Saas-based app promotion result attribution method and system

Abstract

The invention discloses a saas-based app promotion result attribution method and a saas-based app promotion result attribution system. The invention relates to the technical field of network promotion. And configuring a data source D required by a data import algorithm, wherein the data source D comprises a data format FD, a storage position LD and an access right PD. A data source configuration function Configuration (D, FD, LD, PD) receives a data source D, a data format F, a storage position LD and an access right PD as input, and returns a configured data source. Through comprehensive data source configuration and access and strict data import and processing processes, the technical scheme of the invention ensures the integrity and accuracy of data. According to the technical scheme, the tracking tool integration can finely capture key behaviors, such as downloading, starting and using, of the user in the process of using the multi-version App. The evaluation of the promotion effect is more accurate, and the performance difference of different versions of App can be accurately measured.

Description

Technical Field

[0001] The present invention relates to the field of network promotion technology, specifically to the field of promotion attribution technology, and in particular to a SaaS-based app promotion result attribution method and system. Background Art

[0002] Social media platforms have become an important channel for online promotion. Companies can use social media platforms to publish content, interact with users, and build brand image and reputation. Social media marketing emphasizes the fun and interactivity of content to attract and retain users.

[0003] Third-party promoters are organizations that provide online promotion services to enterprises. They can tailor promotion plans according to the needs of enterprises and are responsible for the execution and optimization of the plans. The advantages of third-party promoters are their professionalism and flexibility, and they can quickly respond to market changes and adjust promotion strategies. Third-party promoters will regularly track and analyze the data of promotion activities to evaluate the promotion effect. These data include multiple indicators such as website traffic, conversion rate, keyword ranking, etc. By analyzing these data, promoters can understand the advantages and disadvantages of the promotion activities and adjust strategies accordingly to optimize the promotion effect. For example, if it is found that the conversion rate of a certain keyword is low, the promoter may adjust the bid of the keyword advertisement or optimize the relevant pages of the website to increase the conversion rate.

[0004] Currently, we mainly rely on data provided by third-party promoters to count promotion results, but the tracking data of natural persons after promotion is missing, and it is difficult to implement the download, launch, and use of multiple versions of apps. In addition, the intermediate data is provided offline, and when providing promotion results to tenants, we rely heavily on manual entry of offline data into the system, which is time-consuming, labor-intensive, and extremely prone to errors.

[0005] To this end, the present invention proposes a SaaS-based app promotion result attribution method and system. Summary of the invention

[0006] In view of this, the present invention hopes to provide a SaaS-based app promotion result attribution method and system to solve or alleviate the technical problems existing in the prior art, namely, when it comes to the download, startup and use of multiple versions of apps, the lack of data may lead to inaccurate evaluation of promotion effects and it is difficult to accurately measure the performance of different versions of apps, and at least provide a beneficial option for this; the technical solution of the present invention is implemented as follows:

[0007] First, a SaaS-based app promotion result attribution method:

[0008] (1) Overview;

[0009] The present invention aims to solve the above technical problems, and effectively solves the technical defects involved in the evaluation of the promotion effect of multiple versions of Apps through comprehensive data source configuration, sophisticated tracking tool integration, strict data import and processing, thorough data cleaning and verification, and in-depth data analysis and reporting. It improves data quality and accuracy, enhances the accuracy of promotion effect evaluation, improves decision-making efficiency and effectiveness, and optimizes user experience and App performance.

[0010] (II) Technical solution:

[0011] To achieve the above technical objectives, the following steps S1 to S5 need to be performed.

[0012] 2.1 Step S1, data source configuration and access:

[0013] Configure the data source D required by the data import algorithm, including the data format F D , storage location L D and access rights P D .

[0014] Data source configuration function Configure(D,F D ,L D ,P D ) Accept data source D, data format F, storage location L D and access rights P D Takes as input, and returns a configured data source.

[0015] The data import function Import(D) receives a configured data source D and imports data.

[0016] 2.2 Step S2, tracking tool integration:

[0017] Integrate tracking tool SDKs in multiple versions of the app to ensure that user behavior data can be collected.

[0018] Configure tracking events and tracking logic.

