertising platform.

[0893] For example, if a specific social media advertising platform is a platform with a high initial diffusion effect based on a recommendation algorithm, the device (200) can be evaluated as a platform suitable for advertising operations aimed at expanding advertising exposure. Conversely, if a specific social media advertising platform is a platform with a high effect of maintaining repeated exposure, the device (200) can be evaluated as a platform suitable for advertising operations aimed at maintaining long-term brand awareness.

[0894] In addition, the device (200) can calculate a matching score by taking into account the scope of advertising execution.

[0895] For example, if the executionable budget range of a specific creator shows a high degree of suitability with the ad budget level included in the ad automation control value, the device (200) may evaluate the ad budget suitability as high. Conversely, if the estimated ad unit price of a specific creator exceeds the ad budget range, the ad budget suitability may be evaluated as low.

[0896] Additionally, the device (200) can calculate a matching score by analyzing the suitability between the product category of the advertised product and the characteristics of the creator content.

[0897] For example, if the product being advertised belongs to the beauty category and a specific creator exhibits high content responsiveness and high advertising conversion rates in beauty-related content, the device (200) can calculate a high matching score for the combination of the creator and the social media advertising platform. Conversely, if content that is not highly relevant to the product being advertised is mainly posted or past advertising performance is low, the matching score can be calculated low.

[0898] In addition, the device (200) can calculate a matching score by considering the purpose of advertising operation.

[0899] For example, if the primary goal is to increase short-term sales volume, a high weighting can be applied to purchase conversion rate and ad click-through rate. Conversely, if the primary goal is to expand brand awareness, a high weighting can be applied to content view growth rate, content sharing rate, and user reach.

[0900] In addition, the device (200) can calculate a matching score by considering logistics data together.

[0901] For example, if the risk of excess inventory of the advertised product is high, a high matching score can be calculated for combinations of creators and social media advertising platforms that have a high content diffusion effect within a short period. Conversely, if the risk of shortage inventory is high, the matching score of combinations likely to induce excessive purchase conversions can be limited.

[0902] In addition, the device (200) can calculate a matching score by taking into account user characteristics for each SNS advertising platform.

[0903] For example, if the age group of the main users of a specific social media advertising platform and the age group of the main buyers of the advertised product show a high degree of match, the device (200) can increase the matching score of the social media advertising platform. Conversely, if the characteristics of the main buyers of the advertised product and the characteristics of the social media advertising platform users are significantly different, the matching score can be decreased.

[0904] In addition, the device (200) can calculate a matching score by considering the stability of the advertisement operation.

[0905] For example, if the responsiveness of a specific creator's content fluctuates rapidly within a short period, the device (200) may limit the matching score by determining that the advertising performance stability is low.

[0906] In addition, if the volatility of content exposure on a specific social media advertising platform is very high, the matching score can be adjusted to reflect the potential deviation in expected advertising performance.

[0907] At this time, the criteria for calculating the matching score may include, but are not limited to, criteria for content views, criteria for ad click-through rates, criteria for purchase conversion rates, criteria for ad budget suitability, criteria for product category suitability, criteria for SNS ad platform user characteristics, or criteria for ad operation policies.

[0908] Additionally, the device (200) can correct abnormal reaction data during the process of calculating the matching score.

[0909] For example, if the number of views of a specific creator's content is very high but the actual purchase conversion rate is significantly low, the device (200) can lower the reflection rate of the view count data.

[0910] Additionally, if an abnormal increase in ad clicks pattern or a repetitive false response pattern is detected, the device (200) may limit the matching score of the corresponding creator and social media ad platform combination or exclude it from the candidate group.

[0911] Additionally, the device (200) can provide the calculated matching score through the manager's terminal (100).

[0912] For example, the device (200) can display matching scores by combination of creator and social media advertising platform, content response characteristics, advertising budget suitability, and expected advertising performance information together.

[0913] Additionally, the device (200) may sort and display combinations with high matching scores first, or provide recommended combination information.

[0914] Through this, the device (200) can calculate a matching score for each combination of creators and social media advertising platforms by reflecting the characteristics of the product to be advertised, the purpose of the advertisement operation, and the characteristics of the social media advertising platform. Since the device (200) can select combinations of creators and social media advertising platforms that have a high potential for actual advertising performance while also considering the advertising budget and inventory management status, it can improve the accuracy of the subsequent prediction of the expected advertising unit price and expected advertising performance.

