Sales performance attribution analysis system and method based on store retail terminals

Abstract

The present invention relates to the field of retail data analysis technology, and specifically to a sales performance attribution analysis system and method based on store retail terminals. The data acquisition module provided by the present invention collects comprehensive data from multiple key data sources to ensure that no important information is missed, and provides a comprehensive and in-depth sales performance attribution analysis for store retail terminals. By accurately quantifying the contribution of customer factors, product factors and store operation factors to sales performance, store managers can more effectively allocate market promotion resources and design personalized marketing strategies based on accurate data, optimize product combinations, adjust pricing strategies and improve display methods to improve product sales efficiency and gross profit contribution. In addition, in terms of store operations, evaluating factors such as promotional activities and employee performance is helpful to reasonably arrange promotion plans and motivate employees, thereby improving the overall operation efficiency of stores.

Description

Technical Field

[0001] The present invention relates to the technical field of retail data analysis, and in particular to a sales performance attribution analysis system and method based on store retail terminals. Background Art

[0002] With the rapid development of modern technology and the economy, the retail industry has undergone tremendous changes, and the operating environment of in-store retail terminals has become increasingly complex. On the one hand, consumer demands and purchasing behaviors have become more diverse and unpredictable. Consumers are not only concerned with product price and quality, but also have higher expectations for shopping experience, personalized service, and the social attributes of products. For example, the development of online shopping platforms allows consumers to easily compare prices and product information across stores, posing a challenge to the pricing strategies and product differentiation of traditional stores.

[0003] On the other hand, market competition is becoming increasingly fierce, with the continuous emergence of new retail formats, including offline stores of e-commerce platforms, the expansion of convenience stores, and the rise of various themed specialty stores. These competitors are vying for limited market share, forcing traditional retail outlets to constantly seek ways to increase sales and profits.

[0004] However, current methods for analyzing store sales performance have significant shortcomings. Many traditional analytical methods focus solely on superficial indicators such as sales volume and sales volume, or simply analyze from a single dimension (such as the product dimension or customer dimension). For example, some stores determine the purchase volume of goods based solely on historical sales data, without considering the combined effects of multiple factors such as changes in customer behavior, the impact of promotional activities, and store layout. This one-sided or simplistic analysis method fails to fully and deeply analyze the many factors that affect sales performance, making it difficult for retailers to accurately understand the reasons for performance changes and unable to develop effective strategies to optimize operations and improve performance, putting them at a disadvantage in the fierce market competition.

[0005] Therefore, there is an urgent need for a system and method that can comprehensively and accurately perform attribution analysis on the sales performance of retail terminals in stores. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to propose a sales performance attribution analysis system and method based on store retail terminals, so as to provide a system and method that can comprehensively and accurately perform attribution analysis on the sales performance of store retail terminals.

[0007] Based on the above objectives, the present invention provides a sales performance attribution analysis system and method based on store retail terminals.

[0008] The sales performance attribution analysis system based on store retail terminals includes a data collection module, a data preprocessing module, an analysis module, an attribution model building module, and a result output and visualization module;

[0009] The data acquisition module includes a POS system, a CRM system and an inventory management system;

[0010] The data preprocessing module includes a data cleaning system, a data standardization processing system and a data classification and coding system;

[0011] The analysis module includes a member attribution model, a product attribution model, and a turnover attribution model;

[0012] The result output and visualization module includes a result output system and an interactive visualization system.

[0013] Preferably, the POS system collects transaction data, which includes the time of commodity sales, commodity sales price, commodity sales quantity and commodity payment method;

[0014] The CRM system collects customer data, including customer age, customer gender, customer purchase history and customer membership information;

[0015] The inventory management system collects commodity data and store operation data, wherein the commodity data includes commodity category, commodity brand, commodity cost and commodity inventory level, and the store operation data includes store opening hours, store promotion activity records and store layout changes.

