JP2002140502A - Customer data analysis method for customer retention and development - Google Patents

Abstract

(57) [Summary] [Issues] It is important for retailers to maintain important customers who purchase a large amount of money or to increase the number of customers who purchase large quantities.
To make that marketing approach efficient,
Provided is a method for analyzing customer data using a database of purchase data and the like. SOLUTION: A customer purchase history database is constructed,
Among them, not only the item of the purchase amount but also the annual average number of purchases, the number of years of purchase, the average purchase discount rate, the number of sales floors, the average number of days until the next purchase, and the like are included in the items, and the degree of contribution is evaluated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION When a customer purchases a product at a store or the like, the customer collects the purchase history database collected through a membership card or the like, a point seal to which the applied service point is added, and the like. The present invention relates to a customer data analysis method in which a database such as a customer application history database or a questionnaire response database is constructed in a computer system, and a customer having high importance or expected to have high importance is found.

[0002]

2. Description of the Related Art Conventionally, as a method of maintaining a customer for continuous purchase or mass purchase, a purchase point is provided when a customer purchases a product, and a prize is provided or a product is provided in accordance with the collected points. Discounts and other special services. Keep your customers for a long time,
Further, as a specific method for increasing the number of mass purchasers, there is a method in which the greater the number of points added when purchasing a product, the more the benefits are obtained.

[0003] In addition, the approach to direct mail or the like is used only for customers using purchase points. In addition, the customer is retained and increased after carefully selecting customers. The method for was not specifically considered.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has as its object to increase the degree of customer retention at a store or the like and to increase the number of continuous purchasers and mass purchasers. Another object of the present invention is to provide a method of analyzing customer data for extracting customers expected to become important customers as mentioned above in the future. For more information,
When a customer purchases a product at a store or the like, the customer's purchase history database collected via a membership card,
Build a database such as an application history database or a questionnaire response database collected via a point seal etc. to which the applied service point is added,
It is an object of the present invention to provide an analysis method for extracting customers having high importance or expected to have high importance.

[0005]

In order to solve the above-mentioned problems, a method for analyzing customer data for maintaining and cultivating customers according to the present invention comprises the steps of: The member data and the purchased merchandise data are read by the terminal using the member card of the store in which the data is registered, and the data including the customer code, the merchandise code, the purchase price, the purchase quantity, and the purchase date and time are transmitted via the communication line. A customer data analysis method for calculating the order of importance of important customers to the store by using a customer purchase history database collected and constructed at predetermined time intervals and transmitted to a computer system at a center. The purchase history database is searched for each predetermined period, and the data of the purchase amount item indicating the direct sales contribution within the predetermined period. And a purchase amount extraction table generating means for extracting and totaling for each customer to generate a table, and extracting and totaling data of purchase frequency items indicating a high purchase frequency which is a factor of future sales increase and generating a table. Purchase number extraction table and average purchase amount extraction table generation means for each purchase, and data for purchase duration items indicating the length of the purchase duration period for which future sales can also be expected are extracted and aggregated for each customer to generate a purchase table. Continuation period extraction table generation means and data of average purchase discount rate items that show low reactivity of discounts to purchase at regular prices as well as during the sale period and that can be expected to purchase high-priced products are extracted and aggregated for each customer Means for generating an average purchase discount rate extraction table that generates a purchase table, and the next purchase that shows the high level of purchase planning that is regularly purchased and can be expected to sell in the future Next-day average purchase date extraction table generation means for extracting and tabulating the data of the average number of days item for each customer and generating a table, and purchased products including the possibility of purchasing a large variety of products and becoming important for the store concerned Means for extracting and summarizing the data of the number of purchase counters item indicating the size of the range of each customer for each customer and generating a table. The system automatically generates the table at the end of each predetermined period. A first step of generating a table by extracting from the data of the seven items by each program means of the means, and the system contributes to the increase in sales of the store for each table of the generated data extraction of the seven items. Customer ranking means for rearranging the customers in the order of high or high possibility of becoming high in the future and generating the rearrangement table of the seven item data, and the first step is completed. Then, a second step of generating the permuted tables obtained by rearranging the customers by the customer ranking means using the generated extracted tables, and the system comprises: In each table, the highest to lowest ranks of the customer rankings are classified into a plurality of levels n, and the evaluation value of the highest level is set to n, and the evaluation value of the lowest level is set to l. A customer evaluation unit that calculates each table and generates an evaluation table representing an evaluation value for each customer for each of the tables. When the second step is completed, based on the reordering tables generated therefrom, A third step of generating an evaluation table for each of the seven items by the customer evaluation means, and the system further comprises: a total sum for each customer for which the evaluation values of the evaluation tables for each of the seven items are summed for each customer Value table Generating means, total rank table generating means for arranging the customer rank in the order of the total value based on the total result of each customer of the total total value table, and classifying the rank in the total rank table into m hierarchies, A total evaluation table generating means for each customer for calculating an m-level evaluation value with m being the highest hierarchy, and when the third step is completed, a total total value table and a total evaluation value table are created based on the evaluation tables generated there. A fourth step in which the generating means automatically generates a total total value table for each customer within the predetermined period; and a fifth step in which the total total value table means generates a total ranking table for each customer.
And a sixth step of generating a comprehensive evaluation table for each customer by the comprehensive evaluation table generating means. The list of important customer layers including not only the current period but also the future is stored in the comprehensive evaluation table. It can be created and displayed on the system display unit.