[0915] In step S504, the device (200) can calculate the estimated advertising unit price for each combination of creators and social media advertising platforms using the matching score and the advertising budget included in the advertising automation control value.

[0916] In the present invention, the estimated advertising unit price can be defined as data representing the level of expected advertising execution costs when executing an advertisement using a combination of a specific creator and a specific SNS advertising platform.

[0917] In this case, the estimated advertising cost can be calculated by including content production costs, ad impression costs, platform ad operation costs, estimated click inducement costs, estimated conversion inducement costs, or creator collaboration costs.

[0918] In addition, the advertising budget included in the ad automation control value may refer to information on the amount of advertising execution available for actual ad operations for the advertised product.

[0919] That is, the device (200) can calculate the estimated advertising unit price for each combination of creator and SNS advertising platform by considering the matching score, advertising budget level, SNS advertising platform characteristics and advertising operation purpose, rather than simply calculating the estimated advertising unit price based only on the average advertising cost of the creator.

[0920] Additionally, the device (200) may calculate the estimated advertising unit price not by simply calculating the collaboration cost, but by reflecting the matching score for each combination of creator and SNS advertising platform, the level of the advertising budget included in the advertising automation control value, the content exposure characteristics for each SNS advertising platform, and the creator performance data. For example, in the case of a creator combination with a high matching score and excellent content responsiveness and past advertising performance, the efficiency of the estimated advertising unit price may be evaluated highly by reflecting the possibility of increased advertising exposure efficiency. Conversely, if the level of content responsiveness or advertising performance is low compared to the advertising budget level, the estimated advertising unit price may be adjusted upward by reflecting the possibility of decreased advertising operation efficiency.

[0921] For example, the device (200) can collect matching scores of a specific creator and social media advertising platform combination.

[0922] For example, if a specific creator shows high product category suitability and high advertising performance suitability with the advertised product and a specific social media advertising platform shows high content exposure diffusion effect, the device (200) can evaluate the matching score of the combination highly.

[0923] Additionally, the device (200) can collect advertising budget information included in the advertising automation control value.

[0924] For example, if an advertising budget increase condition is applied to a specific advertising target product, the device (200) can calculate an estimated advertising unit price based on the increased advertising budget level.

[0925] Additionally, the device (200) can analyze the creator's past advertising execution costs and advertising performance data.

[0926] For example, if a specific creator has spent an average of 1,500,000 won in advertising execution costs in past advertising collaborations and has shown an advertising click-through rate and purchase conversion rate above a certain level, the device (200) can reflect the data in the calculation of the expected advertising unit price.

[0927] In addition, the device (200) can calculate an estimated advertising unit price by considering the characteristics of advertising operation costs for each SNS advertising platform.

[0928] For example, if a specific social media advertising platform has a high advertising exposure diffusion effect but a high advertising execution cost, the device (200) can apply a high reflection ratio for the advertising operation costs of the platform. Conversely, if a specific social media advertising platform provides a high user reach with relatively low advertising costs, the estimated advertising cost can be calculated relatively low.

[0929] In addition, the device (200) can calculate an estimated advertising unit price by considering the purpose of advertising operation.

[0930] For example, if the primary goal is to increase short-term sales volume, a higher percentage of operating costs can be applied to purchase conversion-focused ads. Conversely, if the primary goal is to expand brand awareness, a higher percentage of costs for spreading content exposure can be applied.

[0931] Additionally, the device (200) can calculate an estimated advertising unit price by analyzing the suitability between the advertising budget level and the executionable budget range.

[0932] For example, if the executionable budget range of a specific creator shows a high degree of suitability with the ad budget level included in the ad automation control value, the device (200) may limit the increase in the expected ad unit price by determining that the ad operation efficiency is high. Conversely, if the ad budget level exceeds the average ad operation range of a specific creator, the expected ad unit price may be adjusted upward to reflect the possibility of an increase in additional ad operation costs.

[0933] In addition, the device (200) can calculate an estimated advertising unit price by taking into account the content response characteristics.

[0934] For example, if a specific creator's content exhibits a high sharing rate and a high content saving rate, the device (200) can evaluate the expected ad unit price efficiency as high compared to the same ad budget by reflecting the possibility of additional ad exposure effects. Conversely, if the content views are high but the ad click-through rate or purchase conversion rate is low, the ad operation efficiency is determined to be low, and the expected ad unit price can be adjusted upward.