[0016] Preferably, the data cleaning system is used to remove erroneous data, duplicate data and incomplete data;

[0017] The data standardization processing system is used to perform standardization processing on data, and the standardization processing includes unifying date formats and numerical units;

[0018] The data classification and coding system is used to classify and code data.

[0019] Preferably, the analysis process of the member attribution model is as follows:

[0020] Randomly initialize K cluster centers C k =(C kX , C kY , C kZ ), where k = 1, 2, ..., K, where the data points are customers, and each customer is represented by three dimensions: purchase frequency x, consumption amount y, and age level z;

[0021] For each customer i, its data point is represented by (X i , Y i , Zi ,), calculate the distance from the customer to each cluster center using the Euclidean distance formula:

[0022]

[0023] Assign customer i to the cluster with the nearest cluster center, that is, if Then customer i belongs to cluster j;

[0024] Recalculate the cluster center of each cluster using the formula:

[0025] Among them S k is the set of customers belonging to cluster k, n k is the number of customers in cluster k;

[0026] Repeat the above steps of assigning and updating cluster centers for 5-10 iterations.

[0027] Preferably, the product attribution model adopts association rules and performs analysis through support, confidence, lift and certainty:

[0028] Support is used to measure the frequency of a product combination in all transactions, and its formula is: Where Number(M∪N) represents the number of transactions involving both item M and item N, and TotalTransactions is the total number of transactions.

[0029] Confidence is used to measure the probability of purchasing product N when product M is purchased. The formula is:

[0030] Lift is used to evaluate whether the strength of the association between products in the association rules is meaningful. The formula is:

[0031] The formula for confidence is

[0032] Preferably, the turnover attribution model includes a store promotion effect analysis system, an employee performance analysis system and a store layout and display analysis system.

[0033] Preferably, the result output system is used to output the analysis results in a report; the report includes text descriptions and charts; the charts include bar charts, line charts and radar charts;

[0034] The interactive visualization system is used to provide an interactive visualization interface for users to view and analyze sales performance attribution results.

[0035] The sales performance attribution analysis method based on store retail terminals is used to implement the above-mentioned sales performance attribution analysis system based on store retail terminals, and includes the following steps:

[0036] Step S1. Collect comprehensive data from store retail terminals;

[0037] Step S2. Clean, standardize, classify and code the collected data;

[0038] Step S3. Analyze the customer dimension, product dimension, and store operation dimension respectively;

[0039] Step S4. Establish an attribution model based on the analysis results;

[0040] Step S5: Output the analysis results in an intuitive report format and provide an interactive visualization interface.

[0041] Preferably, the data collection in step S1 includes the following steps:

[0042] Step S101. Transaction data is collected through the POS system, including the time of sale, price, quantity, and payment method of the product.

[0043] Step S102. Collect customer data through the CRM system, including customer age, customer gender, customer purchase history, and customer membership information;

[0044] Step S103. Collecting store operation data through the inventory management system, the store operation data includes store opening hours, store promotion activity records and store layout changes;

[0045] The cleaning in step S2 is to remove erroneous data, duplicate data and incomplete data; the standardization in step S2 is to unify the date format and numerical unit;

[0046] The classification coding in step S2 is classification coding of commodity categories and customer types.

[0047] Preferably, the analysis of the product dimensions in step S3 includes the following steps:

[0048] Step S3021 analyzes the sales performance of products in different categories and calculates the sales share and gross profit contribution of each category;

[0049] Step S3022: Analyze the products, identify the best-selling and slow-selling products, find the associations between products through association rules, and provide suggestions for product display and combination sales;

[0050] Step S3023. Perform product price range analysis to determine sales trends and consumer price sensitivity for products in different price ranges;

[0051] The analysis of the store operation dimension in step S3 includes the following steps:

[0052] Step S3031. Evaluate the effectiveness of the promotional activity based on store operation data. By comparing sales data before and after the promotional activity, calculate the sales increase, sales volume increase, and return on investment of the promotional activity.

