Method and system for preventing marketing capital loss

Abstract

The invention relates to the technical field of platform marketing risk detection, and provides a method for preventing marketing asset loss, which comprises the following steps: S1, acquiring platform party preferential data through technologies including web crawler or platform interface docking; s2, combining the preferential data of the platform party with the commodity main file information in the business system obtained through butt joint, and calculating to obtain multi-dimensional hand price prediction data; s3, combining the multi-dimension-to-hand price prediction data with a multi-level and multi-dimension red line price disc set by the system, and obtaining real-time commodity red line price early warning through red line price detection; s4, in combination with historical price data retention, giving a suggestion or an early warning when the hand price is lower than the historical minimum price; s5, analyzing the same-period year-on-year ratio and ring-on-ring ratio of the similar commodities through a marketing risk monitoring large model AI, and giving corresponding suggestions or early warning; and S6, related commodity operators are touched through a multi-stage early warning system. According to the technical scheme, it can be ensured that merchants comply with pricing rules, and marketing asset loss caused by the price setting problem is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of platform marketing risk detection, and more particularly to a method and system for preventing marketing asset losses. Specifically, the system employs platform data crawlers, discount calculations, multi-dimensional discount redline price checks, and AI price checks to establish a multi-dimensional, multi-redline price monitoring system for platform promotional activities to prevent marketing asset losses. Background Art

[0002] With the booming e-commerce industry, merchants face numerous complex price management challenges during product operations. When creating or modifying product prices or configuring marketing campaigns, it's not uncommon for operational errors to result in prices far below cost, posing significant financial risks to merchants.

[0003] From the product creation stage, merchants need to enter a large amount of product information, including basic product names, specifications, inventory, etc., and price setting is undoubtedly the most critical part. Some merchants are prone to various mistakes due to insufficient understanding of the platform's price setting rules or carelessness during operation. For example, some merchants who are new to the platform are unfamiliar with the price unit setting and may mistakenly fill in the price in "cents" instead of "yuan", resulting in a significant reduction in product prices; some merchants also misunderstand the logic of the discount algorithm when setting up promotional activities, and mistakenly set the discount amount as the transaction price, allowing consumers to purchase products at extremely low prices, causing serious financial losses.

[0004] When it comes to price adjustments, the market environment is dynamic. Factors such as raw material price fluctuations, competitor price adjustments, and shifts in consumer demand all require merchants to adjust product prices promptly. However, many merchants, lacking effective price monitoring mechanisms and professional market analysis capabilities, fail to keep pace with market changes. For example, in certain seasonal commodity markets, when raw material shortages lead to significant cost increases, merchants who fail to promptly raise product prices while continuing to implement existing promotional activities may end up losing money during the sales process. Similarly, in the electronics market, with the introduction of new technologies and the acceleration of product upgrades, if merchants fail to promptly reduce the prices of older products to clear inventory and instead continue to invest marketing resources, this results in wasted resources and potential losses.

[0005] Configuring marketing campaigns is also prone to pricing errors. Today, platforms offer a wide variety of promotional tools, such as discounts, instant discounts, coupons, and discounts. While combining these tools can attract more consumers, it also introduces complex logic challenges. Different promotional tools have varying priorities, stacking rules, and applicable scopes. A single, careless combination can cause the actual transaction price of a product to exceed its acceptable price. For example, if a merchant sets up a discount promotion offering a discount of X yuan on purchases above a certain amount while also issuing a coupon with no threshold, and fails to implement proper system settings to prevent these two combinations, consumers could end up using the coupon before receiving the discount, resulting in a final transaction price far lower than expected, resulting in significant losses for the merchant. According to industry statistics, single marketing losses caused by logic errors in the combination of promotional tools can range from tens of thousands of yuan to millions or even tens of millions of yuan.

[0006] Faced with such a complex and challenging price management challenge, traditional manual price monitoring and management methods are no longer sufficient. Manual monitoring is not only inefficient and prone to oversights, but also unable to track massive amounts of market data and price changes in real time. Therefore, there is an urgent need for systems that utilize advanced technologies, such as crawlers and big data analytics, to capture platform-related activity and coupon data. Combined with product pricing information, these systems perform precise calculations based on complex transaction rules, allowing for early predictions of product prices. These predictions are then compared with merchants' pre-set pricing rules (product minimum prices), enabling timely identification and early warning of potential losses. This helps prevent marketing losses caused by price setting errors and ensures reasonable profits and sustainable growth for merchants. Summary of the Invention

[0007] In response to the above problems, the purpose of the present invention is to provide a method and system for preventing marketing capital losses, which obtains platform-related activities and coupon data through crawlers and other technical means, combines product pricing information with the calculation and implementation of relevant transaction rules and predicts product prices, ensures that merchants comply with pricing rules (product red line prices), and avoids marketing capital losses caused by price setting problems.