[0935] In addition, the device (200) can calculate the estimated advertising unit price by taking logistics data into account.

[0936] For example, if the risk of excess inventory for the advertised product is high, the ratio reflecting the estimated ad unit price may be relaxed for combinations of creators and social media advertising platforms with a high potential for increased ad exposure to encourage inventory depletion. Conversely, if the risk of inventory shortage is high, the permissible range of increase in advertising execution costs may be limited by reflecting the possibility of inventory shortages resulting from expanded advertising operations.

[0937] In addition, the device (200) can calculate an estimated advertising unit price by considering the stability of advertising operations.

[0938] For example, if the fluctuation range of an ad unit price of a specific creator is very large, the device (200) can determine that the stability of the ad operation cost is low and calculate the expected ad unit price conservatively.

[0939] In addition, if the advertising operation costs of a specific social media advertising platform increase rapidly during a specific period, the device (200) can adjust the expected advertising unit price by reflecting the variability of the platform advertising costs.

[0940] At this time, the criteria for calculating the estimated advertising unit price may include, but are not limited to, past advertising execution costs, content responsiveness, ad click-through rates, purchase conversion rates, advertising budget suitability, advertising costs per social media advertising platform, or advertising operation policies.

[0941] Additionally, the device (200) can correct abnormal advertising data during the process of calculating the expected advertising unit price.

[0942] For example, if the number of views of a specific creator's content is very high but the actual purchase conversion rate is significantly low, the device (200) can lower the reflection rate of the content response data.

[0943] Additionally, if an abnormal increase in ad clicks pattern or a repetitive false response pattern is detected, the device (200) may limit or correct the expected ad unit price calculation result of the combination of the creator and the social media ad platform.

[0944] Additionally, the device (200) can provide the calculated estimated advertising unit price through the manager's terminal (100).

[0945] For example, the device (200) can display information on the estimated ad unit price, estimated ad exposure range, estimated ad clicks, and ad operation efficiency for each combination of creator and social media advertising platform.

[0946] Additionally, the device (200) can provide the estimated advertising unit price in the form of a ratio to the advertising budget, a step-by-step advertising cost range, or a comparison table.

[0947] Through this, the device (200) can calculate an estimated advertising unit price suitable for an actual advertising operation environment based on the matching score and the advertising budget included in the advertising automation control value. Since the device (200) can calculate the estimated advertising unit price by reflecting the characteristics of the product to be advertised, the purpose of advertising operation, and the characteristics of the SNS advertising platform together, it can improve the accuracy of calculating the estimated sales revenue, estimated return on investment, and cost-effectiveness ratio thereafter.

[0948] In step S505, the device (200) can calculate the expected sales revenue and expected profit rate for each combination of creator and SNS advertising platform using advertising efficiency evaluation data and logistics data.

[0949] In the present invention, the expected sales revenue can be defined as data representing the expected sales revenue of a product when an advertisement is executed using a combination of a specific creator and a specific social media advertising platform.

[0950] In addition, the expected rate of return can be defined as data representing the expected level of profit calculated by reflecting advertising execution costs or advertising operation costs relative to expected sales revenue.

[0951] In this case, the expected rate of return can be calculated by including the ratio of sales to advertising costs, the expected net profit ratio, the return on advertising investment ratio, or the expected margin rate.

[0952] That is, the device (200) can calculate the expected revenue and expected profit margin for each combination of creator and SNS advertising platform by considering not only the number of content views or ad impressions, but also ad efficiency evaluation data, logistics data, ad operation status and inventory operation status together.

[0953] For example, the device (200) can collect the efficiency level per advertisement and the suitability of the advertisement per product included in the advertising efficiency evaluation data.

[0954] For example, if a specific advertising target product exhibits a high ad click-through rate, a high purchase conversion rate, and a high sales level relative to advertising costs, the device (200) can evaluate the potential for advertising performance of the product highly.

[0955] In addition, if the advertising suitability for each product is high, it can be analyzed that there is a high possibility of generating additional revenue when advertising operations are expanded.

[0956] In addition, the device (200) can analyze the expected level of shipment and inventory operation status using logistics data.

[0957] For example, if the current inventory quantity of a specific advertising target product is sufficient and the risk of excess inventory is high, the device (200) can assess the possibility of additional sales generated by expanding advertising operations as high. Conversely, if the current inventory quantity is insufficient or the expected time of inventory depletion is very imminent, the possibility of an increase in expected sales can be calculated as limited.