[0019] 2.2.1 Step S200, integration:

[0020] For different versions V, the tracking tool SDK is integrated to collect user behavior data DataV.

[0021] The configured tracking event Event i,V In , i represents the index of the event and V represents the application version.

[0022] 2.2.2 Step S201, configure tracking logic:

[0023] Tracking logic is used to determine when to trigger a tracking event Event i,V :

[0024] Logic V (s,a)→{0,1};

[0025] Among them, s represents the application state, a represents the user behavior, and the function Logic V Determine whether to trigger a tracking event based on the application version V, current state s, and user behavior a (return 1 for triggering, return 0 for not triggering).

[0026] The configuration method is implemented by configuring function C:

[0027] C(V,{Event i,V},Logic V )→V″;

[0028] Where V represents the original application version, {Event i,V} indicates a collection of tracking events, Logic V represents the tracking logic, and V″ represents the application version after the tracking events and logic are configured.

[0029] 2.3 Step S3, data import and processing:

[0030] Start the data import algorithm to automatically import data from the data source into the system database.

[0031] Monitor the data import process to ensure data integrity and accuracy.

[0032] Handle errors or exceptions that may occur during the import process.

[0033] 2.3.1 Step S300, execute data import algorithm:

[0034] Includes the function ImportAlgorithm, which accepts the data source as input and imports the data into the system database. The Boolean function Integrity(d) represents the integrity of the data d, and returns true 1 if the data is complete, otherwise it returns false 0.

[0035] 2.3.2 Step S301, monitoring and processing:

[0036] The Boolean function Accuracy(d) verifies the accuracy of data d and returns true 1 if the data is accurate, otherwise it returns false 0.

[0037] The error handling function HandleError(e) is responsible for handling the exception e.

[0038] 2.4 Step S4, data cleaning and verification:

[0039] Perform deduplication on the imported data d to remove duplicate, erroneous or invalid data:

[0040]

[0041] 2.5 Step S5, data analysis and reporting:

[0042] Use data analysis algorithms to analyze imported and collected data. Extract key indicators and trends to evaluate the promotion effect and user behavior of multiple versions of apps. Conduct A / B variable testing to compare the performance of different versions of apps.

[0043] Generate reports Generate visual reports and dashboards based on the analysis results, including key indicator information a and trend analysis information b.

[0044] 2.5.1 Step S500, key indicators and trend extraction:

[0045] Update data d to data source D, which contains user behavior data of multiple versions of apps. Extract key indicators a, including user growth rate, activity, and retention rate: a = e(D);

[0046] Where e(D) is the extraction function.

[0047] Use time series analysis method to extract trend information b, including user growth trend and activity change trend: b = E(D);

[0048] Where E(D) is the time series analysis function.

[0049] 2.5.2 Step S501, A / B variable test:

[0050] Read the user data of any two versions A and B, DA and DB respectively. Compare the conversion rates c of the two versions m :

[0051] di=c m (D A ,D B );

[0052] According to the comparison results d i , determine which version performs better.

[0053] 2.5.2 Step S502, report generation:

[0054] The report contains key indicator information a and trend analysis information b.

[0055] (III) Mechanism for solving technical problems:

[0056] This solution first ensures the comprehensiveness and accuracy of the data source. By configuring the data source required by the data import algorithm, including data format, storage location, and access rights, the reliability and consistency of the data are guaranteed from the source. The tracking tool SDK is integrated in multiple versions of the App to achieve detailed tracking of user behavior. By configuring tracking events and tracking logic, every key behavior of the user in the process of using the App can be captured, providing a detailed data foundation for subsequent promotion effect evaluation.

[0057] The imported data is also thoroughly cleaned and verified to remove duplicate, erroneous or invalid data. The strict implementation of this step further improves the quality of the data and lays a solid foundation for subsequent data analysis. At the same time, data analysis algorithms are used to conduct in-depth analysis of the cleaned data to extract key indicators and trends. Through methods such as A / B variable testing, the performance of different versions of the App can be accurately compared, thereby accurately evaluating the promotion effect. Finally, visual reports and dashboards are generated so that decision makers can intuitively understand the analysis results and make wise decisions.

