Intelligent supermarket marketing strategy decision-making system and method based on user behaviors
By collecting multi-source user behavior data in real time to build multi-dimensional user profiles, the problem of the disconnect between supermarket marketing strategies and user scenarios has been solved, enabling precise marketing and adaptive optimization, and improving marketing efficiency and user satisfaction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XINGYI WANFENGLIN TOURISM GROUP CO LTD
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-24
AI Technical Summary
Existing supermarket marketing strategies lack a deep understanding of individual user behavior, resulting in wasted resources and low conversion rates. Furthermore, offline, non-intrusive behavioral data is difficult to integrate, leading to a disconnect between marketing strategies and user scenarios.
By collecting multi-source user behavior data in real time, we construct multi-dimensional user profiles, combine online and offline data to match precise marketing strategies, and optimize profile weights and strategy logic through feedback loops to achieve self-learning and self-adaptation.
It achieves precise matching between marketing strategies and user scenarios, improves the utilization rate of marketing resources and user experience, and has self-learning and self-adaptive capabilities.
Smart Images
Figure CN121921043A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart supermarket marketing strategy technology, specifically to a smart supermarket marketing strategy decision-making system and method based on user behavior. Background Technology
[0002] With the development of new retail formats, traditional supermarkets are accelerating their transformation towards digitalization and intelligence. Existing supermarket marketing relies heavily on human experience or static rules (such as store-wide discounts and member day offers), lacking a deep understanding of individual user behavior, resulting in problems such as wasted marketing resources, low conversion rates, and poor user experience.
[0003] Marketing strategy is a process by which businesses, starting from customer needs and drawing upon experience to obtain information on customer demand and purchasing power, as well as business expectations, systematically organize various business activities. Marketing strategy follows the 4Ps principle: product strategy, pricing strategy, distribution strategy, and promotion strategy, aiming to provide customers with satisfactory goods and services to achieve the company's goals.
[0004] Currently, in order to conduct personalized marketing to users, most companies have tried to introduce user profiling technology. However, data collection is limited to transaction records or APP clicks, making it difficult to integrate offline behaviors and easily leading to marketing strategies being out of touch with the user's current scenario, such as pushing new customer coupons to frequent shoppers.
[0005] Therefore, there is an urgent need for a smart supermarket marketing decision-making system and method that can integrate multi-source behavioral data and dynamically construct user profiles. Summary of the Invention
[0006] The technical problem to be solved by the present invention is to overcome the above-mentioned technical defects and provide a smart supermarket marketing strategy decision-making system and method based on user behavior, which facilitates multi-source data perception, accurate profile construction, and marketing strategy feedback evolution.
[0007] To solve the above-mentioned technical problems, the technical solution provided by the present invention is: a smart supermarket marketing strategy decision-making system based on user behavior, including a user behavior data collection module, a data preprocessing module, a profile building module, a marketing strategy matching module, and an output feedback module; The user behavior data collection module collects multi-source behavior data of users in the smart supermarket environment in real time, and the data preprocessing module processes the multi-source behavior data to obtain user behavior data. The profile building module constructs multi-dimensional user profile tags based on the processed user behavior data. The marketing strategy matching module is connected to the profile building module and matches the user profile tags with the scenario types in the marketing scenario library. The output feedback module will push the matched marketing strategies to supermarket apps, electronic price tags, SMS or in-store digital screens for execution, and collect user response behavior data to the marketing strategies. Based on the response behavior data, it will dynamically optimize the user profile tag weights and update the marketing strategy matching module.
[0008] Preferably, the multi-source behavioral data includes offline shopping trajectories, product browsing records, transaction records, membership card usage information, and mobile supermarket APP usage information; The user behavior data collection module includes one or more of the following devices deployed in supermarkets: cameras, electronic price tags, smart shopping carts, and POS terminal equipment, which collect user behavior data in a non-intrusive manner.
[0009] Preferably, the data preprocessing module includes cleaning, deduplication, completion, and standardization processing.
