Return on investment prediction method, apparatus, device, and computer readable medium

Abstract

The application relates to an investment return rate prediction method, device and equipment and a computer readable medium. The method comprises the following steps: obtaining total access amount data of a target commodity in an e-commerce platform, wherein the target commodity is a commodity for which a target object launches a promotion activity on multiple social media platforms to exchange resources in the e-commerce platform; extracting baseline access amount data from the total access amount data to obtain residual social media total contribution access amount data; determining an investment return rate of each promotion activity in each social media platform based on platform weights of the social media platforms, promotion index indexes of the promotion activities in each platform and the social media total contribution access amount data. The application solves the technical problem that existing investment return rate evaluation needs to collect user personal data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of marketing intelligence, and in particular, to an investment return rate prediction method, device, equipment and computer readable medium. BACKGROUND

[0002] With the advent of the information age, there are more and more types of new advertising marketing, including e-commerce advertising, information flow advertising and KOL content advertising. Information flow advertising is an advertisement located in the friend's dynamic of a social media user, or in the content flow of an information media and audio-visual media. The form of information flow advertising includes pictures, texts and videos, and the characteristics are algorithm recommendation, native experience, and can be targeted through tags, and can select exposure, landing page or application download according to the needs, and the final effect depends on the three key factors of creativity, targeting and bidding. KOL (Key Opinion Leader) refers to a person who has more and more accurate product information, and is accepted or trusted by the relevant group, and has a great influence on the purchase behavior of the group. Through new advertising marketing, the advertiser can often obtain a relatively high ROI (Return On Investment).

[0003] At present, in the related technology, the evaluation of the investment return rate is usually direct conversion and indirect conversion. Direct conversion such as KOL goods carrying short link conversion, information flow direct jump, etc., indirect conversion such as information collision of KOL fan ID in a social platform and member ID in an e-commerce platform, or collision of information flow device ID and e-commerce platform login device ID. However, in the related technology, in order to perform the above conversion evaluation, the user's personal information needs to be collected, used, processed and transmitted, which infringes the user's personal privacy. Moreover, it is difficult to define the investment return rate through a linear relationship of the resource input of the advertiser and the advertising data performance, so it is also unscientific to evaluate the investment return rate directly through the data performance. In addition, the promotion effect of different types of advertisements lacks the same horizontal evaluation standard, so it is difficult to evaluate the marketing effect of different advertisements.

[0004] At present, there is no effective solution to the problem that the existing investment return rate evaluation needs to collect user personal data. SUMMARY

[0005] The present application provides an investment return rate prediction method, device, equipment and computer readable medium to solve the technical problem that the existing investment return rate evaluation needs to collect user personal data.

[0006] According to an aspect of an embodiment of the present application, the present application provides an investment return rate prediction method, comprising:

[0007] Obtaining total visit data for a target product on an e-commerce platform, wherein the target product is a product for which a target object has launched promotional activities on multiple social media platforms in order to exchange resources on the e-commerce platform;

[0008] Extract the baseline visit data from the total visit data to obtain the remaining social media total contribution visit data;

[0009] The return on investment of each promotion activity on each social media platform is determined based on the platform weight of each social media platform, the promotion index index of the promotion activity on each platform, and the total contributed traffic data of social media.

[0010] Optionally, before extracting the baseline pageview data from the total pageview data, the method further includes determining the baseline pageview data in the following manner:

[0011] Determine the backtracking time;

[0012] Find the first visit data from the e-commerce platform's in-site search in the total visit data within the retrospective time range;

[0013] Sort the backtracking time by day based on the size of the first pageview data;

[0014] Select the target time after the target sort position in the sorting results;

[0015] The average value of the first pageview data within the target time is used as the baseline pageview data.

[0016] Optionally, before determining the return on investment of each promotional activity on each social media platform, the method further includes determining a platform weight of each social media platform, wherein the method includes determining the platform weight of any social media platform in the following manner:

[0017] Extracting secondary visit data derived from the target social media platform from the total contributed visit data of social media, and obtaining key indicator response data after the target object has invested resources on the target social media platform;

[0018] In a case where the second pageview data and the key indicator response data have a linear correlation relationship, the platform weight of the target social media platform is determined based on the linear correlation relationship.

[0019] Optionally, determining the platform weight of any social media platform further includes:

[0020] In the case where the second pageview data and the key indicator response data do not have a linear correlation relationship, the key indicator response data is input into a preset nonlinear regression model;

[0021] Obtain visit volume prediction data output by the nonlinear regression model;

[0022] When the similarity between the predicted visit volume data and the second visit volume data is greater than or equal to a target threshold, determining a weight parameter of the nonlinear regression model as a platform weight of the target social media platform;

[0023] When the similarity between the predicted traffic data and the second traffic data is less than the target threshold, the weight parameter of the nonlinear regression model is adjusted until the similarity between the predicted traffic data re-output by the nonlinear regression model and the second traffic data is greater than or equal to the target threshold, and the weight parameter is determined as the platform weight of the target social media platform.