[0958] In addition, the device (200) can calculate the expected sales revenue by utilizing the content response characteristics for each combination of creators and social media advertising platforms.

[0959] For example, if a specific creator consistently exhibits high content click-through rates and high purchase conversion rates, and a specific social media advertising platform is analyzed as a purchase conversion-centric platform, the device (200) can calculate a high expected revenue for that combination. Conversely, if the content view count is high but the actual ad click-through rate or purchase conversion rate is low, the level of expected revenue increase can be calculated to be limited.

[0960] In addition, the device (200) can calculate the expected sales amount by taking into account the ad exposure level and the ad budget level included in the ad automation control value.

[0961] For example, when conditions for increasing the level of ad exposure and increasing the ad budget are applied simultaneously, the device (200) can calculate the expected sales revenue by reflecting the expected increase in ad clicks, the expected increase in purchase conversions, and the expected increase in shipment volume together. Conversely, if the level of the ad budget is limited or the range of increasing the level of ad exposure is low, the expected increase in sales revenue can be calculated relatively low.

[0962] In addition, the device (200) can calculate the expected sales revenue and expected profit margin by considering the characteristics of each product category.

[0963] For example, for product categories with strong seasonality, the projected sales growth level can be adjusted upward to reflect the possibility that the effect of increased advertising exposure will be concentrated during a specific period. Conversely, for household consumer goods with high repeat purchase rates, the projected rate of return can be adjusted upward to reflect the potential for long-term repurchase.

[0964] In addition, the device (200) can calculate expected sales revenue and expected profit margin by considering user characteristics for each SNS advertising platform.

[0965] For example, if the age group of the main users of a specific social media advertising platform and the age group of the main buyers of the advertised product show a high degree of match, the device (200) can adjust the expected purchase conversion rate and expected sales revenue upward. Conversely, if the user characteristics and the characteristics of the advertised product are significantly different, the expected sales revenue and expected rate of return can be adjusted downward.

[0966] In addition, the device (200) can calculate the expected rate of return by taking into account the expected advertising unit price.

[0967] For example, even if projected revenue is high, the projected rate of return may be calculated with limitations if the projected advertising unit price is excessively high. Conversely, if the projected revenue growth level is high relative to the projected advertising unit price, the projected rate of return can be calculated as high.

[0968] In addition, the device (200) can calculate expected sales revenue and expected profit margin by considering the stability of advertising operations.

[0969] For example, if the fluctuation range of a specific creator's advertising performance is very large, the device (200) can determine that the stability of advertising performance is low and calculate the expected increase in sales revenue conservatively.

[0970] In addition, if the volatility of content exposure on a specific social media advertising platform is very high, the expected rate of return can be adjusted to reflect the possibility of deviation in expected advertising performance.

[0971] At this time, the criteria for calculating estimated sales revenue may include, but are not limited to, criteria for ad click-through rates, purchase conversion rates, ad exposure levels, estimated shipment volumes, sales characteristics by product category, or user characteristics by social media advertising platforms.

[0972] In addition, the criteria for calculating the expected rate of return may include, but are not limited to, criteria for expected advertising unit prices, criteria for sales relative to advertising costs, criteria for expected margin rates, criteria for return relative to advertising investment, or criteria for advertising operation policies, and may be set differently depending on the embodiment.

[0973] Additionally, the device (200) can correct abnormal advertising data during the process of calculating expected sales revenue and expected profit margin.

[0974] For example, if the number of views of a specific creator's content is very high but the actual purchase conversion rate is significantly low, the device (200) can lower the reflection rate of the content view count data.

[0975] Additionally, if an abnormal increase in ad clicks pattern or a repetitive false response pattern is detected, the device (200) may limit or correct the expected revenue and expected rate of return calculation results of the combination of the creator and the social media advertising platform.

[0976] Additionally, the device (200) can provide the calculated expected sales revenue and expected rate of return through the manager's terminal (100).

[0977] For example, the device (200) can display information on estimated sales revenue, estimated profit rate, estimated ad clicks, and estimated purchase conversion rate for each combination of creator and social media advertising platform.

[0978] Additionally, the device (200) can provide the expected sales revenue and expected rate of return in the form of a graph, a comparison table, or a comparison of performance against an advertising budget.