In the fourth step, the total sum table generating means for summing the evaluation values of the respective evaluation tables for each of the seven items for each customer, before the sum, the purchase price, the average number of purchases for a predetermined period, Average purchase per transaction,
The evaluation value of each item of the average purchase discount rate is multiplied by a weighting coefficient, and then the respective evaluation values are summed.

In the first step, the purchase price extraction table generating means calculates an average value of the purchase price of the customer during the predetermined period, and extracts the data of the seven items from only the customers having the average value or more. Each table is generated by using

[0008] In the first step, the purchased items to be extracted from the purchase history database are extracted from the data of the seven items with respect to the items of only the specific maker selected from the item code to generate each table. It is characterized by.

[0009] In the first step, the purchase items to be extracted from the purchase history database are extracted from the data of the seven items with respect to only specific items selected from the item code, and each table is generated. And

[0010] Further, the purchase history database further includes electronic transaction purchase data and data in which at least an address and a name are entered for a customer who conducts an electronic transaction using a homepage on the Internet, and the purchase history is stored in the first step. The purchased merchandise extracted from the database is characterized in that the purchased merchandise extracted from the electronic transaction database also includes electronic transaction data.

[0011] Further, for the purpose of having the customer collect in advance, the service points are entered according to the purchase price and the point sticker directly attached to the purchased product or, if there is no point sticker, the product barcode is cut out to provide the customer with the service product. At the time of application to collect the specified quantity in response to, etc., the customer who fills in at least the address and name of the customer, collects it at the dealer, sends the customer data to the computer system at the center, and collects and builds it every predetermined period A customer data analysis method for calculating an order of importance of a customer important to a store by using an application history database, wherein the computer system searches the application history database for each predetermined period, and directly searches the application history database within a predetermined period. Extract and summarize the data of quantity items of application stickers or product barcodes indicating the degree of sales contribution for each customer. Means for generating an application quantity extraction table for generating a table, means for generating an annual application frequency extraction table for indicating the frequency of application that will increase sales in the future, and application continuation for indicating the length of the application continuation period in which future sales can be expected An application continuation period extraction table generating means for extracting and totaling data of period items for each customer and generating a table is provided, and the system automatically executes the data of the three items by each of the three program means at the end of a predetermined period. And generating the respective tables, and the system further comprises, for each of the generated three-item data extraction tables, the customer in the order in which the sales increase of the store increases or increases in the future. A customer ranking means for generating a rearranged table of the rearranged three-item data; when the first step is completed, the extracted A second step of generating a rearranged table in which the tables are rearranged by the customer ranking means, and the system further comprises: Are classified into a plurality of layers, and the evaluation value of the highest hierarchy is set to n, and the evaluation value of the lowest hierarchy is set to l. Is provided, and when the second step is completed, the customer evaluation unit generates each evaluation table for each of the three items based on the generated three-item data rearrangement table. A third step of generating an evaluation table, the system further comprises: a total total value table generating means for each customer for summing the evaluation values of each evaluation table for each of the three items for each customer; Total rank table generating means for arranging the customer ranks in the order of the total value based on the total result of each customer of the table, and classifying the ranks in the total rank table into a plurality of m levels, where m is the highest rank. A total evaluation table generating means for each customer for calculating a graded evaluation value, and when the third step is completed, the total sum table generating means automatically generates the predetermined value based on each evaluation table generated there. A fourth step of generating a total total value table for each customer within the period; and a fifth step of generating a total ranking table for each customer by the total total value table means.
And a sixth step of generating a comprehensive evaluation table for each customer by the comprehensive evaluation table generating means. The important customer layer list including not only the current period but also the future is stored in the comprehensive evaluation table from the comprehensive evaluation table. It can be created and displayed on the system display unit.