[0024] Optionally, before determining the return on investment of each promotion campaign on each social media platform, the method further includes determining a promotion index for the promotion campaign on each platform, wherein the method includes determining the promotion index for any promotion campaign in the following manner:

[0025] Obtaining key indicator response data for a target promotion activity, wherein the target promotion activity includes at least one of a key opinion leader content promotion activity and an information flow promotion activity, and the key indicator response data includes at least one of the actual interaction volume of the key opinion leader content promotion activity and the click volume of the information flow promotion activity;

[0026] Determine the logarithmic value of the key indicator response data;

[0027] Determine the T-score curve of the logarithm;

[0028] The left-tail probability value in the T-score curve is determined as the promotion index of the target promotion activity, where the promotion index is the standardized probability of the key indicator response data.

[0029] Optionally, determining the return on investment of each promotional campaign in each social media platform includes determining the return on investment of any promotional campaign as follows:

[0030] Obtain the first promotion index index of the target key opinion leader content promotion activity and the first platform weight of the social media platform where the target key opinion leader content promotion activity is located;

[0031] The product of the first promotion index and the first platform weight is used as the return on investment of the target key opinion leader content promotion activity; or,

[0032] Obtain the second promotion index of the target information flow promotion activity and the second platform weight of the social media platform where the target information flow promotion activity is located;

[0033] The product of the second promotion index and the second platform weight is taken as the target information flow added to the investment return rate of the promotion activity.

[0034] Optionally, after determining the investment return rate of the promotion activity, the method further comprises determining the visitor quantity contributed by a single promotion activity in a target unit of time in the following manner:

[0035] determining the sum of the investment return rates of all promotion activities;

[0036] determining the quotient of the investment return rate of a single promotion activity and the sum of the investment return rates;

[0037] taking the product of the total social media contributed access quantity data in the target unit of time and the quotient as the visitor quantity contributed by a single promotion activity in the target unit of time.

[0038] According to another aspect of the embodiments of the present application, the present application provides an investment return rate prediction device, comprising:

[0039] a data acquisition module configured to acquire total access quantity data of a target commodity in an e-commerce platform, wherein the target commodity is a commodity for which a target object launches a promotion activity on a plurality of social media platforms to exchange resources in the e-commerce platform;

[0040] an incremental data extraction module configured to extract baseline access quantity data from the total access quantity data to obtain residual social media total contributed access quantity data;

[0041] an investment return rate prediction module configured to determine the investment return rate of each promotion activity in each social media platform based on the platform weight of each social media platform, the promotion index of the promotion activity in each platform, and the social media total contributed access quantity data.

[0042] According to another aspect of the embodiments of the present application, the present application provides an electronic device comprising a memory, a processor, a communication interface, and a communication bus, wherein the memory stores a computer program executable on the processor, the memory, the processor, and the communication interface communicate through the communication bus, and the processor implements the steps of the above method when executing the computer program.

[0043] According to another aspect of the embodiments of the present application, the present application further provides a computer readable medium having non-volatile program codes executable by a processor, wherein the program codes cause the processor to execute the above method.

[0044] The technical solution of the present application can be applied to marketing intelligence technology for prediction and optimization. The above technical solution provided by the embodiments of the present application has the following advantages compared with related technologies:

[0045] The technical scheme of the application is to obtain total access data of a target commodity in an e-commerce platform, wherein the target commodity is a commodity that a target object launches a promotion activity in multiple social media platforms to exchange resources in the e-commerce platform; baseline access data is extracted from the total access data to obtain residual social media total contribution access data; based on platform weights of the multiple social media platforms, promotion index numbers of the promotion activities in each platform, and the social media total contribution access data, the return on investment of each promotion activity in each social media platform is determined. The application starts from the total access of the target commodity in the e-commerce platform, the access of the multiple social media platforms, and other non-user data, cross-domain and standardized, and predicts the return on investment of each promotion activity in each social media platform, solving the technical problem that the existing evaluation of the return on investment needs to collect user personal data. BRIEF DESCRIPTION OF DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application together with the specification.

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the application or the related art, the accompanying drawings needed to be used in the embodiments or the related art description will be briefly introduced. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative labor.

[0048] Figure 1 An optional return on investment prediction method hardware environment schematic diagram provided according to the embodiments of the application;

[0049] Figure 2 An optional return on investment prediction method flowchart provided according to the embodiments of the application;

[0050] Figure 3 An optional return on investment prediction device block diagram provided according to the embodiments of the application;

[0051] Figure 4 An optional electronic device structure schematic diagram provided according to the embodiments of the application. DETAILED DESCRIPTION

[0052] To make the objectives, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the application.