[0979] Through this, the device (200) can calculate an estimated sales revenue and an estimated profit margin suitable for the actual advertising operation environment based on advertising efficiency evaluation data and logistics data. Since the device (200) can predict advertising performance by reflecting the characteristics of the product to be advertised, the inventory operation status, and the characteristics of the SNS advertising platform together, it can improve the accuracy of calculating the cost-to-performance index and generating creator advertising execution data thereafter.

[0980] In step S506, the device (200) can calculate a cost-to-performance index for each combination of creators and social media advertising platforms using the estimated advertising unit price and the estimated sales revenue.

[0981] In the present invention, the cost-performance index can be defined as an evaluation value representing the expected level of advertising performance relative to the advertising costs invested for a combination of a specific creator and a specific social media advertising platform.

[0982] In this case, the cost-effectiveness index can be calculated by including the ratio of expected sales to expected ad unit price, the ratio of sales to advertising costs, an index reflecting expected return on investment, the ratio of advertising costs to expected ad clicks, or the ratio of advertising costs to expected purchase conversions.

[0983] That is, the device (200) can calculate a cost-to-performance index for each combination of creator and SNS advertising platform by reflecting the actual expected advertising performance level relative to the advertising operation cost, rather than independently comparing only the expected advertising unit price or expected sales revenue.

[0984] For example, the device (200) can collect the estimated advertising unit price of a specific creator and social media advertising platform combination.

[0985] For example, when an advertisement is operated by a combination of a specific creator and a specific social media advertising platform, if the estimated advertising unit price is calculated to be around 2,000,000 won, the device (200) can reflect the advertising cost information in the calculation of the cost-to-performance index.

[0986] Additionally, the device (200) can collect the expected sales of the same combination.

[0987] For example, if the expected increase in advertising revenue by the combination is calculated to be at the level of 10,000,000 won, the device (200) can calculate a cost-performance index using the ratio of advertising costs to expected revenue.

[0988] In addition, the device (200) can calculate a cost-to-performance index by taking into account the expected rate of return.

[0989] For example, if projected revenue is high but the projected advertising unit price is also very high, the projected rate of return can be reflected low, allowing for a limited calculation of the cost-effectiveness index. Conversely, if the projected revenue growth level is high relative to the projected advertising unit price, the cost-effectiveness index can be calculated highly.

[0990] In addition, the device (200) can calculate a cost-effectiveness index by considering advertising efficiency evaluation data together.

[0991] For example, if a combination of a specific creator and an SNS advertising platform consistently shows high ad click-through rates and high purchase conversion rates, the device (200) may determine that the stability of ad performance is high and adjust the cost-performance index upward. Conversely, if the number of content views is high but the actual ad click-through rate or purchase conversion rate is low, the cost-performance index may be adjusted downward.

[0992] In addition, the device (200) can calculate a cost-to-performance index by considering logistics data together.

[0993] For example, if the risk of excess inventory for the advertised product is high, the contribution to advertising performance can be evaluated highly by reflecting the effect of inducing inventory depletion. Conversely, if the current inventory is insufficient or the expected time of depletion is very imminent, the cost-effectiveness index can be calculated restrictively by reflecting the possibility of inventory shortages resulting from expanded advertising operations.

[0994] In addition, the device (200) can calculate a cost-effectiveness index by considering the content exposure characteristics and content response characteristics of each SNS advertising platform.

[0995] For example, if a specific social media advertising platform provides a high content diffusion effect and a specific creator exhibits a high sharing rate and a high content saving rate, the device (200) can adjust the cost-performance index upward to reflect the potential for long-term advertising effects. Conversely, in the case of a combination of platforms where content exposure is high but actual purchase conversion is low, the cost-performance index can be adjusted downward by determining that the advertising operation efficiency is low.

[0996] In addition, the device (200) can calculate a cost-effectiveness index by considering the purpose of advertising operations.

[0997] For example, if the primary goal is to increase short-term sales volume, a higher weighting can be applied to the projected purchase conversion rate and projected revenue. Conversely, if the primary goal is to expand brand awareness, a higher weighting can be applied to the projected user reach, content sharing rate, and content diffusion effect.

[0998] In addition, the device (200) can calculate a cost-to-performance index by considering both the advertising budget level and the stability of advertising performance.