When a customer purchases a product at a store,
The member data and the purchased merchandise data are read by a terminal using a member card of the store in which at least the address and name data of the customer are registered in advance, and the customer code, product code, purchase amount, purchase quantity, purchase date and time are read. Is transmitted to the computer system of the center via a communication line, and the customer purchase history database collected and constructed at predetermined time intervals is used to extract the purchase price of the customer of the surveyed product, and the customer within the predetermined time period is At the time of purchase at the store, at least the number of family members, the age of children, and questionnaire response data on the frequency of use of other stores are sent to the center's computer system via a communication line for customers whose purchase amount is equal to or less than the average value of the purchase amount. The questionnaire response database that has been sent and collected at the specified time intervals has potential for growth. A customer data analysis method for calculating an order of importance of a customer, wherein the computer system searches the questionnaire response database for each of the predetermined periods, and extracts data on the number of family members of the customer within the predetermined period for each customer. A family number extraction table generating means for totaling and generating a table, a child age extracting table generating means for extracting and totaling data of customer's child's age items for each customer, and generating a table, and other store use frequency item data. Means for extracting and totaling tables for each customer to generate a table, and a table for generating frequency of use of other stores, wherein the system automatically executes the program for each of the three means at the end of the predetermined period.
A first step of extracting data from the items and generating each table; and the system is arranged such that for each of the generated data extraction tables, the order of the customers in the order that the contribution of the sales increase of the store is likely to increase in the future is high. Customer ordering means for generating each of the rearranged tables obtained by rearranging the tables. When the first step is completed, the extracted tables generated therefrom are sorted by the customer ordering means. A second step of generating a table, and
The system categorizes each of the reordering tables into n pieces of a plurality of hierarchies from the highest order to the lowest order of the customer rank, and calculates an n-level evaluation value where the evaluation value of the highest hierarchy is n for each reordering table. A customer evaluation means for generating an evaluation table representing an evaluation value for each customer for each of the tables;
Is completed, a third step of generating each evaluation table for each of the seven items by the customer evaluation means based on the rearrangement tables generated therein, and the system comprises the steps of: Means for generating a total sum value table for each customer for summing the values for each customer, and Means for generating a total rank table for arranging the customer ranks in descending order of the total value based on the summation result for each customer in the total sum table And a total evaluation table generating means for each customer that classifies the ranks in the total rank table into a plurality of m levels and calculates an m-level evaluation value with the highest rank being m. When the third step is completed, A fourth step in which the total sum table generating means automatically generates a total sum table for each customer within the predetermined period based on each of the generated evaluation tables. And a fifth step of generating a total ranking table for each customer by the total sum table means, and a sixth step of generating a total evaluation table for each customer by the total evaluation table generating means. And an important customer layer list that may be most important in the future can be created from the comprehensive evaluation table and displayed on the system display unit.