[0053] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of this application and have no specific meaning. Therefore, "module" and "component" can be used interchangeably.

[0054] In the related art, the evaluation of return on investment is usually based on direct conversion and indirect conversion. Direct conversion includes conversion through KOL short links, direct jumps in information flows, etc. Indirect conversion includes information collision between the fan ID of the KOL on the social platform and the member ID on the e-commerce platform, or collision between the information flow device ID and the device ID for logging into the e-commerce platform. However, in order to conduct the above-mentioned conversion evaluation in the related art, it is necessary to collect, use, process, and transmit the user's personal information, which infringes on the user's personal privacy. Not only that, the collection of users' personal data is actually to obtain the data performance corresponding to the resources invested by advertisers (such as the number of likes, comments, collections, readings, video views, etc. of KOL ads on social platforms). However, it is difficult to define the advertiser's investment and data performance through a linear relationship. Therefore, it is unscientific to evaluate the return on investment directly through data performance. In addition, there is a lack of uniform horizontal evaluation standards for the promotion effects of different types of advertisements, making it difficult to evaluate the marketing effects of different advertisements.

[0055] In order to solve the problems mentioned in the background technology, according to one aspect of the embodiments of the present application, an embodiment of a method for predicting return on investment is provided.

[0056] Optionally, in the embodiment of the present application, the above investment return rate prediction method can be applied to Figure 1 In the hardware environment composed of the terminal 101 and the server 103 shown in FIG. Figure 1 As shown, the server 103 is connected to the terminal 101 via a network, and can be used to provide services (such as data collection, data extraction, linear regression analysis, nonlinear regression analysis, numerical prediction and other services) for the terminal or the client installed on the terminal. A database 105 can be set on the server or independently of the server to provide data storage services for the server 103. The above-mentioned network includes but is not limited to: wide area network, metropolitan area network or local area network, and the terminal 101 includes but is not limited to PC, mobile phone, tablet computer, etc.

[0057] In the embodiment of the present application, a method for predicting the rate of return on investment can be executed by the server 103, or can be executed jointly by the server 103 and the terminal 101, such as Figure 2 As shown, the method may include the following steps:

[0058] In step S202, total access amount data of a target commodity in an e-commerce platform is acquired, wherein the target commodity is a commodity for which a target object launches a promotion activity on a plurality of social media platforms to exchange resources in the e-commerce platform.

[0059] The essence of the technical solution of the present application is data processing, specifically, processing of non-user privacy data, and on this basis, predicting the return on investment of a promotion activity of an advertiser.

[0060] The e-commerce platform can be directly accessed through a corresponding e-commerce application program, and can also be accessed from an external link on a social media platform, so the total access amount data of the e-commerce platform includes independent access amounts of in-site searches of the e-commerce platform, and also includes independent access amounts from external links of a plurality of social media platforms. The present application can extract access amounts from external links of social media platforms from the total access amount data to analyze the return on investment of the actual promotion activities on the social media platforms and under the social media platforms. Each independent access amount is distinguished by its IP address, and an independent IP address is recorded only once in a day.

[0061] The plurality of social media platforms can be cross-domain platforms, such as instant messaging type social media platforms focusing on chat communication, game social media platforms focusing on games, forum type social media platforms focusing on literary audio-visual product evaluation, and information type social media platforms focusing on news, etc.

[0062] The target object is an advertiser, and the promotion activity launched by the target object is an advertisement of the target commodity. The advertiser launches an advertisement link of the target commodity on a social media platform, and a user can be guided to the e-commerce platform to access the target commodity through the advertisement link.

[0063] In step S204, baseline access amount data is extracted from the total access amount data to obtain residual social media total contribution access amount data.

[0064] In the embodiment of the present application, the baseline access amount data refers to the access amount of the target commodity in the e-commerce platform every day without any promotion activity of the target commodity by the advertiser. This part of access amount is extracted from the total access amount data to obtain the access amount guided from all social media platforms.

[0065] In step S206, based on the platform weight of each social media platform, the promotion index of each promotion activity in each platform, and the social media total contribution access amount data, the return on investment of each promotion activity in each social media platform is determined.

[0066] In the embodiment of the present application, each social media platform has its own corresponding weight, which is used to calculate the number of visits contributed by each social media platform. At the same time, the actual promotion activities on each social media platform also have their own promotion index. Combined with the platform weight of the social media platform, the return on investment of each promotion activity can be calculated.