[0058] Second, a SaaS-based app promotion result attribution system:

[0059] Including: Promotional data synchronization interface used as user click and / or download interface, provided by third-party advertisers and obtained through the promotion link provided by this system;

[0060] The delivery platform and the creation of promotional content in accordance with the promotion plan, the promotional content includes the title, app type, source, content, channel, funding party, tenant and / or remarks of the promoted app or h5;

[0061] The link address generated by the promotion platform and provided to third-party advertisers; if it is an app promotion, after the promotion plan is created, the source and content configurations will be packaged and stored in the app package for tracing when the app is launched;

[0062] Promotion channel, receiving the link address;

[0063] After the promotion channel is released, when the user downloads or clicks, the data collection interface obtains the download and click data and stores it;

[0064] When the app starts, it sends a startup log with source and content parameters. The data collection service stores and saves the data. The BI data analysis service collects log data regularly and attributes the promotion of the app according to the delivery plan and startup log.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. Improve data quality and accuracy: Through comprehensive data source configuration and access, as well as strict data import and processing procedures, the technical solution of the present invention ensures the integrity and accuracy of the data. This avoids the problem of inaccurate promotion effect evaluation caused by missing or incorrect data, and provides a reliable data basis for decision makers.

[0067] 2. Enhance the accuracy of promotion effect evaluation: The tracking tool integration in the technical solution of the present invention can accurately capture the key behaviors of users in the process of using multiple versions of apps, such as downloading, launching, using, etc. This makes the evaluation of promotion effect more accurate and can accurately measure the performance differences of different versions of apps.

[0068] 3. Improve decision-making efficiency and effectiveness: Through in-depth data analysis and reporting, the technical solution of the present invention can extract key indicators and trends and generate visual reports and dashboards. This enables decision makers to intuitively understand the analysis results and make smart decisions quickly, thereby improving decision-making efficiency and effectiveness.

[0069] 4. Optimize user experience and App performance: By comprehensively analyzing the user behavior data of multiple versions of App, the technical solution of the present invention can help developers discover the pain points and needs of users during use, thereby optimizing the user experience. At the same time, the performance of different versions of App can also be analyzed to provide data support for the continuous optimization of App. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0071] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0072] Figure 2 It is a schematic diagram of the system composition of the present invention. DETAILED DESCRIPTION

[0073] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below;

[0074] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0075] Embodiment 1: Figure 1 As shown, this embodiment discloses a SaaS-based app promotion result attribution method, which includes the following steps S1 to S5.

[0076] In this embodiment, regarding step S1, data source configuration and access:

[0077] (1) Data source configuration function Configure(D,F D ,L D ,P D ):

[0078] D: Data source identifier, used to uniquely identify a data source.

[0079] FD: Data format, which defines the structure and type of data to ensure that the data can be correctly parsed and processed after import.

[0080] LD: Storage location, indicating the physical or virtual location where the data is stored, making it easier for the data import function to locate and access the data.

[0081] PD: Access permissions, including read, write and other permission settings, to ensure data security and compliance.

[0082] Configure the data source according to the input parameters to ensure that the data import algorithm can correctly connect to, access, and process the data in the data source.

[0083] Output: Returns a configured data source object that contains all the necessary information and connections for the data import function to use.

[0084] (2) Data import function Import(D):

[0085] D: The configured data source object, returned by the data source configuration function.

[0086] Use the configured data source object to connect to the actual data source and import data from it. The imported data will be used for subsequent attribution analysis.

[0087] Specifically, the processing flow of step S1 is:

[0088] 1) Connect to a data source using the connection information in the data source object, such as storage location and access permissions.

[0089] 2) Parse and import data according to the data format definition in the data source object.

[0090] 3) Perform preliminary quality checks and cleansing on the imported data to ensure data accuracy and consistency.

[0091] It is understandable that by clearly configuring the data format and storage location, it is ensured that the imported data meets the expected format and quality requirements, reducing the possibility of data errors and missing. By configuring access rights, access to sensitive data can be strictly controlled to prevent data leakage and improper use. The configuration and import processes of the data source are clearly separated, so that the imported data can be used more efficiently in subsequent data processing and analysis without repeating the configuration of the data source. When the data source changes (such as location changes, format updates, etc.), only the relevant parameters in the data source configuration function need to be updated without modifying the entire data import and processing process, reducing maintenance costs and improving the scalability of the system.