In addition, the number of service points is entered in advance in accordance with the purchase price for the purpose of having the customer collect, and at least the address and name of the customer are entered when applying for the point seal directly attached to the product of the specified manufacturer. Then, the data is transmitted to the computer system of the center via a communication line, collected at predetermined intervals, and a purchase amount corresponding to a customer's application sticker is extracted from the application history database constructed. For customers whose average purchase price is less than or equal to the average purchase amount at the time of purchase, the survey response data on the total purchase volume, purchase status of similar competitors' products, age, and the number of family members at least at the specified period at the time of purchase at the store The questionnaire response database transmitted to the computer system of the center via the A customer data analysis method for calculating the importance degree order of customers with potential growth potential by,
A total purchase amount table generating means for searching the questionnaire response database for each predetermined period, extracting and totaling data of total purchase amount for each customer within the predetermined period including month or year for each customer and generating a table; Means for generating a table by extracting and totaling data on the purchase amount of similar competitor products for each customer, and generating a customer age table for generating a table by extracting and totaling customer age data for each customer Means, and a family number table generating means for generating a table in which customer family number data is extracted and totaled for each customer, and the system is automatically set by the respective program means of the four means at each end of the predetermined period. A first step of extracting data from the items and generating each table; A customer ranking means for generating the rearranged tables in which the customers are rearranged in the order in which the contribution of the sales increase of the store is likely to increase in the future for each extracted table, and when the first step is completed, And generating a rearranged table in which the extracted tables generated there are rearranged by the customer ranking means.
In addition, the system classifies the n-level evaluation values in which the top-level to the lowest-priority in customer priority are classified into a plurality of layers n and the evaluation value of the highest level is n for each of the rearrangement tables. A customer evaluation means for calculating each replacement table and generating an evaluation table representing an evaluation value for each customer for each table. When the second step is completed, the reordering table generated there is used as a basis. A third step of generating each evaluation table by the customer evaluation means, and the system further comprises: a total total value table generation means for each customer for summing the evaluation values of the evaluation tables for each customer; A total rank table generating means for arranging the customer rank in the order of the total value based on the total result of each customer in the total total value table, and dividing the rank in the total rank table into m layers. And a comprehensive evaluation table generating means for each customer for calculating an m-level evaluation value with the highest hierarchy being m. When the third step is completed, an overall total value table is generated based on each evaluation table generated there. A fourth step in which the generating means automatically generates a total total value table for each customer within the predetermined period; and a fifth step in which the total total value table means generates a total ranking table for each customer. And a sixth step of generating a comprehensive evaluation table for each customer by the comprehensive evaluation table generating means. The important customer layer list which may be most important in the future is stored in the comprehensive evaluation table. It can be created and displayed on the system display unit.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows a flow chart of a method for analyzing customer data for customer maintenance and training according to a first embodiment of the present invention. FIG. 1 shows a flowchart of a customer data analysis 10 using a customer purchase history database.

Here, when the customer purchases a product at the specified store or affiliated store, the purchase history database presents a member card of the store in which data of at least the address and name of the customer is registered in advance, By means of a terminal that reads the member data and the data of the purchased product, data including a customer code, a product code, a purchase amount, a purchase quantity, and a purchase date and time is collected by a computer system via a communication line for a predetermined period of at least one year, It was built. FIG. 5 shows an example in which data items of the purchased merchandise in the purchase history database and some of their contents are displayed.
The contents are omitted and left blank.

The method will be described below with reference to the flowchart of FIG.

First, in step S11, the purchase amounts of all customers or customers are extracted from the purchase history database of the customers, the average value is calculated, and only the customers whose average value is equal to or more than the average value are extracted. However, FIG. 1 shows a case where the latter is selected. The same applies to FIG. 2 described later.

Next, in step S12, customer data is extracted for each of a plurality of items. As these items, in this embodiment, the purchase amount g during a predetermined period (at least one month or several years), the average number of purchases h per year, the average purchase amount i per purchase, the number of years of purchase j, the average purchase discount rate k,
The average number of days until the next purchase is 1 and the number of purchase counters is m.

These seven items are not only large in the amount of the purchase, but also the frequency of purchase, the length of the purchase continuation period, and the low reactivity of the discount at the normal price as well as during the sale period. The data related to these items is selected from the database, assuming that the range of the range of purchased products that purchases various types of products has a good influence on the profitability of the store.

Next, in step S13, for each item, the number, height,
Customer ranking is performed in order of length and breadth, that is, the order in which the contribution of sales increase is high.

Next, in step S14, the customer ranking for each item is classified into a plurality of layers, for example, five layers, and a five-level evaluation value is determined for each item for each customer.

Next, in step S15, the five-level evaluation value for each item is multiplied by a predetermined weighting coefficient according to the importance of the item. As the weighting factor, it is preferable to multiply the evaluation value by setting the purchase amount, the average purchase amount per one average annual number of purchases, and the average purchase discount rate to about 1.5 times. Although the weight coefficient is considered in this embodiment,
Of course, it is not necessary to consider the weight coefficient.

Next, in step S16, the evaluation values of the seven items are summed up for each customer to calculate a total evaluation value.

Next, in step S17, the total evaluation value for each customer is ranked in order of the highest evaluation value.

Next, in step S18, the order is classified into a plurality of layers 5, and evaluation values of five levels are given in order from the top.