[0067] In the embodiments of the present application, promotional activities include but are not limited to KOL (Key Opinion Leader) content advertising and information flow push advertising. A key opinion leader refers to a person who has more and more accurate product information, is accepted or trusted by the relevant group, and has a greater influence on the purchasing behavior of the group, that is, a blogger, big V, etc. who is widely followed on social media platforms and has a certain influence and appeal. KOL content advertising is when advertisers entrust KOLs to publish content posts such as relevant text promotions, audio and video promotions for the target product, attract the fans and passers-by of the KOL to discuss the target product, and generate likes and comments on the content post, which is the purpose of KOL content advertising. Information flow push advertising is when advertisers purchase more traffic for the content post published by the KOL on the basis of KOL content advertising, so that the content post can be more widely disseminated and seen by more people. The ultimate goal is to get more users to click on the link in the content post to enter the access page of the target product on the e-commerce platform.

[0068] Through steps S202 to S206, the present application uses non-user data such as the total number of visits to target products on the e-commerce platform and the number of visits directed by multiple social media platforms to perform cross-domain and standardized ROI predictions for each promotional activity on each social media platform. This solves the technical problem that existing ROI assessments require the collection of user personal data. Without requiring the collection of user privacy data, this application can also conduct targeted ROI analysis on multiple social media platforms in different fields and on promotional activities launched on each social media platform. Based on data processing, the contribution of social marketing activities to business objectives is quantified, providing a data foundation for advertisers to formulate advertising strategies.

[0069] Optionally, before extracting the baseline pageview data from the total pageview data, the method further includes determining the baseline pageview data in the following manner:

[0070] Step 1: Determine the backtracking time;

[0071] Step 2: Find the first pageview data from the e-commerce platform's in-site search within the total pageview data within the retrospective time range;

[0072] Step 3: sort the backtracking time in days according to the size of the first pageview data;

[0073] Step 4: Select the target time after the target sorting position in the sorting results;

[0074] Step 5: The average value of the first pageview data within the target time is used as the baseline pageview data.

[0075] In the embodiment of the present application, the above-mentioned retrospective time can be selected according to actual conditions and actual needs, such as the past three months (P3M), the past six months (P6M), etc. The first visit data derived from the search within the e-commerce platform is the visit data of the target product directly accessed through the search within the e-commerce platform. Therefore, taking the retrospective time of the past three months as an example, in order to determine the baseline visit data, it is necessary to find the first visit data of the site search within the past three months, and sort the first visit data in units of days. The target time after the above-mentioned target sorting position can be the bottom 10% of the days, and finally the average of the first visits per day is calculated for the bottom 10% of the days, and the average value is used as the baseline visit data.

[0076] Because different social media platforms have varying user bases and mechanisms, the key metrics for generating returns on advertisers' investment also vary. Therefore, it's necessary to quantify the weight of each social media platform and the contribution of each platform's key metrics to visitor traffic. The following explains how to quantify platform weights.

[0077] Optionally, before determining the return on investment of each promotional activity on each social media platform, the method further includes determining a platform weight of each social media platform, wherein the method includes determining the platform weight of any social media platform in the following manner:

[0078] Step 1: extracting second visit data derived from the target social media platform from the total contributed visit data of social media, and obtaining key indicator response data after the target object invests resources on the target social media platform;

[0079] Step 2: When the second pageview data and the key indicator response data have a linear correlation, determine the platform weight of the target social media platform based on the linear correlation.

[0080] In this embodiment of the present application, the second visit data is the visit data contributed by the target social media platform. The key indicator response data after the advertiser invests resources on the target social media platform can be the water-free interaction volume of the KOL content advertisement mentioned above, or the click-through rate of the information flow promotion advertisement mentioned above. These are the key indicators of the social media platform. Regression analysis is performed using the linear correlation between the second visit data and the key indicator response data after the advertiser invests resources on the target social media platform, such as:

[0081] Y(Increasementalsearch)=linerregressionanalysis(k)*X

[0082] Among them, Y (Increasemental search) is the second visit data, X is the key indicator response data, that is, the de-watered interaction volume of all KOL content ads and the click-through rate of information flow promotion ads on the target social media platform, and linerregressionanalysis(k) is the linear regression equation. Through the linear regression equation, the platform weight k can be calculated to quantify the contribution of key indicators of different platforms to visit volume.

[0083] However, not all second pageview data and key indicator response data have a linear correlation relationship. For second pageview data and key indicator response data with a nonlinear relationship, the platform weight can be calculated through nonlinear regression analysis.

[0084] Optionally, determining the platform weight of any social media platform further includes:

[0085] Step 1: When the second pageview data and the key indicator response data do not have a linear correlation relationship, the key indicator response data is input into a preset nonlinear regression model;

[0086] Step 2: Obtain the visit volume prediction data output by the nonlinear regression model;

[0087] Step 3: When the similarity between the predicted pageview data and the second pageview data is greater than or equal to a target threshold, a weight parameter of the nonlinear regression model is determined as the platform weight of the target social media platform;

[0088] Step 4: When the similarity between the predicted visit data and the second visit data is less than the target threshold, adjust the weight parameter of the nonlinear regression model until the similarity between the predicted visit data re-output by the nonlinear regression model and the second visit data is greater than or equal to the target threshold, and determine the weight parameter as the platform weight of the target social media platform.