[0008] The above-mentioned object of the present invention is achieved through the following technical solutions: A method for preventing marketing asset losses, comprising the following steps: S1: Obtain platform discount data through technologies including web crawlers or platform interface docking; S2: The platform's preferential data is combined with the commodity master file information in the business system obtained through the connection to calculate multi-dimensional price prediction data; S3: The multi-dimensional price forecast data is combined with the multi-level and multi-dimensional red line price list set by the system to obtain a real-time red line price warning for the commodity through red line price detection; S4: Combined with the retained historical price data, provide suggestions or warnings when the price is lower than the historical lowest price; S5: Use AI to analyze the year-on-year and month-on-month performance of similar products through a large marketing risk monitoring model to provide corresponding suggestions or warnings; S6: Reach relevant product operators through a multi-level early warning system.

[0009] Furthermore, in step S1, the platform's preferential data is obtained through technologies including web crawlers or platform interface docking, specifically: Through web crawler technology, accurately locate the web page elements where platform data, including platform coupons, product activities, platform member discounts, and platform store discounts, use HTML tags, CSS selectors, or XPath expressions to lock the location of data nodes, use the network request library to send HTTP requests, reasonably set the request header to simulate real browser behavior, process the request response, and parse the obtained web page content using library files including the BeautifulSoup library and the json library. After cleaning the data, store it in the specified format. At the same time, adopt strategies including setting request intervals, using proxy IP pools, and simulating user behavior to deal with the platform's anti-crawler mechanism to obtain the platform's discount data; or Apply to the platform for interface use, obtain platform authentication credentials, call the interface in accordance with the functions, input parameter requirements, and output data format specifications specified in the interface document, receive and parse the platform discount data returned by the interface, perform data verification, pay attention to interface version updates and adjust the code in a timely manner, and obtain the platform's discount data through the platform interface.

[0010] Furthermore, in step S2, the platform's preferential data is combined with the commodity master file information in the business system obtained through the connection to calculate the multi-dimensional price prediction data, specifically: Connecting with the business system to obtain detailed information about the product master file, including product category and product price; The obtained platform discount data is combined with the main file information of the product, and according to the relevant industry and platform rules, the price after the inclusive coupon and the estimated lowest price are calculated based on the product price to obtain the multi-dimensional price prediction data; The price after the inclusive coupon is calculated based on the current public inclusive discount of the product, and is often used as a reference for the price when registering for investment promotion. The calculation does not take into account discounts that do not meet the threshold except for cross-store discounts, discounts for specific groups and targeted channels, consumer assets, and platform investment; The estimated minimum price is the lowest price after the maximum discount is placed within a preset time in the future, taking into account the discounts that can be combined for all groups of people across all channels. It is used to determine whether there is a risk of product loss. The calculation includes discounts that do not meet the threshold and are converted proportionally, as well as discounts for groups and targeted channels, but does not include consumer assets or platform contributions.

[0011] Furthermore, before step S3, the method further includes setting the multi-level and multi-dimensional red line price list, specifically: Multiple price lists are preset, and different price dimensions, including the price after the inclusive coupon or the estimated lowest price, are selected in each price list for red line price comparison and warning. At the same time, when setting the red line price in the price list, options are provided for setting a unified store discount or importing specific amounts by product; In addition, different warning reminder channels are set for each of the prices, including DingTalk group notifications and DingTalk phone notifications, as well as corresponding warning levels and penalty conditions.

[0012] Furthermore, in step S3, the multi-dimensional price forecast data is combined with the multi-level and multi-dimensional red line price list set by the system to obtain a real-time commodity red line price warning through red line price detection, specifically: The calculated multi-dimensional price prediction data including the post-price of the inclusive coupon or the estimated minimum price is compared with the red line price set by the user in the price disk. When the red line price set by the user in the price disk is higher than the set multi-dimensional price prediction data, different warning levels are triggered according to pre-set trigger conditions.

[0013] Furthermore, in step S4, combined with the retained historical price data, when the price is lower than the historical lowest price, suggestions or warnings are given, specifically: When the price or data of the commodity master file information and platform data in the business system changes, the original data will be automatically saved in the historical data retention module; The multi-dimensional price prediction data is compared with the historical lowest price in the historical lowest price setting. When the price is lower than the historical lowest price, suggestions are given or warnings are triggered. At the same time, warning data including the red line price and the price are retained when performing red line price detection.