Next, in step S19, the highest stage in the list is regarded as the most important customer, and a customer list is created. Analyzing in this way, you can find out which customers are most important and must be maintained at present.
As a result, it is possible to restrict the number of customers and approach them efficiently without approaching all customers, which leads to a reduction in marketing costs.

FIG. 6 shows an analysis method according to FIG.
For B, C, D, E, and 欄, a column for entering the respective 5-grade evaluation values for the above seven items (the description of the evaluation values is omitted in this figure), and a 5-grade evaluation value for that column , The total evaluation value is calculated, and based on the calculated total evaluation value, the total evaluation value of the five grades is shown in the final comprehensive column. Note that the comprehensive column is called a loyalty level in the figure, and the customer importance level is divided into five levels. High-level customers,
For example, it is efficient to approach from a group of level 5 customers and to make a difference in the content of the approach.

Also, in step S11 in FIG. 1, customer data may be analyzed based on the product code in the purchase history database and limited to products related to only a specific manufacturer. At this time, the level of importance of each customer regarding the product of the manufacturer can be found.

Further, in the step S11 in FIG. 1, if only a specific product is limited from the product code in the purchase history database, the level of importance of each customer regarding the product can be found.

FIG. 2 shows a flowchart of a customer data analysis method according to a second embodiment of the present invention. FIG. 2 is a flowchart of the customer data analysis 20 using the customer application history database. Here, in the application history database, the number of service points is entered according to the purchase price for the purpose of having the customer collect in advance, and the point sticker directly attached to the purchased product and the product barcode correspond to the customer's service item etc. At the time of application to collect the designated quantity, paste it on a postcard, etc., and mail it to a predetermined place, at least fill in the address and name of the customer and collect it for a predetermined period of time.

The method will be described below with reference to the flowchart of FIG.

First, in step S21, the purchase amount of all customers or customers is extracted from the application history database, the average value is calculated, and only the customers whose average value is equal to or more than the average value are extracted.

Next, in step S22, each data is extracted from three items of the number of application stickers, the number of application times per year, and the application continuation period for each customer within a predetermined period.

Next, in step S23, the order of the number and length of the items which are determined to have contributed for each item, that is,
Rank customers in the order of sales contribution.

Next, in step S24, the customer ranking is classified into five levels for each item, and a five-level evaluation value is determined for each item for each customer.

Next, in step S25, the evaluation values of the three items are summed up for each customer to calculate a total evaluation value.

Next, in step S26, the customers are ranked in descending order of the total evaluation value for each customer.

Next, in step S27, the order is classified into five levels, and a five-level evaluation value for each customer is determined.

Finally, a customer layer list in order of importance is created. In this way, you can find the customers that are most important to your dealer and must be maintained. As a result, it is possible to limit the number of customers and approach them efficiently without approaching all customers, thereby reducing marketing costs.

FIG. 7 shows an analysis method according to FIG.
For b, cc, dd, ee, and 欄, a column for entering a 5-level evaluation value for each of the above three items (the evaluation value is omitted in FIG.
The total evaluation value is calculated from the grade evaluation value, and the five-grade evaluation value obtained by totaling based on the calculated value is shown in the royalty level column.

FIG. 3 shows a flowchart of a customer data analysis method according to a third embodiment of the present invention. FIG. 3 shows a flowchart of customer data analysis 30 using a purchase history database and a questionnaire response database at the time of purchase. Here, the questionnaire response database at the time of purchase is constructed by collecting at least the number of family members, the age of children, and the frequency of use of other stores in the computer system for a predetermined period at the time of purchase at the store. This data analysis is a customer data analysis method that uses the survey response database to find important customers who have potential growth potential for retailers from customers with a small purchase price based on purchase data and calculates the order of importance of customers. is there.

First, in step S31, customers whose purchase amount is equal to or less than the average purchase amount of all customers are extracted from the purchase history database.

Next, at step S32, the number of families, the age of children,
The three items of the frequency of use at another store are extracted.

Next, in step S33, the customers are ranked in order of the number of items contributing to the sales, that is, the order expected to contribute to the sales increase for each item.

Next, in step S34, the customer ranking is classified into five layers for each item, and a five-level evaluation value of each item is determined for each customer.

Next, in step S35, the evaluation values of the three items for each customer are added up to calculate a total evaluation value.

Next, in step S36, the ranking of the customers is ranked in descending order of the total evaluation value for each customer.

Next, in step S37, the order is classified into five levels, and a five-level evaluation value for each customer is determined.