[0053] Step S3032: Analyze employee performance, calculate individual sales, average order value, and sales completion rate based on employee sales data, and evaluate employee contributions to sales performance;

[0054] Step S3033. Study the impact of changes in store layout and display methods on sales, and determine the optimal store layout and display plan by analyzing customer flow path data and product sales data within the store;

[0055] The attribution model in step S4 is established as follows:

[0056] A multiple regression model S=aW1+bW2+cW3 is established with sales performance S as the dependent variable, customer factor W1, product factor W2 and store factor W3 as independent variables;

[0057] According to the analysis results, the weight coefficients of customer factor W1, product factor W2 and store factor W3 are determined as a=0.45, b=0.3 and c=0.2.

[0058] Beneficial effects of the present invention:

[0059] 1. This invention provides comprehensive and in-depth sales performance attribution analysis for retail stores. By accurately quantifying the contribution of customer, product, and store operation factors to sales performance, store managers can more effectively allocate marketing resources and design personalized marketing strategies based on accurate data. They can also optimize product mixes, adjust pricing strategies, and improve display methods to increase product sales efficiency and gross profit contribution. Furthermore, in terms of store operations, the evaluation of factors such as promotional activities and employee performance helps to rationally plan promotions and motivate employees, thereby improving overall store operational efficiency.

[0060] 2. The data acquisition module provided by the present invention collects comprehensive data from multiple key data sources, ensuring that no important information is missed. The data preprocessing module ensures data quality and improves the accuracy and reliability of subsequent analysis. The multi-dimensional analysis step deeply explores the hidden information in different dimensions, helping to identify potential problems and opportunities.

[0061] 3. This system and method helps stores better adapt to the complex and ever-changing retail market environment. Compared with traditional simple analysis methods, it can more accurately reveal the drivers and obstacles of sales performance, thereby improving customer satisfaction and loyalty, and enhancing the store's competitiveness in the market;

[0062] 4. The result output and visualization module provided by this invention presents analysis results using intuitive charts and an interactive visual interface, allowing store managers to view and analyze sales performance attribution from different dimensions. This visualization significantly simplifies data understanding and improves decision-making efficiency. Without complex data processing skills, managers can quickly obtain key information, make timely strategic adjustments, and grasp market trends. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0064] Figure 1 This is a module diagram of the sales performance attribution analysis system based on store retail terminals in the present invention;

[0065] Figure 2 The figure is a flow chart of the sales performance attribution analysis method based on store retail terminals in the present invention.

[0066] Figure 3 It is a schematic block diagram of the analysis module in the present invention. In the figure: 1. Data acquisition module; 2. Data preprocessing module; 3. Analysis module; 4. Attribution model establishment module; 5. Result output and visualization module. DETAILED DESCRIPTION

[0067] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0068] like Figure 1-Figure 2 As shown, the sales performance attribution analysis system based on store retail terminals includes a data collection module 1, a data preprocessing module 2, an analysis module 3, an attribution model establishment module 4, and a result output and visualization module 5;

[0069] Data collection module 1 includes POS system, CRM system and inventory management system;

[0070] The POS system is the store sales terminal system, and the CRM system is the customer relationship management system;

[0071] The types of data collected by the POS system include transaction data, such as the time of product sale, product sales price, product sales quantity, and product payment method;

[0072] The types of data collected by the CRM system include customer data, such as customer age, customer gender, customer purchase history, and customer membership information;

[0073] The data collected by the inventory management system includes product data and store operation data;

[0074] Product data includes product category, product brand, product cost, and product inventory level;

[0075] Store operation data includes store opening hours, store promotion records, and store layout changes;

[0076] First, perform system initialization and data acquisition configuration:

[0077] Install and configure the system, set the connection parameters with the POS system, CRM system and inventory management system data sources, ensure that the data collection module can work properly, and determine the data scope and collection frequency to be collected. For example, configure the POS system interface to collect sales transaction data on a daily basis.