[0014] Furthermore, in step S5, the marketing risk monitoring model AI analyzes the same period and month-on-month performance of similar products and provides corresponding suggestions or warnings, specifically: When the multi-dimensional price prediction data, including the price after the inclusive coupon and the estimated lowest price, passes the red line price detection process, the data is synchronized to the historical price data retention. The marketing risk monitoring large model AI obtains the historical data retention analysis and outputs the red line price set by the AI virtual price disk. The formula and logic used are as follows: Get the average price of the same period in the preset historical time , calculate the historical price standard deviation, which represents the degree of deviation of price data from the historical average , the calculation formula is: in, is the historical statistical average price, is the i-th price data, N is the total amount of data; By historical price standard deviation Calculating the coefficient of volatility , the calculation formula is: Among them, the coefficient of 0.8 is a risk adjustment parameter to avoid interference from extreme fluctuations; Calculate dynamic adjustment factors based on inventory pressure and market demand , the calculation formula is: The final discount amount is: .

[0015] Furthermore, in step S6, the multi-level early warning system is used to contact the relevant product operators, specifically: During the red line price detection process, when the preset warning trigger conditions are met, the relevant commodity operators are contacted through a multi-level warning system; the multi-level warning system includes setting different warning levels and configuring corresponding contact methods for each warning level to ensure that commodity operators can obtain relevant information on price anomalies in a timely manner.

[0016] A system for preventing marketing asset loss for executing the above-mentioned method for preventing marketing asset loss comprises: The preferential data acquisition module is used to obtain preferential data from the platform through technologies including web crawlers or platform interface docking; The price calculation module is used to combine the platform's preferential data with the commodity master file information obtained from the docking business system to calculate multi-dimensional price prediction data; The red line price warning module is used to combine the multi-dimensional price forecast data with the multi-level and multi-dimensional red line price disk set by the system to obtain real-time commodity red line price warnings through red line price detection; The historical data warning module is used to combine the retained historical price data and give suggestions or warnings when the price is lower than the historical lowest price; The AI model early warning module is used to analyze the same period and month-on-month performance of similar products through the marketing risk monitoring large model AI to provide corresponding suggestions or early warnings; The early warning reminder contact module is used to reach relevant product operators through a multi-level early warning system.

[0017] A computer-readable storage medium stores computer code. When the computer code is executed, the above method is performed.

[0018] Compared with the prior art, the present invention has at least one of the following beneficial effects: (1) Comprehensive and accurate data acquisition: The platform's preferential data is obtained through two technical means: web crawlers and platform interface docking. Web crawler technology can accurately locate web page elements and effectively obtain various preferential information such as platform coupons, product activities, etc. Even if the platform data display format is complex, it can be accurately captured through HTML tags, etc.; platform interface docking follows standardized procedures to ensure the officiality and stability of the acquired data. The combination of the two ensures the diversity and comprehensiveness of data sources, laying a solid foundation for the subsequent accurate calculation of the purchase price and risk warning. In the data acquisition process, whether it is the web crawler's parsing, cleaning, and storage of data, or the data verification during interface docking, professional and sophisticated processing methods are adopted to remove invalid and erroneous data, ensure the high quality and accuracy of the data, and make subsequent analysis and calculation more reliable.

[0019] (2) The calculation of the final price is scientific and reasonable: Based on industry and platform rules, combined with the platform's preferential data and the main information of the product, the price after the inclusive coupon and the estimated minimum final price are calculated respectively. The price after the inclusive coupon is in line with the investment registration scenario, and the estimated minimum final price takes into account all kinds of superimposed discounts, which is used to accurately determine the risk of capital loss. It provides merchants with a clear and practical price reference from different dimensions, so that merchants can more accurately grasp the actual selling price of the product. The rules for including and excluding various discounts in the calculation of the two final prices are clearly stipulated, avoiding ambiguity and arbitrariness in the calculation, so that merchants and the system can operate and judge according to unified and clear standards.

[0020] (3) Flexible and efficient red line price setting and early warning: It supports presetting multiple price plates, and each price plate can select different dimensions of prices for red line price comparison and early warning. The red line price setting provides a variety of methods such as unified store discounts or importing specific amounts by product, which meets the diverse needs of different merchants and different products in different marketing scenarios. Merchants can flexibly customize the price bottom line according to their own business strategies and product characteristics. Clear red line price detection rules, that is, when the red line price is higher than the multi-dimensional hand-in price forecast data, different warning levels are triggered according to the preset trigger conditions, which can timely and accurately capture the risk of the hand-in price being lower than the red line price, providing strong protection for merchants to timely discover and deal with potential asset loss risks. At the same time, the multi-level early warning system sets different warning levels and corresponding contact methods to ensure that product operators can obtain price anomaly information in a timely and accurate manner, facilitating rapid response and processing.