[0999] For example, if the fluctuation range of the advertising unit price of a specific creator is very large or the variability of the advertising operation costs of a specific social media advertising platform is high, the device (200) may determine that the stability of the advertising performance prediction is low and calculate the cost-to-performance index conservatively. Conversely, if the advertising performance deviation is low and stable advertising efficiency is maintained in the past advertising operation results, the cost-to-performance index may be adjusted upward.

[1000] At this time, the criteria for calculating the cost-effectiveness ratio may include, but are not limited to, criteria for estimated advertising unit price, criteria for estimated sales revenue, criteria for sales relative to advertising costs, criteria for estimated rate of return, criteria for ad click-through rate, criteria for purchase conversion rate, or criteria for ad operation policy.

[1001] Additionally, the device (200) can correct abnormal advertising data during the process of calculating the cost-performance index.

[1002] For example, if the number of views of a specific creator's content is very high but the actual purchase conversion rate is significantly low, the device (200) can lower the reflection rate of the content view count data.

[1003] Additionally, if an abnormal increase in ad clicks pattern or a repetitive false response pattern is detected, the device (200) may limit or correct the cost-performance index of the combination of the creator and the social media ad platform.

[1004] Additionally, the device (200) can provide the calculated cost-to-performance index through the manager's terminal (100).

[1005] For example, the device (200) can display information on cost-to-performance ratios, estimated advertising unit prices, estimated sales revenue, and estimated rate of return for each combination of creator and social media advertising platform.

[1006] Additionally, the device (200) can provide a cost-performance index in the form of a ranking, a comparison graph, or an efficiency comparison relative to an advertising budget.

[1007] Through this, the device (200) can quantitatively evaluate the potential for actual advertising performance relative to advertising operation costs based on the expected advertising unit price and expected sales revenue. Since the device (200) can calculate a performance index relative to costs by reflecting advertising efficiency evaluation data, logistics data, and SNS advertising platform characteristics together, it can improve the accuracy of selecting a combination of creators and SNS advertising platforms and the efficiency of advertising operations.

[1008] In step S507, the device (200) can provide an interface that displays the matching score, estimated advertising unit price, estimated revenue, and cost-to-performance index for each combination of creator and SNS advertising platform through the manager's terminal.

[1009] In the present invention, the interface that displays for comparison can be defined as a user interface that provides advertising performance prediction results for a combination of multiple creators and social media advertising platforms in a form that allows for comparison with one another.

[1010] At this time, the interface may include a table form, a card form, a graph form, a ranking form, a comparison list form, or a multiple combination simultaneous comparison form, but is not limited thereto and may be configured differently depending on the embodiment.

[1011] That is, the device (200) does not simply display advertising performance data for each combination individually, but can provide an interface that allows an administrator to intuitively compare differences in advertising performance, differences in advertising costs, and suitability of advertising operations between multiple creator and SNS advertising platform combinations.

[1012] For example, the device (200) can display matching scores for each combination of creator and social media advertising platform.

[1013] For example, if the matching score of a specific combination is higher than that of other combinations, the device (200) may prioritize displaying that combination in the top area or provide it in a highlighted form.

[1014] In addition, the device (200) can display the estimated advertising unit price.

[1015] For example, if the estimated ad unit price of a specific creator and social media advertising platform combination falls within the ad budget range, the device (200) may display the information along with the ad operation availability status information. Conversely, if the estimated ad unit price exceeds the ad budget range included in the ad automation control value, it may display the budget overrun warning information or the ad execution restriction information together.

[1016] In addition, the device (200) can display information on expected sales revenue and expected rate of return together.

[1017] For example, if the expected sales growth level and expected rate of return of a specific combination are higher than those of other combinations, the device (200) can display the potential for advertising performance of that combination in the form of highlighting information. Conversely, if the expected advertising unit price is high but the expected rate of return is low, the device can display the potential for reduced advertising operation efficiency in the form of warning information.

[1018] In addition, the device (200) can also display a cost-performance index.

[1019] For example, if the cost-effectiveness ratio of a specific creator and social media advertising platform combination is high, the device (200) can display the combination as a combination with excellent advertising operation efficiency.

[1020] In addition, for combinations with a low cost-performance index, the advertising performance efficiency relative to advertising costs can be displayed as low.

[1021] Additionally, the device (200) can provide a visualization interface for comparison between multiple combinations.

[1022] For example, the device (200) can display the matching score, estimated advertising unit price, estimated sales revenue, and cost-performance index in the form of a graph that can be compared on the same screen.