[0010] Preferably, the multi-dimensional profile tags include user spending power tags, product preference tags, shopping frequency tags, price sensitivity tags, and scenario demand tags; in: Statistical analysis of user payment amounts generates spending power tags; Product preference tags are generated by analyzing product interaction data and payment data; A shopping frequency tag is generated based on the number of purchases within a preset time period; Price sensitivity tags are generated by analyzing users' preferences for promotional products. The matching relationship between user location trajectory and the functions of the shopping mall area is obtained to generate scenario requirement tags.
[0011] Preferably, the marketing strategy matching module includes a marketing scenario library; The marketing scenario library includes new product promotion scenarios, inventory clearance scenarios, member-exclusive scenarios, holiday promotion scenarios, and real-time traffic generation scenarios. Each scenario is configured with scenario feature conditions, and matching is performed when the multi-dimensional profile tags meet the scenario feature conditions.
[0012] Preferably, the output feedback module calculates strategy effectiveness evaluation indicators after collecting user response behavior data to marketing strategies; The strategy effectiveness evaluation metrics include at least one of the following: coupon redemption rate, target product sales increase rate, change in user dwell time, and reduction in repurchase interval days.
[0013] Another aspect of the present invention discloses a marketing strategy decision-making method, comprising the following steps: S1: After a user enters the smart supermarket environment, continuously capture the interaction events generated by the user in the offline physical space and the mobile digital platform; S2: Perform spatiotemporal alignment and identity normalization on interactive events, remove noise and complete missing behavior chains to form continuous and consistent user behavior logs; S3: Quantify metrics based on behavior logs and map them to comparable profile tag values; S4: Input the profile tag value into the preset scenario trigger rule engine to determine whether the activation conditions of any of the marketing scenarios, such as new product trial, clearance of slow-moving products, high-value member awakening, holiday-themed marketing, or instant customer traffic guidance, are met. S5: When the conditions are met, the corresponding personalized intervention plan is retrieved from the strategy template library and pushed through the channel that the user is most likely to reach. S6: Monitor whether users generate positive behavioral feedback related to the intervention within a preset observation period, and generate an effect metric based on the feedback; S7: Based on the effect metric as a reinforcement signal, the update rate of the profile label and the scene matching threshold are adjusted in reverse.
[0014] The advantages of this invention compared to existing technologies are as follows: by integrating online and offline multi-source user behavior data, a dynamic and multi-dimensional user profile is constructed, enabling precise matching of marketing strategies with real-time user scenarios. This invention adopts non-intrusive data collection technology to fully perceive user trajectories, preferences, and consumption characteristics, and continuously optimizes profile weights and strategy logic based on feedback loops. It has self-learning and adaptive capabilities and is easy to promote and use. Attached Figure Description
[0015] Figure 1 This is a flowchart illustrating the decision-making process for smart supermarket marketing strategies. Detailed Implementation
[0016] The present invention will now be described in further detail with reference to the accompanying drawings.
[0017] Combined with appendix Figure 1 As shown, a smart supermarket marketing strategy decision-making system based on user behavior includes a user behavior data collection module, a data preprocessing module, a profile building module, a marketing strategy matching module, and an output feedback module.
[0018] In use, the user behavior data collection module collects multi-source behavior data of users in the smart supermarket environment in real time, and the data preprocessing module processes the multi-source behavior data to obtain user behavior data. The profile building module constructs multi-dimensional user profile tags based on the processed user behavior data. The marketing strategy matching module is connected to the profile building module and matches the user profile tags with the scenario types in the marketing scenario library. The output feedback module will push the matched marketing strategies to supermarket apps, electronic price tags, SMS or in-store digital screens for execution, and collect user response behavior data to the marketing strategies. Based on the response behavior data, it will dynamically optimize the user profile tag weights and update the marketing strategy matching module.
[0019] This includes the following steps: S1: After a user enters the smart supermarket environment, continuously capture the interaction events generated by the user in the offline physical space and the mobile digital platform; S2: Perform spatiotemporal alignment and identity normalization on interactive events, remove noise and complete missing behavior chains to form continuous and consistent user behavior logs; S3: Quantify metrics based on behavior logs and map them to comparable profile tag values; S4: Input the profile tag value into the preset scenario trigger rule engine to determine whether the activation conditions of any of the marketing scenarios, such as new product trial, clearance of slow-moving products, high-value member awakening, holiday-themed marketing, or instant customer traffic guidance, are met. S5: When the conditions are met, the corresponding personalized intervention plan is retrieved from the strategy template library and pushed through the channel that the user is most likely to reach. S6: Monitor whether users generate positive behavioral feedback related to the intervention within a preset observation period, and generate an effect metric based on the feedback; S7: Based on the effect metric as a reinforcement signal, the update rate of the profile label and the scene matching threshold are adjusted in reverse.