[0021] (4) Historical data and AI assist in risk prevention and control: The original data of business system data and platform data changes are automatically retained in the historical data retention module, which not only provides a basis for comparing the actual price with the historical lowest price, and enables timely warnings or suggestions when the actual price is lower than the historical lowest price, but also provides a rich data foundation for the continuous optimization and analysis of the system. The marketing risk monitoring model AI uses historical data retention for analysis and outputs the red line price set by the AI virtual price disk. Through a series of scientific formula calculations, such as considering the historical average price, standard deviation, volatility coefficient, dynamic adjustment factor, etc., it comprehensively and dynamically analyzes price trends and market conditions, providing merchants with more forward-looking and scientific price setting suggestions or risk warnings, improving the system's adaptability to market changes and the intelligent level of risk prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is an overall flow chart of the method for preventing marketing capital loss according to the present invention; Figure 2 A detailed flow chart of the method for preventing marketing capital loss according to the present invention; Figure 3 This is the overall structural diagram of the system for preventing marketing capital loss in the present invention. DETAILED DESCRIPTION

[0023] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a," "an," "said," and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0025] First embodiment like Figure 1 and 2 As shown, this embodiment provides a method for preventing marketing asset loss, comprising the following steps: S1: Obtain platform discount data through technologies including web crawlers or platform interface docking.

[0026] In this embodiment, step S1 is specifically as follows: Through web crawler technology, accurately locate the web page elements where platform data, including platform coupons, product activities, platform member discounts, and platform store discounts, use HTML tags, CSS selectors, or XPath expressions to lock the location of data nodes, use the network request library to send HTTP requests, reasonably set the request header to simulate real browser behavior, process the request response, and parse the obtained web page content using library files including the BeautifulSoup library and the json library. After cleaning the data, store it in the specified format. At the same time, adopt strategies including setting request intervals, using proxy IP pools, and simulating user behavior to deal with the platform's anti-crawler mechanism to obtain the platform's discount data; or Apply to the platform for interface use, obtain platform authentication credentials, call the interface in accordance with the functions, input parameter requirements, and output data format specifications specified in the interface document, receive and parse the platform discount data returned by the interface, perform data verification, pay attention to interface version updates and adjust the code in a timely manner, and obtain the platform's discount data through the platform interface.

[0027] S2: The platform's preferential data is combined with the commodity master file information in the business system obtained by docking to calculate multi-dimensional price prediction data.

[0028] In this embodiment, step S2 is specifically as follows: Connecting with the business system to obtain detailed information about the product master file, including product category and product price; The obtained platform discount data is combined with the main file information of the product, and according to the relevant industry and platform rules, the price after the inclusive coupon and the estimated lowest price are calculated based on the product price to obtain the multi-dimensional price prediction data; The price after the inclusive coupon is calculated based on the current public inclusive discount of the product, and is often used as a reference for the price when registering for investment promotion. The calculation does not take into account discounts that do not meet the threshold except for cross-store discounts, discounts for specific groups and targeted channels, consumer assets, and platform investment; The estimated minimum price is the lowest price after the maximum discount is placed within a preset time in the future, taking into account the discounts that can be combined for all groups of people across all channels. It is used to determine whether there is a risk of product loss. The calculation includes discounts that do not meet the threshold and are converted proportionally, as well as discounts for groups and targeted channels, but does not include consumer assets or platform contributions.

[0029] The following is a specific calculation example of the price after the inclusive coupon and the estimated lowest price: (1) Example of calculating the post-price of inclusive coupons Consider a sports backpack priced at 300 yuan. The platform currently offers the following public discounts: Cross-store discount: 30 off for every 200 spent; Store coupon (for all users): 20 off for purchases over 199.

[0030] Because cross-store discounts and store coupons available to all users are considered universal discounts, other discounts, such as coupons exclusive to specific members or coupons available only to users in certain channels, are not included in the calculation. Calculation process: First, consider the cross-store discount. A purchase of 300 yuan meets the 30 yuan discount requirement for purchases over 200 yuan, resulting in a 30 yuan discount. Then, use the store coupon for a 20 yuan discount on purchases over 199 yuan. The price after the universal discount coupon = 300 yuan - 30 yuan - 20 yuan = 250 yuan. This price can serve as a reference for merchants participating in platform investment promotions.

[0031] (2) Example of calculating the estimated minimum price Still taking this sports backpack as an example, in addition to the above general discounts, there are also the following discounts: The platform's targeted coupon for new users is 30 off for purchases over 150 (assuming it can be combined); Platform member exclusive coupon: 25 off for purchases over 200 (assuming it can be stacked); There are also some small discounts that do not meet the threshold, such as 5 off for purchases over 50. Although it does not meet the threshold, it is converted proportionally (assuming a 50% discount here).