[1023] Additionally, the device (200) may provide the ratio of expected sales to the advertising unit price, the ratio of expected return to the advertising budget, or the ranking of advertising performance efficiency in the form of a comparison table.

[1024] Additionally, the device (200) can set different priority display criteria for comparison items depending on the purpose of advertising operation.

[1025] For example, if the primary goal is to increase short-term sales volume, information related to projected revenue and purchase conversion rates can be displayed first. Conversely, if the primary goal is to expand brand awareness, information related to the diffusion effect of content exposure, projected user reach, and content sharing rates can be displayed first.

[1026] In addition, the device (200) can adjust the interface display method by taking into account the characteristics of each product category.

[1027] For example, for product categories with strong seasonality, information on the expected inventory depletion effect before the end of the season can be displayed. Conversely, for household consumer goods with high repeat purchase rates, information on the expected repurchase effect can be displayed during long-term advertising campaigns.

[1028] In addition, the device (200) can provide an interface by taking logistics data into account.

[1029] For example, if there is a high risk of excess inventory for a specific advertised product, combinations of creators and social media advertising platforms with a high effectiveness in inducing inventory depletion can be prioritized. Conversely, if there is a high risk of inventory shortage, warning indicators can be provided for combinations where the likelihood of shortage increases due to the expansion of advertising operations.

[1030] In addition, the device (200) can provide an interface considering the stability of advertising operations.

[1031] For example, if the fluctuation range of a specific creator's advertising performance is very large or the variability of advertising costs of a specific social media advertising platform is high, the device (200) can display advertising performance stability risk information together.

[1032] In addition, for combinations with a high probability of deviation in advertising performance prediction, the expected advertising performance range or the expected rate of return deviation range can be displayed together.

[1033] At this time, the interface display criteria may include, but are not limited to, matching score criteria, estimated ad unit price criteria, estimated revenue criteria, cost-effectiveness index criteria, ad operation purpose criteria, or ad operation policy criteria.

[1034] Additionally, the device (200) can correct abnormal advertising data during the interface provision process.

[1035] For example, if the number of views of a specific creator's content is very high but the actual purchase conversion rate is significantly low, the device (200) may display advertising performance information of the combination in a limited way or provide warning information together.

[1036] Additionally, if an abnormal increase in ad clicks or a repetitive false response pattern is detected, the device (200) may exclude the corresponding creator and social media ad platform combination from the comparison interface or display it in a separate risk indicator area.

[1037] Additionally, the device (200) can provide recommended combination information along with a comparison interface.

[1038] For example, the device (200) can display the combination with the highest cost-to-performance index, the combination with the highest expected sales revenue, or the combination with the highest advertising operation stability in the form of recommended combinations.

[1039] Through this, the device (200) can provide advertising performance prediction results for a combination of multiple creators and social media advertising platforms in a comparable form. Since the device (200) can provide a comparison interface that reflects advertising efficiency evaluation data, logistics data, and advertising operation objectives together, it can support an administrator in selecting an appropriate advertising combination by considering both advertising operation efficiency and inventory operation stability.

[1040] In step S508, the device (200) may provide a selection interface element for selecting a combination of creators and social media advertising platforms on the interface.

[1041] In the present invention, a selection interface element may be defined as a user interface component for selecting a specific combination among a plurality of creator and SNS advertising platform combinations or designating it as an advertising operation target.

[1042] At this time, the selection interface elements may include a selection button, a checkbox, a priority application selection menu, a combination comparison selection area, an ad execution approval button, or a step-by-step selection interface, but are not limited thereto and may be configured differently depending on the embodiment.

[1043] That is, the device (200) is not limited to simply displaying the results of advertising performance prediction, but can also provide selection interface elements so that an administrator can select a combination of creators and social media advertising platforms to apply to actual advertising operations.

[1044] For example, the device (200) can display selection interface elements in the form of selection buttons for each creator and social media advertising platform combination.

[1045] For example, the administrator can configure it so that specific combinations can be designated as ad operation targets via a selection button.

[1046] Additionally, the device (200) may provide a selection interface element in the form of a checkbox for comparing multiple combinations simultaneously.

[1047] For example, after an administrator selects a combination of multiple creators and social media advertising platforms, the system can be configured to allow comparison of estimated ad rates, estimated revenue, and cost-effectiveness indices.