[0089] In the embodiment of the present application, weight calculation can be performed by a nonlinear regression model such as XGBoost, which is a supervised model and is essentially a pile of CART trees. The calculation idea is to use a pile of trees to perform feature screening and make predictions, and finally add the predicted values ​​of each tree together as the final predicted value. By building a tree, the feature weight of each feature can be obtained. Therefore, it is only necessary to input the key indicator response data of different social media platforms into the nonlinear regression model, compare the visit prediction value output by the nonlinear regression model with the real second visit data, and when the similarity between the two is greater than or equal to the target threshold, use the feature weight calculated by the nonlinear regression model as the platform weight of the current social media platform. Otherwise, continue to train the nonlinear regression model until the visit prediction data output is greater than or equal to the target threshold, and use the feature weight calculated by the nonlinear regression model as the platform weight of the current social media platform. This is a method for quantifying the contribution factors of the key indicators of different platforms to the visit volume by means of nonlinear regression analysis.

[0090] In the embodiment of the present application, the objective function of the nonlinear regression model can be:

[0091]

[0092] That is, the square of the deviation between the actual number of visits (the second number of visits data) and the predicted number of visits.

[0093] Due to the different user scales and mechanisms of different social media platforms, the scales of key indicators such as interaction volume and click volume on different social media platforms will also vary. Therefore, for the key indicators of each social media platform, it is necessary to standardize the key indicators within the platform to obtain the promotion indicator index, so as to prepare for horizontal comparison and rationalization of the contribution value (return on investment) calculation of the next cross-platform promotion activities.

[0094] Optionally, before determining the return on investment of each promotion campaign on each social media platform, the method further includes determining a promotion index for the promotion campaign on each platform, wherein the method includes determining the promotion index for any promotion campaign in the following manner:

[0095] Step 1: Obtain key indicator response data for a target promotion campaign, wherein the target promotion campaign includes at least one of a key opinion leader content promotion campaign and an information flow promotion campaign, and the key indicator response data includes at least one of the actual interaction volume of the key opinion leader content promotion campaign and the click volume of the information flow promotion campaign;

[0096] Step 2, determining the logarithmic value of the key indicator response data;

[0097] Step 3, determine the T score curve of the logarithmic value;

[0098] Step 4: Determine the left-tail probability value in the T-score curve as the promotion index of the target promotion activity, where the promotion index is the normalized probability of the key indicator response data.

[0099] In the present embodiment, the key indicator response data, i.e., the de-watered interactions of all KOL content ads and click-throughs of information flow promotion ads on the target social media platform, is normalized in three steps: calculating the logarithm of the key indicator response data, performing a T-score calculation on the logarithm, and finally taking the left-tail probability value of the T-score curve to determine the promotion indicator index of the target promotion activity. The promotion indicator index is the normalized probability value.

[0100] Alternatively, key indicator normalization can be performed through other methods, such as Min-Max normalization and Z-Score normalization. Min-Max normalization is also known as 0-1 normalization, linear function normalization, and deviation normalization. The specific method to choose can be determined based on the key indicators of the specific category. For example, categories with a wide range of numerical performance, such as beauty products and maternal and child products, are suitable for Min-Max normalization. Min-Max normalization is a linear transformation of the original data so that the result falls within the range [0, 1]. Its conversion function is as follows:

[0101]

[0102] Among them, x is the key indicator response data before standardization, x * is the normalized index, min is the minimum value of all key indicator response data, and max is the maximum value of all key indicator response data.

[0103] Optionally, determining the return on investment of each promotional campaign in each social media platform includes determining the return on investment of any promotional campaign as follows:

[0104] Obtain the first promotion index index of the target key opinion leader content promotion activity and the first platform weight of the social media platform where the target key opinion leader content promotion activity is located;

[0105] The product of the first promotion index and the first platform weight is used as the return on investment of the target key opinion leader content promotion activity; or,

[0106] Obtain the second promotion index of the target information flow promotion activity and the second platform weight of the social media platform where the target information flow promotion activity is located;

[0107] The product of the second promotion indicator index and the second platform weight is used as the return on investment of the target information flow promotion activity.

[0108] In the embodiments of the present application, whether it is a horizontal comparison between multiple KOL content advertisements and information flow promotion advertisements in cross-domain social media platforms or a horizontal comparison between multiple KOL content advertisements and information flow promotion advertisements on the same social media platform, cross-domain comparison can be achieved by predicting the standardized return on investment.