[0078] Data preprocessing module 2 includes data cleaning system, data standardization processing system and data classification and coding system;

[0079] The data cleaning system cleans the collected data to remove erroneous data, duplicate data, and incomplete data;

[0080] The data standardization processing system standardizes the data, such as unifying the date format and numerical units, to facilitate subsequent analysis;

[0081] The data classification and coding system classifies and codes data, such as product categories and customer types, to improve data processing efficiency;

[0082] Assume that some of the date formats in the collected transaction data are inconsistent. Use the date format conversion function in the data preprocessing module to unify them into the "YYYY-MM-DD" format.

[0083] For missing values ​​in customer data, such as missing age information for some customers, they can be estimated based on their purchasing behavior and other relevant information or marked as missing, and then specially handled in subsequent analysis;

[0084] Analysis module 3 includes member attribution model, product attribution model and turnover attribution model;

[0085] The membership attribution model is segmented based on customer data, using clustering algorithms to divide customers into different groups, such as purchase frequency, spending amount, and age level;

[0086] By analyzing the contribution of new and old customers to sales performance, calculate new customer acquisition rate, old customer retention rate and repurchase rate indicators, and evaluate their impact on sales and profits;

[0087] Preferably, the analysis process of the member attribution model is as follows:

[0088] Randomly initialize K cluster centers C k =(C kX , C kY , C kZ ), where k = 1, 2, ..., K, where the data points are customers, and each customer is represented by three dimensions: purchase frequency x, consumption amount y, and age level z;

[0089] For each customer i, its data point is represented by (X i , Y i , Z i ,), calculate the distance from the customer to each cluster center using the Euclidean distance formula:

[0090]

[0091] Assign customer i to the cluster with the nearest cluster center, that is, if Then customer i belongs to cluster j;

[0092] Recalculate the cluster center of each cluster using the formula:

[0093] Among them S k is the set of customers belonging to cluster k, n k is the number of customers in cluster k;

[0094] Repeat the above steps of assigning and updating cluster centers for 5-10 iterations to divide customers into different groups: high-consumption frequency and high-amount customers, high-consumption frequency and low-amount customers, low-consumption frequency and high-amount customers, and low-consumption frequency and low-amount customers. By analyzing the changes in purchasing behavior of different groups over a certain period of time, their impact on sales performance can be evaluated.

[0095] The product attribution model analyzes the sales performance of different product categories based on product data, calculating the sales share and gross profit contribution of each category; analyzes individual products to identify best-selling and slow-selling products, and uses association rules to find the correlation between products to provide suggestions for product display and combination sales; and conducts product price range analysis to determine the sales trends and consumer price sensitivity of products in different price ranges.

[0096] The sales ratio analysis formula is as follows:

[0097] Category sales share = (category sales ÷ total sales) × 100%. For example, if the sales of a clothing category are 50,000 yuan, and the total store sales are 200,000 yuan, then the sales share of the clothing category = (50,000 ÷ 200,000) × 100% = 25%. This formula can clearly understand the contribution of each category to the overall sales.

[0098] The formula for gross profit contribution analysis is as follows:

[0099] Gross profit contribution of a category = gross profit of the category ÷ total gross profit, where category gross profit = sales of the category - cost of the category. Assuming that the sales of the food category are 80,000 yuan and the cost is 50,000 yuan, then the gross profit of the food category is 30,000 yuan. If the total gross profit is 100,000 yuan, then the gross profit contribution of the food category = (30,000 ÷ 100,000) × 100% = 30%. This helps determine which categories are the main sources of profit;

[0100] The formula for analyzing the best-selling / slow-selling products is as follows:

[0101] Best-selling product ranking: Sort products by sales or volume. For example, products are sorted from high to low by sales volume. Products ranked high can be considered best-sellers to a certain extent, while products ranked low may be slow-moving products.