[0020] In specific implementation of the present invention, Multi-source behavioral data includes offline shopping trajectories, product browsing records, transaction records, membership card usage information, and mobile supermarket APP usage information. The user behavior data collection module includes one or more of the following devices deployed in the supermarket: cameras, electronic price tags, smart shopping carts, and POS terminal devices, which collect user offline behavior in a non-intrusive manner.
[0021] The data preprocessing module includes cleaning, deduplication, completion, and standardization. The multi-dimensional profile tags include user spending power tags, product preference tags, shopping frequency tags, price sensitivity tags, and scenario demand tags. in: Statistical analysis of user payment amounts generates spending power tags; Product preference tags are generated by analyzing product interaction data and payment data; A shopping frequency tag is generated based on the number of purchases within a preset time period; Price sensitivity tags are generated by analyzing users' preferences for promotional products. The matching relationship between user location trajectory and the functions of the shopping mall area is obtained to generate scenario requirement tags.
[0022] In one embodiment: The marketing strategy matching module includes a marketing scenario library; The marketing scenario library includes new product promotion scenarios, inventory clearance scenarios, member-exclusive scenarios, holiday promotion scenarios, and real-time traffic generation scenarios. Each scenario is configured with scenario feature conditions, and matching is performed when the multi-dimensional profile tags meet the scenario feature conditions.
[0023] The output feedback module collects user response data to marketing strategies and then calculates strategy effectiveness evaluation metrics. The strategy effectiveness evaluation metrics include at least one of the following: coupon redemption rate, target product sales increase rate, change in user dwell time, and reduction in repurchase interval days.
[0024] In practical use, this invention: Once a user enters the supermarket, the user behavior data collection module seamlessly collects their offline shopping trajectory, product browsing history, transaction records, and membership card usage information through devices such as cameras, smart shopping carts, electronic price tags, and POS terminals deployed within the store. Simultaneously, it acquires their operational behavior on the supermarket's mobile app, forming multi-source behavioral data covering all channels. This approach avoids reliance on user input, significantly improving data integrity and real-time performance. The data preprocessing module performs cleaning, deduplication, missing value completion, and standardization on the heterogeneous data, achieving spatiotemporal alignment and user identity normalization. This outputs a clear, continuous, and consistent user behavior log, laying a high-quality data foundation for subsequent analysis. Based on the processed logs, the profile building module calculates quantitative indicators for users in five dimensions: spending power (based on payment amount), product preference (combining product interaction and purchase records), shopping frequency (based on the number of visits or transactions at a preset period), price sensitivity (based on the tendency to choose promotional products), and scenario needs (matching location trajectory and supermarket functional areas). These indicators are then mapped to comparable multi-dimensional profile tags to achieve a refined characterization of user features. The marketing strategy matching module inputs the above profile tags into the preset scenario triggering rule engine and matches them with scenario types in the marketing scenario library, such as new product promotion, inventory clearance, member-exclusive, holiday promotion, and real-time traffic generation. Each scenario is configured with clear tag combinations and threshold conditions to ensure that the strategy activation accurately matches the user's current status and the supermarket's operational goals. After a successful match, the output feedback module retrieves the corresponding personalized intervention plan from the strategy template library, including targeted coupons, limited-time discounts, and product recommendations. During the preset observation period, the system continuously monitors whether users exhibit positive response behaviors such as redemption, purchase of target products, extended dwell time, or shortened repurchase intervals. Based on these behaviors, the system calculates at least one strategy effectiveness evaluation indicator to quantify marketing effectiveness. The evaluation results are used as reinforcement signals to dynamically adjust the update rate and confidence weight of each profile tag and optimize the matching rules and thresholds in the marketing scenario library, enabling the entire decision-making mechanism to have closed-loop learning and adaptive evolution capabilities.