[0109] Optionally, after determining the return on investment of the promotion activity, the method further includes determining the number of visitors contributed by the single promotion activity within the target unit time in the following manner:

[0110] Step 1: Determine the total ROI of all promotional activities;

[0111] Step 2: Determine the quotient of the return on investment of a single promotion activity and the total return on investment;

[0112] Step 3: The product of the total contributed visit data of social media in the target unit time and the quotient value is used as the visitor volume contributed by a single promotion activity in the target unit time.

[0113] Finally, to determine the daily contribution of a single campaign to visits, you can divide the ROI of that single campaign by the ratio of the ROI of all campaigns on that day, and multiply it by the total incremental visits contributed by the social media platform on that day to get the total visits contributed by the single campaign on that day.

[0114] By adopting the technical solution of the present application, based on the data processing method, not only can the return on investment analysis be performed without collecting user privacy data, but it can even conduct a unified and standardized cross-domain horizontal comparison of the return on investment of different social media platforms, actual KOL content advertisements on each platform, and information flow push advertisements, thereby achieving quantitative measurement of cross-domain attribution, opening up the connection between the grass-roots and push advertisements of the social marketing platform and the post-link e-commerce sales conversion platform, eliminating the natural attribute differences between the various social platforms, and allowing the contribution of diversified social marketing methods to commercial output to be compared horizontally.

[0115] According to another aspect of the embodiment of the present application, Figure 3 As shown, a device for predicting return on investment is provided, comprising:

[0116] The data acquisition module 301 is used to obtain the total visit data of the target product on the e-commerce platform, wherein the target product is a product for which the target object launches promotion activities on multiple social media platforms in order to exchange resources on the e-commerce platform;

[0117] The incremental data extraction module 303 is used to extract the baseline visit data from the total visit data to obtain the remaining social media total contribution visit data;

[0118] The ROI prediction module 305 is used to determine the ROI of each promotion activity on each social media platform based on the platform weight of each social media platform, the promotion index index of the promotion activity on each platform, and the total contributed visit volume data of the social media.

[0119] It should be noted that the data acquisition module 301 in this embodiment can be used to execute step S202 in the embodiment of this application, the incremental data extraction module 303 in this embodiment can be used to execute step S204 in the embodiment of this application, and the return on investment prediction module 305 in this embodiment can be used to execute step S206 in the embodiment of this application.

[0120] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments. Figure 1 In the hardware environment shown, it can be implemented by software or by hardware.

[0121] Optionally, the return on investment prediction device further includes a baseline visit data determination module, which is used to:

[0122] Determine the backtracking time;

[0123] Find the first visit data from the e-commerce platform's in-site search in the total visit data within the retrospective time range;

[0124] Sort the backtracking time by day based on the size of the first pageview data;

[0125] Select the target time after the target sort position in the sorting results;

[0126] The average value of the first pageview data within the target time is used as the baseline pageview data.

[0127] Optionally, the return on investment prediction device further includes a platform weight determination module, which is used to:

[0128] Extracting secondary visit data derived from the target social media platform from the total contributed visit data of social media, and obtaining key indicator response data after the target object has invested resources on the target social media platform;

[0129] In a case where the second pageview data and the key indicator response data have a linear correlation relationship, the platform weight of the target social media platform is determined based on the linear correlation relationship.

[0130] Optionally, the platform weight determination module is further configured to:

[0131] In the case where the second pageview data and the key indicator response data do not have a linear correlation relationship, the key indicator response data is input into a preset nonlinear regression model;

[0132] Obtain visit volume prediction data output by the nonlinear regression model;

[0133] When the similarity between the predicted visit volume data and the second visit volume data is greater than or equal to a target threshold, determining a weight parameter of the nonlinear regression model as a platform weight of the target social media platform;

[0134] When the similarity between the predicted traffic data and the second traffic data is less than the target threshold, the weight parameter of the nonlinear regression model is adjusted until the similarity between the predicted traffic data re-output by the nonlinear regression model and the second traffic data is greater than or equal to the target threshold, and the weight parameter is determined as the platform weight of the target social media platform.

[0135] Optionally, the investment return rate prediction device further includes a promotion indicator index determination module, which is used to:

[0136] Obtaining key indicator response data for a target promotion activity, wherein the target promotion activity includes at least one of a key opinion leader content promotion activity and an information flow promotion activity, and the key indicator response data includes at least one of the actual interaction volume of the key opinion leader content promotion activity and the click volume of the information flow promotion activity;

[0137] Determine the logarithmic value of the key indicator response data;

[0138] Determine the T-score curve of the logarithm;

[0139] The left-tail probability value in the T-score curve is determined as the promotion index of the target promotion activity, where the promotion index is the standardized probability of the key indicator response data.