[0092] Furthermore, the python execution program of step S1 is as follows:

[0093]

[0094]

[0095] print(imported_data)

[0096] In the above program, the DataSource class is a data container that is used to store the configuration information of the data source, such as the data source identifier, data format, storage location, and access permissions.

[0097] The configure_data_source function accepts the configuration parameters of the data source and returns a configured DataSource object. The import_data function accepts a configured data source object and connects to the actual data source to import data. The function returns a placeholder indicating that the data has been imported, or returns the imported data itself in actual applications. Configure the data source by calling the configure_data_source function, and import the data by calling the import_data function.

[0098] In this embodiment, regarding step S2, tracking tool integration: integrating the tracking tool SDK in the multi-version app to ensure that user behavior data can be collected, and configuring tracking events and tracking logic. The goal of tracking tool integration is to integrate the tracking tool SDK in the multi-version App so that user behavior data can be collected, and tracking events and tracking logic can be configured, so as to accurately measure the performance of different versions of the App.

[0099] Specifically, step S200, integration: for different versions V, integrate the tracking tool SDK to collect user behavior data DataV. The configured tracking event Event i,V In , i represents the index of the event, and V represents the application version. The operation steps are:

[0100] 1) Select a tracking tool SDK that supports cross-version compatibility and can adapt to different versions of apps.

[0101] 2) Integrate the SDK into each version of the App to ensure that the SDK can run stably and collect user behavior data.

[0102] 3) Configure an independent tracking event set Eventi,V for each version V, where i represents the index of the event

[0103] Specifically, step S201, configure the tracking logic:

[0104] Tracking logic is used to determine when to trigger a tracking event Event i,V :

[0105] Logic V (s,a)→{0,1};

[0106] Among them, s represents the application state, a represents the user behavior, and the function Logic V Determine whether to trigger a tracking event based on the application version V, current state s, and user behavior a (return 1 for triggering, return 0 for not triggering).

[0107] Furthermore, the configuration method is implemented by configuring the function C:

[0108] C(V,{Event i,V},Logic V )→V″;

[0109] Where V represents the original application version, {Event i,V} indicates a collection of tracking events, Logic V represents the tracking logic, and V″ represents the application version after the tracking events and logic are configured.

[0110] It is understandable that by integrating the tracking tool SDK and configuring the tracking logic, it is possible to ensure that the tracking event is accurately triggered when the user performs a specific behavior, thereby collecting more accurate user behavior data. It helps to reduce data missing and errors and improve the accuracy of promotion effect evaluation. The integration of the tracking tool SDK enables user behavior data to be collected and recorded in real time, which is convenient for subsequent data analysis and attribution. It helps to track the changes in user behavior in different versions of the App, so as to more accurately evaluate the performance of different versions. By configuring function C, tracking events and logic can be flexibly configured for different versions of the App. This makes it easy to adjust the tracking strategy when the App version is updated or iterated to adapt to new business needs and data collection goals. Based on the collected user behavior data, a more accurate promotion effect evaluation can be performed on different versions of the App. It helps to discover the differences and advantages and disadvantages between different versions, so as to optimize the promotion strategy and improve the overall promotion effect.

[0111] Furthermore, the Python execution program of step S2 is as follows:

[0112]

[0113]

[0114] #Configure the tracking tool

[0115] tracking_sdk=configure_tracking(app_version,events,logic)

[0116] #Read user behavior

[0117] tracking_sdk.track_event("event1","home_screen","click")

[0118] tracking_sdk.track_event("event2","settings_screen","click")

[0119] tracking_sdk.track_event("event3","home_screen","scroll")

[0120] In the above program, TrackingToolSDK class is a class that represents the Tracking Tool SDK. It contains version information, event list, and tracking logic. The configure_events method is used to configure the events to be tracked. The configure_logic method is used to configure the tracking logic, that is, to determine when to trigger the tracking event. The track_event method is used to actually trigger the tracking event and decide whether to print the event information based on the tracking logic (in actual applications, the event data will be sent to the tracking server).