Finally, in step S38, a customer group having potential growth room can be found, and a list thereof can be created. Therefore, unlike the customer who is important at present as in the data analysis of the customer in FIGS. 1 and 2, it is possible to perform a sophisticated marketing approach suitable for the important customer in the future. FIG. 8 shows an analysis method according to FIG. 3 for customer V, W, X, Y, Z,.
A total evaluation value is calculated from the five-level evaluation values for each of the columns (3) and the five-level evaluation values in these columns (the description of the evaluation values is omitted in this figure), and based on the values. , The integrated five-level evaluation value is shown in the last column. This column is called a potential level in the sense that it has potential growth potential. The analysis of the potential level shown in FIGS. 3 and 8 is particularly effective in the following cases. In general food-related products and food stores, customers who answered that “the number of family members is large (for example, 5 or more)” despite the small purchase amount are “the number of family members is small (for example, 2)”. Is more likely to purchase another product or use a different store than the customer who answered "No."

FIG. 4 shows a flowchart of a customer data analysis method according to a fourth embodiment of the present invention. FIG. 4 is a flowchart of the customer data analysis 40 using the application history database and the questionnaire response database at the time of application. Here, the questionnaire response database at the time of application is constructed by collecting questionnaire responses of at least a monthly or annual total purchase amount, purchase status of similar competitor products, age, and the number of families for a predetermined period. This data analysis is based on this questionnaire response database to find important customers who have potential growth potential from the customers who have a small number of application seals, that is, the purchase amount with the purchase data, and sort the importance order of the customers. This is a customer data analysis method to be calculated.

First, in step S41, customers whose purchase amounts are equal to or less than the average value of the purchase amounts of all customers are extracted from the application history database, and the data is further limited to the products of the manufacturer specified from the extracted customers.

Next, in step S42, data is extracted from the questionnaire response database from the four items of the total purchase amount per month or year, the purchase amount of similar competitor's products, the age, and the number of families.

Next, in step S43, the customers are ranked in the order in which they contribute to the sales for each item, that is, the order in which the contribution of the sales increase is expected to be high.

Next, in step S44, the customer ranking is classified into five layers for each item, and a five-level evaluation value of each item is determined for each customer.

Next, in step S45, the evaluation values of the four items are summed up for each customer to calculate a total evaluation value.

Next, in step S46, the customers are ranked in descending order of the total evaluation value for each customer.

Next, in step S47, the order is classified into five levels, and a five-level evaluation value for each customer is determined.

Next, in step S48, a list of customer classes having potential growth potential can be created. Therefore, unlike the currently important customers as shown in the data analysis of FIG. 1 and FIG. 2, it is possible to perform a sophisticated marketing approach suitable for the customers. FIG. 9 shows an analysis method according to FIG. 4 in which a column is provided in which five-level evaluation values for at least four items are written as questionnaire response data at the time of application for customers A, B, C, D, E, and （ The values are omitted), and a total evaluation value is calculated from the five-level evaluation values in those columns, and a total of five-level evaluation values based on the calculated values are shown in the potential level column. The analysis of the potential levels shown in FIGS. 4 and 9 has the following effects. Although the purchase price of the target product in the specified period, that is, the number of application stickers, is small, the questionnaire to answer the purchase price and purchase amount of similar products including the target product already knows the purchase price and purchase amount. If the purchase amount or purchase amount is significantly higher than the target product, it is likely that the customer is likely to purchase other similar products frequently, and the target product will Then, the customer whose profit is high, that is, the customer with a high potential level can be clearly identified.

[0059]

The customer data analysis method for retaining and cultivating customers according to the present invention has the following effects. (1) From the purchase history database, this data analysis method can be used to find out the most important customer groups that must be maintained for dealers, manufacturers and products at present and to take an efficient approach such as limiting them to those customer groups. it can.

(2) From the application history database, the present data analysis method is used to find the most important customer group that needs to be maintained at present as described above, and to take an efficient approach focusing on that customer group. Can be.

(3) By using the purchase history database and the questionnaire response database based on the questionnaire response at the time of purchase, a customer group having potential growth potential, which is likely to become the most important customer group in the future, is found. It is possible to take an appropriate and efficient approach to

(4) From the application history database and the questionnaire response database based on the questionnaire response at the time of application, similarly to the above, a customer group having potential growth potential, which may become the most important customer group in the future, is found. It will be possible to take an appropriate and efficient approach to that customer segment.