[1048] Additionally, the device (200) may provide a selection interface element in the form of a selection menu for specifying a priority application combination.

[1049] For example, ...

Claims

Claim 1 A method for an artificial intelligence-based advertising evaluation and an advertising automation method using the same, performed by a device, comprising: collecting advertising data and logistics data stored in an ERP system; extracting at least one of the number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement from the advertising data, and extracting at least one of the shipment volume, inventory quantity, and expected time of inventory depletion per product from the logistics data; calculating an advertising performance indicator based on at least one of the extracted number of impressions, clicks, advertising costs, conversions, and sales revenue per advertisement; inputting at least one of the data extracted from the advertising performance indicator and the logistics data into an artificial intelligence model trained based on past advertising data and logistics data stored in the ERP system to generate advertising efficiency evaluation data including an efficiency level per advertisement and an advertising suitability degree per product; generating advertising planning data including advertising target products, advertising execution priorities, and advertising budget allocation information based on the advertising efficiency evaluation data; and generating an advertising automation control value including at least one of an advertising exposure level adjustment value, an advertising budget adjustment value, and an advertising target product change value based on the advertising planning data. Artificial intelligence-based ad evaluation and ad automation method using the same, comprising the step of changing at least one of an ad exposure level, an ad budget, and an ad target product according to the ad automation control value. Claim 2 In claim 1, the step of generating the advertising automation control value comprises: generating inventory status data including the current inventory quantity, shipment history, available sales period, and expected time of inventory depletion for each product from the logistics data; classifying the shipment history by time unit to generate product-specific shipment time series data; analyzing the shipment time series data to calculate demand pattern data including demand increase trend, demand decrease trend, and demand volatility for each product; aggregating the product-specific inventory status data and the demand pattern data by category to generate category-specific inventory level and category-specific demand pattern data; calculating product-specific safety stock standards and product-specific overstock standards based on the inventory quantity, the expected time of inventory depletion, and demand volatility; using the inventory status data, the demand pattern data, the category-specific inventory level, and the category-specific demand pattern data, and using the safety stock standards and overstock standards as comparison criteria to calculate inventory shortage risk and inventory overstock risk; generating an advertising exposure increase condition that induces priority shipment for products whose remaining available sales period is below a preset standard; and when the inventory shortage risk is above a preset standard, the advertising efficiency A step of generating a first ad control condition including a reduction in ad exposure and an increase in ad budget when the efficiency level per ad included in the evaluation data is above a preset standard; a step of generating a second ad control condition including an increase in ad exposure and an increase in ad budget when the excess inventory risk is above a preset standard and the ad suitability level per product included in the ad efficiency evaluation data is above a preset standard; and a condition selected according to a preset priority standard among the first ad control condition, the second ad control condition, and the ad exposure increase condition based on the remaining period of the sales period, to obtain an ad exposure level adjustment value,AI-based ad evaluation and ad automation method using the same, comprising the step of generating ad automation control values ​​including ad budget adjustment values ​​and ad target product change values. Claim 3 In claim 1, the step of generating advertising plan data based on the advertising efficiency evaluation data comprises: a step of calculating an advertising priority value for each target product using the advertising efficiency level and product-specific advertising suitability included in the advertising efficiency evaluation data and the logistics data; a step of displaying together the advertising priority value for each target product and adjustment interface elements for adjusting the advertising exposure level and advertising budget level corresponding to each product through the manager's terminal; a step of generating manager setting values ​​for the advertising exposure level and advertising budget level according to the manager's input to the adjustment interface elements; a step of calculating changes in advertising performance and inventory quantity corresponding to the manager setting values ​​using the advertising efficiency evaluation data and the logistics data; a step of providing the changes in advertising performance and inventory quantity through the manager's terminal so as to correspond to the manager setting values ​​for the advertising exposure level and advertising budget level; a step of setting constraints so that the advertising exposure level and advertising budget are included within a preset allowable range using the inventory quantity and the expected time of inventory depletion included in the logistics data; a step of generating a recommended control value by automatically adjusting the manager setting value according to the manager's input to satisfy the constraints; and using the manager input history and manager setting information collected from the manager's terminal to generate the recommended control value Artificial intelligence-based ad evaluation and ad automation method using the same, comprising the steps of: a correction step; and selecting one of the recommended control value and the administrator setting value according to an administrator selection input or a preset selection criterion, and generating ad plan data by reflecting the selected value.

Images