[0140] Optionally, the return on investment prediction module is specifically used to:

[0141] Obtain the first promotion index index of the target key opinion leader content promotion activity and the first platform weight of the social media platform where the target key opinion leader content promotion activity is located;

[0142] The product of the first promotion index and the first platform weight is used as the return on investment of the target key opinion leader content promotion activity; or,

[0143] Obtain the second promotion index of the target information flow promotion activity and the second platform weight of the social media platform where the target information flow promotion activity is located;

[0144] The product of the second promotion indicator index and the second platform weight is used as the return on investment of the target information flow promotion activity.

[0145] Optionally, the return on investment prediction device further includes a contributed visitor quantity determination module, which is configured to:

[0146] Determine the total ROI of all promotional activities;

[0147] Determine the quotient of the ROI of a single campaign and the total ROI;

[0148] The product of the total social media contribution visit data within the target unit time and the quotient value is taken as the number of visitors contributed by a single promotion activity within the target unit time.

[0149] By adopting the technical solution of the present application, based on the data processing method, not only can the return on investment analysis be performed without collecting user privacy data, but it can even conduct a unified and standardized cross-domain horizontal comparison of the return on investment of different social media platforms, actual KOL content advertisements on each platform, and information flow push advertisements, thereby achieving quantitative measurement of cross-domain attribution, opening up the connection between the grass-roots and push advertisements of the social marketing platform and the post-link e-commerce sales conversion platform, eliminating the natural attribute differences between the various social platforms, and allowing the contribution of diversified social marketing methods to commercial output to be compared horizontally.

[0150] According to another aspect of the embodiment of the present application, the present application provides an electronic device, such as Figure 4 As shown, it includes a memory 401, a processor 403, a communication interface 405 and a communication bus 407. The memory 401 stores a computer program that can be run on the processor 403. The memory 401 and the processor 403 communicate through the communication interface 405 and the communication bus 407. When the processor 403 executes the computer program, the steps of the above method are implemented.

[0151] The memory and processor in the electronic device communicate via a communication bus and a communication interface. The communication bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus may be divided into an address bus, a data bus, a control bus, and the like.

[0152] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0153] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0154] According to another aspect of the embodiments of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of any of the above embodiments.

[0155] Optionally, in an embodiment of the present application, the computer-readable medium is configured to store program codes for the processor to execute the following steps:

[0156] Obtaining total visit data for a target product on an e-commerce platform, wherein the target product is a product for which a target object has launched promotional activities on multiple social media platforms in order to exchange resources on the e-commerce platform;

[0157] Extract the baseline visit data from the total visit data to obtain the remaining social media total contribution visit data;

[0158] The return on investment of each promotion activity on each social media platform is determined based on the platform weight of each social media platform, the promotion index index of the promotion activity on each platform, and the total contributed traffic data of social media.

[0159] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0160] When implementing the embodiments of the present application, reference may be made to the above embodiments, which have corresponding technical effects.

[0161] It is understood that the embodiments described herein may be implemented using hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit may be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or a combination thereof.

[0162] For software implementation, the technology described herein can be implemented by a unit that performs the functions described herein. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or outside the processor.

[0163] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. 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 beyond the scope of this application.

[0164] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0165] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0166] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0167] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0168] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application are essentially or partly contributed to the prior art or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard drive, a ROM, a RAM, a magnetic disk, or an optical disk. It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without further constraints, an element defined by the phrase "comprises a..." does not preclude the existence of additional identical elements in the process, method, article or apparatus that includes the element.

[0169] The foregoing is merely a list of specific embodiments of the present application, intended to enable those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the broadest scope consistent with the principles and novel features of the present application.

Claims

1. A method for predicting return on investment, characterized in that: include: Obtaining total visit data for a target product on an e-commerce platform, wherein the target product is a product for which a target object has launched promotional activities on multiple social media platforms in order to exchange resources on the e-commerce platform; Extracting baseline visit data from the total visit data to obtain the remaining social media total contribution visit data, wherein the baseline visit data refers to the daily visit data of the target product on the e-commerce platform without any promotion activities for the target product; extracting the baseline visit data from the total visit data, including determining the baseline visit data in the following manner: determining a lookback time; finding the first visit data from the total visit data that originated from the search on the e-commerce platform within the lookback time range; sorting the lookback time in units of days according to the size of the first visit data; selecting a target time after the target sorting position in the sorting result; and taking the average value of the first visit data within the target time as the baseline visit data; Based on the platform weights of various social media platforms, the promotion index index of the promotion activities on each platform, and the total contributed visit data of the social media, the return on investment of each promotion activity on each social media platform is determined, wherein the promotion index index is a normalized probability value of the key indicator response data of the promotion activities on each platform, and the key indicator response data includes at least one of the actual interaction volume of the key opinion leader content promotion activity and the click volume of the information flow promotion activity; determining the return on investment of each promotion activity on each social media platform includes determining the return on investment of any promotion activity in the following manner: obtaining a first promotion index index of a target key opinion leader content promotion activity and a first platform weight of the social media platform where the target key opinion leader content promotion activity is located; and taking the product of the first promotion index and the first platform weight as the return on investment of the target key opinion leader content promotion activity; or obtaining a second promotion index index of a target information flow promotion activity and a second platform weight of the social media platform where the target information flow promotion activity is located; and taking the product of the second promotion index and the second platform weight as the return on investment of the target information flow promotion activity.