[0102] Sales growth rate: Product sales growth rate = [(Current period sales volume - Previous period sales volume) ÷ Previous period sales volume] × 100%. For example, if a mobile phone product sold 1,200 units this period and 1,000 units last period, its sales growth rate = [(1,200 - 1,000) ÷ 1,000] × 100% = 20%. Products with higher sales growth rates generally have better market performance and may be best-sellers. Conversely, products with negative and declining growth rates may be slow-moving products.

[0103] Association rules are analyzed by support, confidence, lift, and certainty:

[0104] Support is used to measure the frequency of a product combination in all transactions, and its formula is: Number(M∪N) represents the number of transactions that include both item M and item N, and TotalTransactions represents the total number of transactions. Assume there are 1000 transactions in total, of which 200 transactions include both item M and item N. Support reflects how frequently product M and product N are purchased at the same time;

[0105] Confidence is used to measure the probability of purchasing product N when product M is purchased. The formula is: For example, if there are 500 transactions containing product M and 200 of them also contain product N, then Confidence indicates the probability of purchasing product N when product M is purchased;

[0106] Lift is used to evaluate whether the strength of the association between products in the association rules is meaningful. The formula is: , assuming that the support of product N is Support(N) = 0.3, and the previously calculated Confidence(M→N) = 40% = 0.4, then A lift greater than 1 indicates that the presence of item M increases the probability of item N being purchased. A lift of 1 indicates that items M and N are independent of each other. A lift less than 1 indicates that the presence of item M decreases the probability of item N being purchased.

[0107] The formula for confidence is Confidence is another indicator to measure the strength of association rules. It indicates the probability that product N will not be purchased if there is no association rule M→N. For example, if Support(N)=0.4,Confidence(M→N)=0.7, then The higher the confidence, the more reliable the association rules. These formulas can help retailers determine the associations between products, thereby optimizing product display and combination sales strategies.

[0108] The formula for commodity price band analysis is as follows:

[0109] Sales volume by price range:

[0110] Sales volume percentage of a certain price range = (sales volume of the price range ÷ total sales volume) × 100%

[0111] For example, consider product prices divided into price ranges of 0-100 yuan, 101-200 yuan, and so on. If sales in the 0-100 yuan price range are 30,000 yuan, and total sales are 100,000 yuan, then the sales percentage of this price range = (30,000 ÷ 100,000) × 100% = 30%. This formula can be used to understand the sales performance of products in different price ranges.

[0112] Price elasticity coefficient, used to measure consumer price sensitivity:

[0113] Price elasticity coefficient Percentage change in sales volume Percentage change in price

[0114] For example, if the price of a product drops by 10% and sales increase by 20%, then the price elasticity coefficient = 20% ÷ (-10%) = -2. If the absolute value of the price elasticity coefficient is greater than 1, it means that consumers are price sensitive; if the absolute value is less than 1, it means that consumers are not price sensitive.

[0115] The result output and visualization module 5 includes a result output system and an interactive visualization system;

[0116] The result output system generates daily or weekly sales performance attribution analysis reports, outputting the analysis results in an intuitive report format, including text descriptions and charts;

[0117] Charts include bar charts, line charts, and radar charts;

[0118] The bar chart shows sales of different categories;

[0119] A line chart shows the trend of sales performance over time;

[0120] The radar chart shows the contribution of each factor to performance;

[0121] The interactive visualization system provides an interactive visualization interface that allows users to view and deeply analyze sales performance attribution results according to different dimensions (such as time, store, category, etc.), providing decision support for store managers. At the same time, based on the attribution results, targeted marketing strategies, product management strategies, employee incentive strategies and store operation improvement strategies are formulated. For example, if it is found that the decline in sales of a certain category of goods is mainly due to unreasonable product display, the display location and method of the category can be adjusted. If it is found that insufficient new customers are acquired, market promotion efforts can be increased or new customer attraction strategies can be optimized.