[0121] The configure_tracking function accepts the application version, event list, and tracking logic as input and returns a configured TrackingToolSDK instance.

[0122] In the above program, the application version app_version, the list of events to be tracked events, and the tracking logic logic are defined. The tracking logic is a lambda function that determines whether to trigger a tracking event based on the application version, current state, and user behavior. The track_event method is called to simulate different user behaviors in different states, and the tracking logic is used to determine whether to trigger a tracking event.

[0123] In this embodiment, regarding step S3, data import and processing: the goal of data import and processing is to automatically import data from the data source into the system database, and monitor the data import process to ensure the integrity and accuracy of the data, while handling errors or abnormal situations that may occur during the import process.

[0124] Specifically, in step S300, a data import algorithm is executed:

[0125] 1) Define the function ImportAlgorithm, which accepts the data source as an input parameter.

[0126] 2) The ImportAlgorithm function performs a data import operation to import data from the data source into the system database.

[0127] 3) Use the Boolean function Integrity(d) to verify the integrity of the data d. If the data is complete, it returns true (1); otherwise, it returns false (0).

[0128] Specifically, step S301, monitoring and processing:

[0129] 1) Define a Boolean function Accuracy(d) that verifies the accuracy of data d. If the data is accurate, it returns true (1); otherwise, it returns false (0).

[0130] 2) Define the error handling function HandleError(e), which is responsible for handling the exception e.

[0131] 3) During the data import process, monitor the integrity and accuracy of the data in real time.

[0132] 4) If the data is found to be incomplete or inaccurate, use the HandleError(e) function to handle the corresponding error or exception.

[0133] It is understandable that by monitoring the data import process and using

[0134] Using the Integrity(d) and Accuracy(d) functions to verify the integrity and accuracy of the data can ensure that the data imported into the system database is of high quality. This helps reduce data missing and errors and improve the accuracy of promotion effect evaluation. Define the HandleError(e) function to handle errors or abnormal situations that may occur during data import. This helps to promptly discover and solve problems and ensure the stability and reliability of the data import process. Automating the execution of data import algorithms can reduce the time and labor costs required for manual data import. At the same time, real-time monitoring and handling of problems during the data import process can further improve the efficiency of data import. Moreover, this data import and processing solution is suitable for data management of multiple versions of apps. By ensuring the integrity and accuracy of data from different versions of apps, the performance of different versions of apps can be more accurately measured, providing strong support for the formulation of promotion strategies.

[0135] Furthermore, the Python execution program of step S3 is as follows:

[0136] import pandas as pd

[0137] from sqlalchemy import create_engine

[0138]

[0139]

[0140] In the above program:

[0141] (1) The ImportAlgorithm function is responsible for reading data from a data source (assuming it is a CSV file) and importing it into the system database.

[0142] Use the pandas library to read the CSV file and return the data in the form of a DataFrame.

[0143] Use the SQLAlchemy library to create a database connection engine and import the DataFrame data into the database.

[0144] If an exception occurs during the import process, the HandleError function will be called to handle the error.

[0145] (2) The Integrity function is used to check the integrity of data and returns a Boolean value indicating whether the data is complete.

[0146] (3) The Accuracy function is used to check the accuracy of data. It returns a Boolean value indicating whether the data is accurate.

[0147] (4) The HandleError function is responsible for handling errors or exceptions. The execution flow is:

[0148] (4.1) First, call the ImportAlgorithm function to import data from the data source.

[0149] (4.2) Then, use the Integrity function to check the integrity of the data and print the corresponding message.

[0150] (4.3) Next, use the Accuracy function to check the accuracy of the data and print the corresponding message.

[0151] (4.4) If an error occurs during the import process, the HandleError function will be called to handle the error.

[0152] In this embodiment, regarding step S4, data cleaning and verification: the main goal of data cleaning and verification is to deduplicate the imported data to remove duplicate, erroneous or invalid data, ensure the accuracy and reliability of the data, and thus improve the accuracy of the promotion effect evaluation.

[0153] Deduplication: Remove duplicate records from a data set to ensure that each record is unique.