[0025] The contents not described in detail in this specification are existing technologies known to those skilled in the art.
[0026] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0027] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.
Claims
1. A smart supermarket marketing strategy decision-making system based on user behavior, characterized in that: It includes a user behavior data collection module, a data preprocessing module, a profile building module, a marketing strategy matching module, and an output feedback module; The user behavior data collection module collects multi-source behavior data of users in the smart supermarket environment in real time, and the data preprocessing module processes the multi-source behavior data to obtain user behavior data. The profile building module constructs multi-dimensional user profile tags based on the processed user behavior data. The marketing strategy matching module is connected to the profile building module and matches the user profile tags with the scenario types in the marketing scenario library. The output feedback module will push the matched marketing strategies to supermarket apps, electronic price tags, SMS or in-store digital screens for execution, and collect user response behavior data to the marketing strategies. Based on the response behavior data, it will dynamically optimize the user profile tag weights and update the marketing strategy matching module.
2. The smart supermarket marketing strategy decision-making system based on user behavior according to claim 1, characterized in that: The multi-source behavioral data includes offline shopping trajectories, product browsing records, transaction details, membership card usage information, and mobile supermarket APP usage information; The user behavior data collection module includes one or more of the following devices deployed in supermarkets: cameras, electronic price tags, smart shopping carts, and POS terminal equipment, which collect user behavior data in a non-intrusive manner.
3. The intelligent supermarket marketing strategy decision-making system based on user behavior according to claim 1, characterized in that: The data preprocessing module includes cleaning, deduplication, completion, and standardization.
4. The intelligent supermarket marketing strategy decision-making system based on user behavior according to claim 1, characterized in that: The multi-dimensional profile tags include user spending power tags, product preference tags, shopping frequency tags, price sensitivity tags, and scenario-based demand tags; in: Statistical analysis of user payment amounts generates spending power tags; Product preference tags are generated by analyzing product interaction data and payment data; A shopping frequency tag is generated based on the number of purchases within a preset time period; Price sensitivity tags are generated by analyzing users' preferences for promotional products. The matching relationship between user location trajectory and the functions of the shopping mall area is obtained to generate scenario requirement tags.
5. The intelligent supermarket marketing strategy decision-making system based on user behavior according to claim 1, characterized in that: The marketing strategy matching module includes a marketing scenario library; The marketing scenario library includes new product promotion scenarios, inventory clearance scenarios, member-exclusive scenarios, holiday promotion scenarios, and real-time traffic generation scenarios. Each scenario is configured with scenario feature conditions, and matching is performed when the multi-dimensional profile tags meet the scenario feature conditions.
6. The intelligent supermarket marketing strategy decision-making system based on user behavior according to claim 1, characterized in that: The output feedback module calculates strategy effectiveness evaluation indicators after collecting user response behavior data to marketing strategies. The strategy effectiveness evaluation metrics include at least one of the following: coupon redemption rate, target product sales increase rate, change in user dwell time, and reduction in repurchase interval days.
7. The marketing strategy decision-making method according to any one of claims 1-6, characterized in that: Includes the following steps: S1: After a user enters the smart supermarket environment, continuously capture the interaction events generated by the user in the offline physical space and the mobile digital platform; S2: Perform spatiotemporal alignment and identity normalization on interactive events, remove noise and complete missing behavior chains to form continuous and consistent user behavior logs; S3: Quantify metrics based on behavior logs and map them to comparable profile tag values; S4: Input the profile tag value into the preset scenario trigger rule engine to determine whether the activation conditions of any of the marketing scenarios are met, such as new product trial, clearance of slow-moving products, high-value member reactivation, holiday-themed marketing, or instant customer traffic guidance. S5: When the conditions are met, the corresponding personalized intervention plan is retrieved from the strategy template library and pushed through the channel that the user is most likely to reach. S6: Monitor whether users generate positive behavioral feedback related to the intervention within a preset observation period, and generate an effect metric based on the feedback; S7: Based on the effect metric as a reinforcement signal, the update rate of the profile label and the scene matching threshold are adjusted in reverse.