[Brief description of the drawings]

FIG. 1 is a flowchart of a first embodiment of customer data analysis using a purchase history database.

FIG. 2 is a flowchart of a second embodiment of customer data analysis using an application history database.

FIG. 3 is a flowchart of a third embodiment of customer data analysis using a purchase history database and a purchase questionnaire response database.

FIG. 4 is a flowchart of a fourth embodiment of customer data analysis using an application history database and an application questionnaire response database.

FIG. 5 is a diagram showing items and the like as an example of product data in a purchase history database.

FIG. 6 is a diagram showing an evaluation value list of important customer classes in the first embodiment.

FIG. 7 is a diagram showing an evaluation value list of important customer classes in the second embodiment.

FIG. 8 is a diagram showing an evaluation value list of a customer group having potential growth potential in the third embodiment.

FIG. 9 is a diagram showing an evaluation value list of a customer group having potential growth room in the fourth embodiment.

[Explanation of symbols]

 g Purchase amount h Average number of purchases per year i Average purchase amount per purchase j Years of purchase k Average purchase discount rate l Average days until next purchase m Number of purchase counters 10 Flowchart of the first embodiment 20 Flowchart of the second embodiment 30 Flowchart of Third Embodiment 40 Flowchart of Fourth Embodiment

Claims (9)