[0032] Calculation process: First, calculate the general discount portion, using the same calculation as above for general discount coupons. A 30 yuan discount applies to cross-store purchases, and a 20 yuan discount applies to store coupons. New user coupons can be combined for an additional 30 yuan discount; member-only coupons can be combined for a 25 yuan discount. Discounts below the threshold are calculated proportionally: 5 yuan off for purchases above 50 yuan, resulting in a total discount of 5 x 50% = 2.5 yuan. Estimated minimum price = 300 yuan - 30 yuan - 20 yuan - 30 yuan - 25 yuan - 2.5 yuan = 192.5 yuan. This price represents the lowest price over a period of time, assuming a combination of discounts, and is used to determine whether the product is at risk of financial loss.

[0033] S3: The multi-dimensional price prediction data is combined with the multi-level and multi-dimensional red line price disk set by the system to obtain a real-time commodity red line price warning through red line price detection.

[0034] In this embodiment, before step S3, the multi-level and multi-dimensional red line price disk is also included, specifically: multiple price disks are preset, and different dimensional prices including the price after the inclusive coupon or the estimated lowest price are selected in each price disk for red line price comparison and warning. At the same time, when setting the red line price in the price disk, the option of providing a unified store discount or setting a specific amount by importing the product is provided; and different warning reminder channels including DingTalk group notification and DingTalk phone notification are set for each price, as well as corresponding warning levels and penalty conditions.

[0035] In this embodiment, step S3 is specifically as follows: The calculated multi-dimensional price prediction data, including the future price of the inclusive coupon or the estimated minimum price, is compared with the red line price set by the user in the price order. When the red line price set by the user in the price order is higher than the multi-dimensional price prediction data, different warning levels are triggered according to pre-set trigger conditions. For example, a level 1 warning is triggered when the red line price is greater than 5% of the red line price (red line price - price) or the red line price is greater than 5%.

[0036] For example, a sporting goods store sells a sports backpack. To effectively monitor price risks, the store sets two price lists: Price list 1: used for daily sales monitoring Compare price dimensions: Select the price after discount coupons for redline price comparison alerts. In daily sales, the price after discount coupons can better reflect the selling price of the product under regular discounts.

[0037] Redline Price Setting Method: Use a storewide discount method. The store sets the redline price for this sports backpack at 20% off the list price. If the sports backpack is listed at 300 yuan, then the redline price = 300 × 0.8 = 240 yuan.

[0038] Early Warning Alert Channels and Levels: Set up DingTalk group notifications as early warning alert channels. When the price after the coupon drops below the red line price of 240 yuan, set an early warning level based on the degree of deviation. If the price after the coupon drops between 220 and 240 yuan, a Level 2 alert is triggered, with a message posted in the DingTalk group reminding operations staff to monitor price fluctuations and check for unusual discounts. If the price after the coupon drops below 220 yuan, a Level 1 alert is triggered. In addition to the DingTalk group notification, the relevant person in charge will be notified via a DingTalk phone call, requiring them to immediately adjust the discount strategy to avoid losses. Furthermore, for Level 1 alerts, the employee responsible for setting the discount for the product will be warned and required to submit an optimization plan within one week.

[0039] Price list 2: used for monitoring during promotional activities Compare Price Dimension: Select the estimated lowest price to compare against the redline price for early warning. During promotional events, discounts can be complex and overlapping, so the estimated lowest price better reflects the lowest possible selling price for an item, helping to mitigate financial loss risks.

[0040] The redline price is set based on the specific amount of money imported into the product. Considering the cost and expected profit of the promotion, the store sets the redline price of this sports backpack at 200 yuan during the promotion.

[0041] Warning reminder channels and levels: Set the warning reminder channels to DingTalk phone notifications and email notifications. When the estimated lowest price is lower than the red line price of 200 yuan, if the estimated lowest price is between 180-200 yuan, a second-level warning is triggered, and the operations staff is notified via DingTalk phone that the price is lower than the red line price and that they need to pay attention to subsequent sales. If the estimated lowest price is lower than 180 yuan, a first-level warning is triggered. In addition to the phone notification, an email will be sent to notify the store management and relevant operations team, requesting the immediate suspension of some promotional activities involving the product and the urgent evaluation and adjustment of the promotion plan. At the same time, for the first-level warning situation, the entire operations team will be criticized and a portion of the team's performance bonus for that month will be deducted.