[0122] The sales performance attribution analysis method based on store retail terminals includes the following steps:

[0123] Step S1. Collect comprehensive data from store retail terminals;

[0124] Step S2. Clean, standardize, classify and code the collected data;

[0125] Step S3. Analyze the customer dimension, product dimension, and store operation dimension respectively;

[0126] Step S4. Establish an attribution model based on the analysis results;

[0127] Step S5: Output the analysis results in an intuitive report format and provide an interactive visualization interface.

[0128] Collecting data in step S1 includes the following steps:

[0129] Step S101. Transaction data is collected through the POS system. The transaction data includes the time of commodity sales, commodity sales price, commodity sales quantity and commodity payment method;

[0130] Step S102. Collect customer data through the CRM system, including customer age, customer gender, customer purchase history, and customer membership information;

[0131] Step S103. Collect store operation data through the inventory management system, including store operating hours, store promotion activity records and store layout changes;

[0132] In step S2, cleaning is to remove erroneous data, duplicate data and incomplete data.

[0133] In step S2, the date format and numerical unit are standardized;

[0134] The classification coding in step S2 is to classify and code the commodity categories and customer types.

[0135] The analysis of the product dimension in step S3 includes the following steps:

[0136] Step S3021 analyzes the sales performance of products in different categories and calculates the sales share and gross profit contribution of each category;

[0137] Step S3022: Analyze the products, identify the best-selling and slow-selling products, find the associations between products through association rules, and provide suggestions for product display and combination sales;

[0138] Step S3023. Perform product price range analysis to determine sales trends and consumer price sensitivity for products in different price ranges;

[0139] The analysis of the store operation dimension in step S3 includes the following steps:

[0140] Step S3031. Evaluate the effectiveness of the promotional activity based on store operation data. By comparing sales data before and after the promotional activity, calculate the sales increase, sales volume increase, and return on investment of the promotional activity.

[0141] Step S3032: Analyze employee performance, calculate individual sales, average order value, and sales completion rate based on employee sales data, and evaluate employee contributions to sales performance;

[0142] Step S3033. Study the impact of changes in store layout and display methods on sales, and determine the optimal store layout and display plan by analyzing customer flow path data and product sales data within the store;

[0143] The attribution model in step S4 is established as follows:

[0144] A multiple regression model S=aW1+bW2+cW3 is established with sales performance S as the dependent variable, customer factor W1, product factor W2 and store factor W3 as independent variables;

[0145] According to the analysis results, the weight coefficients of customer factor W1, product factor W2 and store factor W3 are determined as a=0.45, b=0.3 and c=0.2.

[0146] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Within the scope of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0147] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. The sales performance attribution analysis system based on store retail terminals is characterized by: It includes data collection module, data preprocessing module, analysis module, attribution model building module and result output and visualization module; The data acquisition module includes a POS system, a CRM system and an inventory management system; The data preprocessing module includes a data cleaning system, a data standardization processing system and a data classification and coding system; The analysis module includes a member attribution model, a product attribution model and a turnover attribution model; The result output and visualization module includes a result output system and an interactive visualization system.

2. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The POS system collects transaction data, which includes commodity sales time, commodity sales price, commodity sales quantity and commodity payment method; The CRM system collects customer data, including customer age, customer gender, customer purchase history and customer membership information; The inventory management system collects commodity data and store operation data, wherein the commodity data includes commodity category, commodity brand, commodity cost and commodity inventory level, and the store operation data includes store business hours, store promotion activity records and store layout changes.

3. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The data cleaning system is used to remove erroneous data, duplicate data and incomplete data; The data standardization processing system is used to perform standardization processing on data, and the standardization processing includes unifying date formats and numerical units; The data classification and coding system is used to classify and code data.

4. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The analysis process of the member attribution model is as follows: Randomly initialize K cluster centers C k =(C kX , C kY , C kZ ), where k = 1, 2, ..., K, where the data points are customers, and each customer is represented by three dimensions: purchase frequency x, consumption amount y, and age level z; For each customer i, its data point is represented by (X i , Y i , Z i ,), calculate the distance from the customer to each cluster center using the Euclidean distance formula: Assign customer i to the cluster to which the nearest cluster center belongs, and recalculate the cluster center of each cluster. The formula is: Where S k is the set of customers belonging to cluster k, n k is the number of customers in cluster k; Repeat the above steps of assigning and updating cluster centers for 5-10 iterations.

5. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The product attribution model adopts association rules and performs analysis through support, confidence, lift and certainty: Support is used to measure the frequency of a product combination in all transactions. The formula is: Where Number(M∪N) represents the number of transactions involving both item M and item N, and TotalTransactions is the total number of transactions; Confidence is used to measure the probability of purchasing product N when product M is purchased. The formula is: Lift is used to evaluate whether the strength of association between commodities in association rules is meaningful. The formula is: The formula for confidence is 6. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The turnover attribution model includes a store promotion effect analysis system, an employee performance analysis system and a store layout and display analysis system.

7. The sales performance attribution analysis system based on store retail terminals according to claim 1 is characterized in that: The result output system is used to output the analysis results in the form of a report; the report includes text descriptions and charts; the charts include bar charts, line charts and radar charts; The interactive visualization system is used to provide an interactive visualization interface for users to view and analyze sales performance attribution results.

8. The sales performance attribution analysis method based on store retail terminals is characterized by: It is used to execute the sales performance attribution analysis system based on store retail terminals as described in any one of claims 1 to 7, comprising the following steps: Step S1. Collect comprehensive data from store retail terminals; Step S2. Clean, standardize and classify the collected data; Step S3. Analyze in customer dimension, product dimension and store operation dimension respectively; Step S4. Establish an attribution model based on the analysis results; Step S5: Output the analysis results in an intuitive report format and provide an interactive visualization interface.

9. The sales performance attribution analysis method based on store retail terminals according to claim 8 is characterized in that: The data collection in step S1 includes the following steps: Step S101. Transaction data is collected through the POS system, wherein the transaction data includes the time of commodity sales, commodity sales price, commodity sales quantity and commodity payment method; Step S102. Collect customer data through the CRM system, the customer data including customer age, customer gender, customer purchase history and customer membership information; Step S103. Collecting store operation data through the inventory management system, the store operation data includes store business hours, store promotion activity records and store layout changes; The cleaning in step S2 is to remove erroneous data, duplicate data and incomplete data; the standardization in step S2 is to unify the date format and numerical unit; The classification coding in step S2 is classification coding of commodity categories and customer types.

10. The sales performance attribution analysis method based on store retail terminals according to claim 8, characterized in that: The analysis of the product dimension in step S3 includes the following steps: Step S3021 analyzes the sales performance of commodities of different categories and calculates the sales share and gross profit contribution of each category; Step S3022. Analyze the items, identify the best-selling items and slow-selling items, find out the association between the items through association rules, and provide suggestions for product display and combination sales; Step S3023. Perform commodity price range analysis to determine sales trends of commodities in different price ranges and consumer price sensitivity; The analysis of the store operation dimension in step S3 includes the following steps: Step S3031. Evaluate the effectiveness of the promotional activities based on the store operation data, and calculate the sales increment, sales volume increment and return on investment of the promotional activities by comparing the sales data before and after the promotional activities; Step S3032. Analyze employee performance, calculate individual sales, average order value, and sales achievement rate based on the employee's sales data, and evaluate the employee's contribution to sales performance; Step S3033. Study the impact of changes in store layout and display methods on sales, and determine the best store layout and display plan by analyzing customer flow path data and product sales data in the store; The attribution model in step S4 is established as follows: A multiple regression model S = aW1+bW2+cW3 is established with sales performance S as the dependent variable, customer factor W1, product factor W2 and store factor W3 as independent variables; According to the analysis results, the weight coefficients of customer factor W1, product factor W2 and store factor W3 are determined as a=0.45, b=0.3 and c=0.2.

Images