[0154] Use the Set data structure to store data because sets automatically remove duplicate elements.

[0155] Alternatively, use the database's unique constraint to ensure data uniqueness.

[0156] You can also use a hash table to record the data that has already appeared, thereby filtering out duplicate data:

[0157] For each element d in the data set D, if there is another element d' equal to it (d = d'), then d is still equal to itself (d =

[0158] d), which is obvious. But in the context of deduplication, the intention of this formula is to express "keep unique data elements".

[0159] It is understandable that by removing duplicate, erroneous or invalid data, the accuracy of the data set can be ensured, thereby improving the accuracy of the promotion effect evaluation. This is because duplicate or invalid data may lead to deviations in the evaluation results. Removing duplicate data can reduce the space requirements for data storage and reduce storage costs. In the subsequent data analysis and processing process, using the cleaned data set can improve processing efficiency because there is no need to process duplicate or invalid data. For the download, launch and use of multiple versions of App, data cleaning and verification can ensure that the data of different versions of App are accurate and reliable, thereby more accurately measuring the performance of different versions of App. It helps to formulate more effective promotion strategies.

[0160] Furthermore, the Python execution program of step S4 is as follows:

[0161]

[0162]

[0163] In the above program: The DataFrame contains columns such as user ID, App version, and download date.

[0164] The clean_data function takes a DataFrame as input and returns a cleaned DataFrame. Use the drop_duplicates method to remove duplicate rows from the DataFrame. This method will keep the first occurrence of the row and remove subsequent occurrences of the duplicate rows.

[0165] Call the clean_data function and pass the original data data as a parameter to it. The function returns the cleaned data cleaned_data. Print the cleaned DataFrame cleaned_data to see the cleaning effect.

[0166] In this embodiment, regarding step S5, data analysis and reporting: use data analysis algorithms to analyze the imported and collected data. Extract key indicators and trends to evaluate the promotion effect and user behavior of multiple versions of apps. Perform A / B variable testing to compare the performance of different versions of apps.

[0167] Generate reports Generate visual reports and dashboards based on the analysis results, including key indicator information a and trend analysis information b.

[0168] Specifically, step S500, key indicators and trend extraction: update data d to data source D, including user behavior data of multiple versions of apps. Extract key indicators a, including user growth rate, activity and retention rate: a = e(D);

[0169] Where e(D) is the extraction function.

[0170] Use time series analysis method to extract trend information b, including user growth trend and activity change trend: b = E(D);

[0171] Where E(D) is the time series analysis function.

[0172] Specifically, step S501, A / B variable test: read the user data of any two versions A and B, DA and DB respectively. Compare the conversion rates c of the two versions m :

[0173] di=c m (D A ,D B );

[0174] According to the comparison results d i , determine which version performs better.

[0175] Specifically, step S502, report generation: the report includes key indicator information a and trend analysis information b.

[0176] Furthermore, the Python execution program of step S5 is as follows:

[0177]

[0178]

[0179]

[0180]

[0181] In the above program: Data source D is a Pandas DataFrame that contains user behavior data of multiple versions of apps. The data columns include version, user ID, registration date, and active days.

[0182] The extract_key_metrics function calculates user growth rate, activity, and retention rate. The extract_trends function uses time series analysis to extract user growth trends and activity change trends. Here, the time series analysis is simplified and only the monthly user growth and activity changes are calculated. The compare_metrics function compares the conversion rates of two versions A and B. The conversion rate is the ratio of the number of active users to the total number of users. The generate_report function prints key indicators, trends, and A / B test results.

[0183] By writing automated data analysis scripts, you can quickly extract key indicators and trends, and conduct A / B testing, thereby improving the efficiency of data analysis. The generated report contains key indicators, trends, and A / B test results, which are essential for developing promotion strategies and optimizing app performance.

[0184] Decision makers can quickly make judgments based on the information in the report and formulate corresponding promotion strategies. By analyzing user behavior and app performance, potential problems and improvement points can be discovered to optimize the user experience. For example, if the retention rate of a certain version is found to be low, the reasons can be further analyzed and corresponding measures can be taken to improve the retention rate.