[0042] S4: Combined with the retained historical price data, provide suggestions or warnings when the price is lower than the historical lowest price.

[0043] In this embodiment, step S4 is specifically as follows: When the price or data of the commodity master file information and platform data in the business system changes, the original data is automatically retained in the historical data retention module; the multi-dimensional final price prediction data is compared with the historical lowest price in the historical lowest price setting. When the final price is lower than the historical lowest price, suggestions are given or an early warning is triggered. At the same time, the early warning data including the red line price and final price will be retained during the red line price detection.

[0044] S5: Through the AI ​​marketing risk monitoring model, the same period and month-on-month analysis of similar products are conducted to provide corresponding suggestions or warnings.

[0045] The above steps represent the traditional price risk detection and early warning process. Simultaneously, a large-scale AI-powered price detection mechanism will be implemented to address unpriced products and provide recommendations or price risk warnings for those with priced products.

[0046] In this embodiment, step S5 is specifically as follows: When the multi-dimensional price prediction data including the inclusive coupon price and the estimated lowest price pass the red line price detection process (note: the red line price here does not necessarily exist for every product, as long as the inclusive coupon or the lowest price is available for the product on the platform), the data is synchronized to the historical price data retention, and the marketing risk monitoring model AI obtains the historical data retention analysis and outputs the red line price set by the AI virtual price disk. The formula and logic used are as follows: Get the average price of the same period in the preset historical time (For example, take the weighted average of the price data for the 15 days before and after the same period in the past three years, with the weight decreasing as the time is closer. For example, the weight of the year before last is 0.5, and the weight of the previous two years is 0.3 and 0.2 respectively.) Calculate the historical price standard deviation, which represents the degree of deviation of the price data from the historical average. , the calculation formula is: in, is the historical statistical average price, is the i-th price data, N is the total amount of data; By historical price standard deviation Calculating the coefficient of volatility , the calculation formula is: Among them, the coefficient of 0.8 is a risk adjustment parameter to avoid interference from extreme fluctuations, and the risk adjustment coefficient is not fixed at 0.8 and can be adjusted according to needs.

[0047] Calculate dynamic adjustment factors based on inventory pressure and market demand , the calculation formula is: (Parameter weights need to be fine-tuned according to product categories) The final discount amount is: .

[0048] The following is a specific calculation example of a large-scale AI model for marketing risk monitoring: (1) Calculate the average price for the same period in history The prices in the past three years were 300 yuan, 320 yuan, and 310 yuan respectively. Assuming the weights are the same, then: Yuan (2) Calculate the standard deviation of historical prices 1. Calculate variance 2. Calculate the standard deviation Yuan (3) Determine the coefficient of fluctuation (4) Calculating reasonable price thresholds The upper limit of reasonable price is the fluctuation range of historical average: Yuan (5) Determine whether the current price needs a discount The current price of 330 yuan is greater than the threshold of 316.51, so the discount can be released (6) Calculate the optimal discount amount (the maximum amount that can be released) S6: Reach relevant product operators through a multi-level early warning system.

[0049] In this embodiment, step S6 is specifically as follows: During the red line price detection process, when the preset warning trigger conditions are met, the relevant commodity operators are contacted through a multi-level warning system; the multi-level warning system includes setting different warning levels and configuring corresponding contact methods for each warning level to ensure that commodity operators can obtain relevant information on price anomalies in a timely manner.

[0050] For example, if a sports backpack is priced at 300 yuan, the red line price after the inclusive coupon is set at 240 yuan in Price List 1, and the red line price of the estimated lowest price is set at 200 yuan in Price List 2. A three-level early warning system is established: (1) Level 1 warning Trigger conditions: Triggered when the price of the inclusive coupon is lower than 220 yuan (91.67% of the red line price of 240 yuan), or the estimated lowest price is lower than 180 yuan (90% of the red line price of 200 yuan).

[0051] Contact method: Immediately notify the product operations manager and store manager via DingTalk call, and send SMS reminders to ensure that key personnel are aware of it as soon as possible.

[0052] Example scenario: Due to an operator's mistake, a store coupon was set to offer 100 yuan off for purchases over 100 yuan, causing the coupon's value to drop to 200 yuan, triggering a Level 1 alert. Upon receiving the notification, the relevant person in charge can quickly review the discount settings and suspend the abnormal activity to avoid further losses.

[0053] (2) Second level warning Trigger conditions: Triggered when the price of the inclusive coupon is between 220-235 yuan (close to the red line price but not breaking the limit), or the estimated lowest price is between 180-195 yuan.

[0054] Contact method: Push warnings via DingTalk group messages and send emails to the product operations team, detailing price deviations and potential risks.