2. The method according to claim 1, characterized in that Before determining the return on investment of each promotional activity on each social media platform, the method further includes determining a platform weight of each social media platform, wherein the method includes determining the platform weight of any social media platform in the following manner: Extracting second pageview data derived from the target social media platform from the total contributed pageview data of the social media, and obtaining key indicator response data after the target object invests resources on the target social media platform; In a case where the second pageview data and the key indicator response data have a linear correlation relationship, the platform weight of the target social media platform is determined based on the linear correlation relationship.

3. The method according to claim 2, characterized in that Determining the platform weight of any social media platform also includes: In a case where the second pageview data and the key indicator response data do not have a linear correlation relationship, inputting the key indicator response data into a preset nonlinear regression model; Obtaining visit volume prediction data output by the nonlinear regression model; When the similarity between the predicted pageview data and the second pageview data is greater than or equal to a target threshold, determining the weight parameter of the nonlinear regression model as the platform weight of the target social media platform; When the similarity between the predicted visit data and the second visit data is less than the target threshold, the weight parameter of the nonlinear regression model is adjusted until the similarity between the predicted visit data re-output by the nonlinear regression model and the second visit data is greater than or equal to the target threshold, and the weight parameter is determined as the platform weight of the target social media platform.

4. The method according to any one of claims 1 to 3, characterized in that: Before determining the return on investment of each promotional activity on each social media platform, the method further includes determining a promotional index for the promotional activity on each platform, wherein the method includes determining the promotional index for any promotional activity in the following manner: Acquiring key indicator response data for a target promotion activity, wherein the target promotion activity includes at least one of a key opinion leader content promotion activity and an information flow promotion activity; Determining the logarithmic value of the key indicator response data; determining a T-score curve for the logarithmic values; The left-tail probability value in the T-score curve is determined as the promotion indicator index of the target promotion activity.

5. The method according to claim 1, characterized in that After determining the return on investment of the promotional activity, the method further includes determining the number of visitors contributed by the single promotional activity within the target unit time in the following manner: Determine the total ROI of all promotional activities; Determining a quotient of the return on investment of the single promotional activity and the sum of the return on investments; The product of the total contributed visit volume data of the social media within the target unit time and the quotient value is taken as the visitor volume contributed by the single promotion activity within the target unit time.

6. A device for predicting return on investment, characterized in that: include: A data acquisition module is used to obtain total visit data of target products on the e-commerce platform, wherein the target products are products for which the target object launches promotion activities on multiple social media platforms in order to exchange resources on the e-commerce platform; An incremental data extraction module is used to extract baseline visit data from the total visit data to obtain the remaining social media total contribution visit data, wherein the baseline visit data refers to the daily visit data of the target product on the e-commerce platform in the absence of any promotional activities for the target product; extracting the baseline visit data from the total visit data includes determining the baseline visit data in the following manner: determining a lookback time; finding the first visit data in the total visit data that originated from the search on the e-commerce platform within the lookback time range; sorting the lookback time in units of days according to the size of the first visit data; selecting a target time after the target sorting position in the sorting result; and taking the average value of the first visit data within the target time as the baseline visit data; A return on investment prediction module is configured to determine the return on investment of each promotional activity on each social media platform based on the platform weight of each social media platform, the promotional index index of the promotional activity on each platform, and the total contributed visit data of the social media, wherein the promotional index is a normalized probability value of the key indicator response data of the promotional activity on each platform, and the key indicator response data includes at least one of the actual interaction volume of the key opinion leader content promotion activity and the click volume of the information flow promotion activity; determining the return on investment of each promotional activity on each social media platform includes determining the return on investment of any promotional activity in the following manner: obtaining a first promotional index index of a target key opinion leader content promotion activity and a first platform weight of the social media platform where the target key opinion leader content promotion activity is located; and using the product of the first promotional index and the first platform weight as the return on investment of the target key opinion leader content promotion activity; or obtaining a second promotional index index of a target information flow promotion activity and a second platform weight of the social media platform where the target information flow promotion activity is located; and using the product of the second promotional index and the second platform weight as the return on investment of the target information flow promotion activity.

7. An electronic device comprising a memory, a processor, a communication interface, and a communication bus, wherein the memory stores a computer program that can be run on the processor, and the memory and the processor communicate via the communication bus and the communication interface, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable medium having a non-volatile program code executable by a processor, characterized in that The program code enables the processor to execute the method according to any one of claims 1 to 5.

Images