[0185] Embodiment 2: Figure 2 As shown, this embodiment discloses a SaaS-based app promotion result attribution system, which implements the following functions on the SaaS platform:

[0186] (1) Promotional data synchronization interface: The system serves as a promotional data synchronization interface for user clicks and / or downloads. The interface is provided by third-party advertisers. When a user clicks or downloads through a promotional link provided by the system, the relevant data will be synchronized to the system through this interface.

[0187] The purpose of this interface is to achieve real-time data synchronization and ensure that the system can capture all user behavior data generated through promotion links. This is the basis for subsequent data analysis and attribution.

[0188] Real-time synchronization of promotional data ensures the integrity and accuracy of the data, providing a reliable basis for subsequent data analysis and attribution. Through real-time synchronization, the system can capture all user behaviors generated through promotional links, avoiding data omissions and distortions, thereby improving data quality and reliability.

[0189] (2) Delivery platform and promotional content creation:

[0190] The delivery platform is responsible for creating the promotion content according to the promotion plan. The promotion content includes the title, app type, source, content, channel, funding party, tenant and / or remarks of the promoted app or h5.

[0191] The delivery platform is the starting point of the promotion activities. By creating detailed promotion content, it provides a clear direction and basis for subsequent promotion activities and data analysis.

[0192] Detailed promotion content settings ensure the pertinence and effectiveness of promotion activities, and also provide detailed dimensions and basis for subsequent data analysis and attribution. Through detailed promotion content settings, the system can more accurately capture the interests and needs of users, thereby formulating more targeted promotion strategies. At the same time, detailed promotion content also provides more dimensions and basis for subsequent data analysis and attribution, making the analysis results more accurate and comprehensive.

[0193] (3) Link address generation and app package configuration:

[0194] The promotion platform generates a link address for providing to third-party advertisers. If it is an app promotion, after creating the promotion plan, the source and content configurations will be packaged and stored in the app package for tracing when launching the app.

[0195] The purpose of generating a unique link address is to ensure that each promotion channel and user behavior can be accurately tracked and attributed. The source and content configurations are packaged and stored in the app package so that these promotion information can be obtained when the app is launched, thereby achieving more accurate attribution.

[0196] The unique link address and configuration information in the app package ensure the traceability and attributability of the promotion activities, providing an accurate data source for subsequent data analysis and attribution.

[0197] Through the unique link address and configuration information in the app package, the system can accurately track each promotion channel and user behavior, thus ensuring the accuracy and reliability of the data. This provides a solid foundation for subsequent data analysis and attribution.

[0198] (4) Promotion channels and data collection:

[0199] The promotion channel receives the link address, and when the user downloads or clicks, the data collection interface obtains and stores the download and click data. The promotion channel is a bridge connecting users and the system. The link address spread through the channel can capture the user's download and click behavior. The data collection interface is responsible for obtaining and storing this data in real time.

[0200] Real-time collection of user behavior data ensures the integrity and timeliness of the data, and provides timely and accurate data support for subsequent data analysis and attribution. By collecting user behavior data in real time, the system can capture the user's download and click behavior in a timely manner, thereby ensuring the integrity and timeliness of the data. This provides timely and accurate data support for subsequent data analysis and attribution, and improves the accuracy of promotion effect evaluation.

[0201] (5) App startup log and data analysis:

[0202] When the app starts, it sends a startup log with source and content parameters, and the data collection service stores the data. The BI data analysis service collects log data regularly and attributes the app's promotion according to the launch plan and the startup log; the startup log sent when the app starts contains promotion information (such as source and content), which enables the system to associate the user's startup behavior with specific promotion activities. The data collection service is responsible for storing this data, while the BI data analysis service is responsible for regularly collecting and analyzing this data, so as to achieve accurate attribution of the promotion effect.

[0203] Through the launch log and data analysis service, the system can accurately associate the user's launch behavior with specific promotion activities, thereby achieving accurate attribution of the promotion effect. This improves the accuracy of promotion effect evaluation and provides strong data support for optimizing promotion strategies. The launch log carries promotion information, allowing the system to associate the user's launch behavior with specific promotion activities. The data analysis service achieves accurate attribution of promotion effects through in-depth analysis of these data. This improves the accuracy of promotion effect evaluation and provides strong data support for optimizing promotion strategies.