[0055] Example scenario: The platform launches a new discount promotion combined with store coupons, resulting in an estimated minimum purchase price of 190 yuan, triggering a Level 2 alert. After receiving the notification, the operations team can promptly evaluate the effectiveness of the promotion and adjust the discount strategy to control costs.

[0056] (3) Level 3 warning Trigger conditions: Triggered when the price of the inclusive coupon is between 235-240 yuan (slightly close to the red line price), or the estimated lowest price is between 195-200 yuan.

[0057] Contact method: Send an early warning message to all members in the DingTalk group to remind operators to pay attention to the price dynamics of the product. No emergency action is required.

[0058] Example scenario: As market competition intensifies, competitors offer price cuts and promotions. The post-coupon price of this sports backpack is 238 yuan, triggering a Level 3 alert. Operations personnel can conduct routine inspections to further analyze the market situation and determine whether pricing strategies need to be adjusted.

[0059] Second embodiment like Figure 3 As shown, this embodiment provides a system for preventing marketing asset loss for executing the method for preventing marketing asset loss in the first embodiment, including: Discount data acquisition module 1, used to obtain platform discount data through technologies including web crawlers or platform interface docking; The price calculation module 2 is used to combine the platform's preferential data with the commodity master file information in the business system obtained by docking to calculate the multi-dimensional price prediction data; Redline price warning module 3 is used to combine the multi-dimensional price forecast data with the multi-level and multi-dimensional redline price disk set by the system to obtain real-time commodity redline price warning through redline price detection; Historical data warning module 4 is used to combine the retained historical price data and give suggestions or warnings when the price is lower than the historical lowest price; AI model early warning module 5 is used to analyze the same period and month-on-month performance of similar products through the marketing risk monitoring large model AI to provide corresponding suggestions or early warnings; The early warning reminder contact module 6 is used to contact relevant product operators through a multi-level early warning system.

[0060] A computer-readable storage medium stores computer code. When the computer code is executed, the above-described method is performed. Those skilled in the art will appreciate that all or part of the steps in the various methods of the above-described embodiments can be performed by a program instructing related hardware. The program can be stored in a computer-readable storage medium. The storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0061] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

[0062] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention. It should be pointed out that those skilled in the art can make several improvements and modifications without departing from the principles of the present invention, and such improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for preventing marketing capital loss, characterized in that: The following steps are involved: S1: Obtain platform discount data through technologies including web crawlers or platform interface docking; S2: The platform's preferential data is combined with the commodity master file information in the business system obtained through the connection to calculate multi-dimensional price prediction data; S3: The multi-dimensional price forecast data is combined with the multi-level and multi-dimensional red line price list set by the system to obtain a real-time red line price warning for the commodity through red line price detection; S4: Combined with the retained historical price data, provide suggestions or warnings when the price is lower than the historical lowest price; S5: Use AI to analyze the year-on-year and month-on-month performance of similar products through a large marketing risk monitoring model to provide corresponding suggestions or warnings; S6: Reach relevant product operators through a multi-level early warning system.

2. The method for preventing marketing capital loss according to claim 1, characterized in that: In step S1, the platform's preferential data is obtained through technologies including web crawlers or platform interface docking, specifically: Through web crawler technology, accurately locate the web page elements where platform data, including platform coupons, product activities, platform member discounts, and platform store discounts, use HTML tags, CSS selectors, or XPath expressions to lock the location of data nodes, use the network request library to send HTTP requests, reasonably set the request header to simulate real browser behavior, process the request response, and parse the obtained web page content using library files including the BeautifulSoup library and the json library. After cleaning the data, store it in the specified format. At the same time, adopt strategies including setting request intervals, using proxy IP pools, and simulating user behavior to deal with the platform's anti-crawler mechanism to obtain the platform's discount data; or Apply to the platform for interface use, obtain platform authentication credentials, call the interface in accordance with the functions, input parameter requirements, and output data format specifications specified in the interface document, receive and parse the platform discount data returned by the interface, perform data verification, pay attention to interface version updates and adjust the code in a timely manner, and obtain the platform's discount data through the platform interface.

3. The method for preventing marketing capital loss according to claim 2, characterized in that: In step S2, the platform's preferential data is combined with the commodity master file information in the business system obtained by docking to calculate multi-dimensional price prediction data, specifically: Connecting with the business system to obtain detailed information about the product master file, including product category and product price; The obtained platform discount data is combined with the main file information of the product, and according to the relevant industry and platform rules, the price after the inclusive coupon and the estimated lowest price are calculated based on the product price to obtain the multi-dimensional price prediction data; The price after the inclusive coupon is calculated based on the current public inclusive discount of the product, and is often used as a reference for the price when registering for investment promotion. The calculation does not take into account discounts that do not meet the threshold except for cross-store discounts, discounts for specific groups and targeted channels, consumer assets, and platform investment; The estimated minimum price is the lowest price after the maximum discount is placed within a preset time in the future, taking into account the discounts that can be combined for all groups of people across all channels. It is used to determine whether there is a risk of product loss. The calculation includes discounts that do not meet the threshold and are converted proportionally, as well as discounts for groups and targeted channels, but does not include consumer assets or platform contributions.

4. The method for preventing marketing capital loss according to claim 3, characterized in that: Before step S3, the multi-level and multi-dimensional red line price list is set, specifically: Multiple price lists are preset, and different price dimensions, including the price after the inclusive coupon or the estimated lowest price, are selected in each price list for red line price comparison and warning. At the same time, when setting the red line price in the price list, options are provided for setting a unified store discount or importing specific amounts by product; In addition, different warning reminder channels are set for each of the prices, including DingTalk group notifications and DingTalk phone notifications, as well as corresponding warning levels and penalty conditions.

5. The method for preventing marketing capital loss according to claim 4, characterized in that: In step S3, the multi-dimensional price forecast data is combined with the multi-level and multi-dimensional red line price disk set by the system to obtain a real-time commodity red line price warning through red line price detection, specifically: The calculated multi-dimensional price prediction data including the post-price of the inclusive coupon or the estimated minimum price is compared with the red line price set by the user in the price disk. When the red line price set by the user in the price disk is higher than the set multi-dimensional price prediction data, different warning levels are triggered according to pre-set trigger conditions.

6. The method for preventing marketing capital loss according to claim 3, characterized in that: In step S4, combined with the retained historical price data, when the price is lower than the historical lowest price, suggestions or warnings are given, specifically: When the price or data of the commodity master file information and platform data in the business system changes, the original data will be automatically saved in the historical data retention module; The multi-dimensional price prediction data is compared with the historical lowest price in the historical lowest price setting. When the price is lower than the historical lowest price, suggestions are given or warnings are triggered. At the same time, warning data including the red line price and the price are retained when performing red line price detection.

7. The method for preventing marketing capital loss according to claim 3, characterized in that: In step S5, the AI ​​marketing risk monitoring model analyzes the same period and month-on-month performance of similar products and provides corresponding suggestions or warnings, specifically: When the multi-dimensional price prediction data, including the price after the inclusive coupon and the estimated lowest price, passes the red line price detection process, the data is synchronized to the historical price data retention. The marketing risk monitoring large model AI obtains the historical data retention analysis and outputs the red line price set by the AI virtual price disk. The formula and logic used are as follows: Get the average price of the same period in the preset historical time , calculate the historical price standard deviation, which represents the degree of deviation of price data from the historical average , the calculation formula is: in, is the historical statistical average price, is the i-th price data, N is the total amount of data; By historical price standard deviation Calculating the coefficient of volatility , the calculation formula is: Among them, the coefficient of 0.8 is a risk adjustment parameter to avoid interference from extreme fluctuations; Calculate dynamic adjustment factors based on inventory pressure and market demand , the calculation formula is: The final discount amount is: .

8. The method for preventing marketing capital loss according to claim 1, characterized in that: In step S6, the multi-level early warning system is used to contact the relevant product operators, specifically: During the red line price detection process, when the preset warning trigger conditions are met, the relevant commodity operators are contacted through a multi-level warning system; the multi-level warning system includes setting different warning levels and configuring corresponding contact methods for each warning level to ensure that commodity operators can obtain relevant information on price anomalies in a timely manner.

9. A system for preventing marketing asset loss for executing the method for preventing marketing asset loss according to any one of claims 1 to 8, characterized in that: include: The preferential data acquisition module is used to obtain preferential data from the platform through technologies including web crawlers or platform interface docking; The price calculation module is used to combine the platform's preferential data with the commodity master file information obtained from the docking business system to calculate multi-dimensional price prediction data; The red line price warning module is used to combine the multi-dimensional price forecast data with the multi-level and multi-dimensional red line price disk set by the system to obtain real-time commodity red line price warnings through red line price detection; The historical data warning module is used to combine the retained historical price data and give suggestions or warnings when the price is lower than the historical lowest price; The AI model early warning module is used to analyze the same period and month-on-month performance of similar products through the marketing risk monitoring large model AI to provide corresponding suggestions or early warnings; The early warning reminder contact module is used to reach relevant product operators through a multi-level early warning system. 10 . A computer-readable storage medium storing computer code, wherein when the computer code is executed, the method according to claim 1 is performed.

Images