[0204] All the above embodiments only express the implementation methods of the relevant practical applications of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

[0205] For those skilled in the art, it can be further appreciated that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0206] At the same time, those skilled in the art can understand that all or part of the processes in all the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

Claims

1. A SaaS-based app promotion result attribution method, characterized in that: The steps include: S1, configure the data source D required by the data import algorithm; S2, integrate the tracking tool SDK in multiple versions of the app to ensure that user behavior data can be collected; configure tracking events and tracking logic; S3, starts the data import algorithm, automatically imports data from the data source into the system database, and monitors the data import process; S4, performing a deduplication operation on the imported data d; S5, use data analysis algorithms to analyze the imported and collected data; extract key indicators and trends, and evaluate the promotion effect and user behavior of multiple versions of apps; Conduct A / B testing to compare the performance of different versions of your app. Generate reports Generate visual reports and dashboards based on the analysis results, including key indicator information a and trend analysis information b.

2. The method for attributing app promotion results according to claim 1, characterized in that: In S1, the data source D includes a data format F D , storage location L D and access rights P D ; Data source configuration function Configure(D,F D ,L D ,P D ) Accept data source D, data format F, storage location L D and access rights P D Takes as input, and returns a configured data source.

3. The method for attributing app promotion results according to claim 1, characterized in that: In said S2, the method of said integration and said tracking logic is: For different versions V, the integrated tracking tool SDK collects user behavior data DataV; the configured tracking event Event i,V In , i represents the index of the event, and V represents the application version; Tracking logic is used to determine when to trigger a tracking event Event i,V : Logic V (s,a)→{0,1}; Among them, s represents the application state, a represents the user behavior, and the function Logic V Determine whether to trigger a tracking event based on the application version V, current state s, and user behavior a.

4. The method for attributing app promotion results according to claim 3 is characterized by: In S2, the above configuration method is implemented by configuring function C: C(V,{Event i,V },Logic V )→V″; Where V represents the original application version, {Event i,V } indicates a collection of tracking events, Logic V represents the tracking logic, and V″ represents the application version after the tracking events and logic are configured.

5. The method for attributing app promotion results according to claim 1, characterized in that: In S3, the implementation method of the data import algorithm is: It includes the function ImportAlgorithm, which accepts the data source as input and imports the data into the system database; the Boolean function Integrity(d) represents the integrity of the data d, and returns true 1 if the data is complete, otherwise returns false 0.

6. The method for attributing app promotion results according to claim 5, characterized in that: In S3, the monitoring method is: The Boolean function Accuracy(d) verifies the accuracy of data d and returns true 1 if the data is accurate, otherwise it returns false 0; The error handling function HandleError(e) is responsible for handling the exception e.

7. The method for attributing app promotion results according to claim 1, characterized in that: In S4, the deduplication operation is to remove duplicate, erroneous or invalid data:

8. The method for attributing app promotion results according to claim 1, characterized in that: In S5, the extraction method of the data analysis algorithm is: Update data d to data source D, which contains user behavior data of multiple versions of apps; extract key indicators a, including user growth rate, activity, and retention rate: a = e(D); Among them, e(D) is the extraction function; Use time series analysis method to extract trend information b, including user growth trend and activity change trend: b = E(D); Where E(D) is the time series analysis function.

9. The method for attributing app promotion results according to claim 8, characterized in that: In S5, the method of the A / B variable test is: read the user data of any two versions A and B, DA and DB respectively. Compare the conversion rates c of the two versions m : di=c m (D A ,D B ); According to the comparison results d i , determine which version performs better.

10. A SaaS-based app promotion result attribution system using the app promotion result attribution method according to any one of claims 1 to 9, characterized in that: include: Promotion data synchronization interface used as user click and / or download interface; Placement platform and create promotional content according to promotion plan The link address generated by the promotion platform and provided to third-party advertisers; Promotion channel, receiving the link address; After the promotion channel is released, when the user downloads or clicks, the data collection interface obtains and stores the download and click data.

Images