[Claims] When a customer purchases a product at a store, the terminal reads the member data and the purchased product data with a member card of the store in which at least the address and name data of the customer are registered in advance. , Its customer code,
Data including the product code, purchase price, purchase quantity, and purchase date and time is transmitted to the computer system of the center via a communication line, collected at predetermined time intervals, and constructed by a customer purchase history database. A customer data analysis method for calculating an order of importance, wherein the computer system searches the purchase history database for each predetermined period, and purchase price item data indicating a direct sales contribution within the predetermined period and a customer. A purchase amount extraction table generation means for extracting and totaling each time and generating a table, and a purchase frequency for extracting and totaling data of purchase frequency items indicating a high purchase frequency which is a factor of future sales increase for each customer and generating a table The extraction table and the average purchase amount extraction table generation means per time, and the length of the purchase continuation period in which future sales can be expected. A purchase duration extraction table generating means for extracting and summarizing the data of the purchase duration item shown for each customer and generating a table, and showing low and high discount reactivity for purchase not only during the low price period but also at the normal price. Shows the average purchase discount rate extraction table generation means that extracts and aggregates data of average purchase discount rate items that can be expected to purchase price products for each customer and generates a table, and shows the high level of purchase planning that always purchases regularly. A next-purchase-average-days-extraction-table generating means for extracting and totaling data of the average number of days until the next purchase at which future sales can be expected for each customer and generating a table;
The number of purchase sales floors where the data of the purchase sales floor items that indicate the breadth of the range of purchased products that include the possibility of purchasing various types of products and becoming important to the store concerned are extracted for each customer, totaled, and a table is generated. A first step in which the system automatically extracts and generates a table from the seven items of data by each of the seven program means at the end of each predetermined period; and Is a customer that rearranges the customers in the order in which the contribution of the sales increase of the store is high or likely to increase in the future for each of the generated seven-item data extraction tables. When the first step is completed, when the first step is completed, the extracted tables generated by the first step are sorted by the customer ranking means into the rearranged tables obtained by rearranging the customers. A second step of generating a table, the system classifies the customer order from the highest rank to the lowest rank into a plurality of layers n for each of the reordering tables, and evaluates the evaluation value of the highest rank. a customer evaluation means for calculating an n-level evaluation value, where n is the evaluation value of the lowest hierarchy, and 1 for each of the rearranged tables, and generates an evaluation table representing an evaluation value for each customer for each of the tables. , When the second step is completed, a third step of generating each evaluation table for each of the seven items by the customer evaluation means based on the rearrangement tables generated therein, and the system comprises: Means for generating a total sum value table for each customer for summing the evaluation values of each evaluation table for each of the seven items for each customer, and an array of customer ranks in the order of the total value based on the summation result for each customer in the total sum value table Gross A ranking table generating means, and a comprehensive evaluation table generating means for each customer, which classifies the ranks in the overall ranking table into a plurality of m levels and calculates an m-level evaluation value with the highest rank being m. After 3 steps,
A fourth step of automatically generating a total total value table for each customer within the predetermined period by a total total value table and generating means based on each of the generated evaluation tables; And a sixth step of generating a comprehensive evaluation table for each customer by the comprehensive evaluation table generating means. The fifth step is to generate a comprehensive ranking table for each customer. A customer data analysis method for maintaining and cultivating customers, wherein an important customer layer list including the above can be created from the comprehensive evaluation table and displayed on a system display unit. 2. A total sum table generating means for summing the evaluation values of each evaluation table for each of the seven items in the fourth step for each customer, wherein the total purchase price is calculated before the summation.
The evaluation value according to each item of the average number of purchases, the average purchase amount per one time, and the average purchase discount rate for a predetermined period is multiplied by a weighting coefficient, and then the respective evaluation values are summed up. Customer data analysis method for customer retention and training. 3. The purchase price extraction table generating means of the first step calculates an average value of purchase prices of the customers in the predetermined period, and extracts the data of the seven items from only the customers having the average value or more. The customer data analysis method for customer maintenance and training according to claim 1 or 2, wherein each table is generated by using the method. 4. In the first step, the purchased items to be extracted from the purchase history database are extracted from the data of the seven items with respect to the items of only the specific manufacturer selected from the item code, and each table is generated. The customer data analysis method for customer maintenance and training according to claim 1, 2, or 3, characterized in that: 5. A purchase product to be extracted from the purchase history database in the first step, wherein only specific products selected from the product code are extracted from the data of the seven items to generate each table. The customer data analysis method for customer maintenance and training according to claim 1, 2 or 3. 6. The purchase history database further includes electronic transaction purchase data and data in which at least an address and a name of a customer who conducts an electronic transaction using an Internet homepage are added. The purchased product extracted from the database includes electronic transaction data included in the purchased product extracted from the electronic transaction database.
2. The customer data analysis method for customer maintenance and training described in 2, 3, 4 or 5. 7. A point seal that is filled in with service points in accordance with the purchase price for the purpose of having the customer collect in advance, or a point sticker directly attached to the purchased product or a product bar code cut out when there is no point seal, and the service product is provided to the customer. At the time of application to collect the specified quantity in response to, etc., the customer who fills in at least the address and name of the customer, collects it at the dealer, sends the customer data to the computer system at the center, and collects and builds it every predetermined period A customer data analysis method for calculating an order of importance of a customer important to a store by using an application history database, wherein the computer system searches the application history database for each predetermined period, and directly searches within the predetermined period. Extract and tabulate data on quantity items of application stickers or product barcodes indicating sales contributions for each customer. Means for generating an application quantity extraction table, which generates an application frequency, an annual application frequency extraction table generation means for indicating the frequency of application frequency which will be a factor for increasing future sales, and an application continuation period indicating the length of the application continuation period in which future sales can be expected An application continuation period extraction table generation means for extracting and totaling item data for each customer and generating a table is provided, and the system automatically generates the table at the end of each predetermined period.
A first step of extracting the three items of data by means of each program means and generating each table; and a system for increasing the sales of the store for each of the generated three items of data extraction table. Customer ordering means for generating a rearranged table of the three item data in which the customers are rearranged in the order of higher contribution or higher in the future, and when the first step is completed, the extracted respective tables generated there are replaced. A second step of generating sorting tables in which the customers are sorted by the customer ranking means, respectively, and the system further comprises: Classification into n layers, an evaluation value of the highest hierarchy being n, an evaluation value of the lowest hierarchy being 1 and an n-stage evaluation value being calculated for each rearrangement table,
Customer evaluation means for generating an evaluation table representing an evaluation value for each customer for each of the tables, and when the second step is completed, the customer evaluation is performed based on the three item data rearrangement tables generated there. A third step of generating each of the evaluation tables for each of the three items by the means; and the system generates a total total value table for each of the customers for summing the evaluation values with the evaluation tables for each of the three items for each of the customers. Means, total rank table generating means for arranging the customer rank in the order of the total value based on the sum total of each customer in the total sum table, and classifying the rank in the total rank table into a plurality of m levels, A total evaluation table generating means for each customer for calculating an m-level evaluation value with the hierarchy being m, and when the third step is completed, each evaluation table generated there is A fourth step of automatically generating a total total value table for each customer within the predetermined period by the total total value table generating means; and a total ranking table for each customer by the total total value table means. A fifth step of generating
A sixth step of generating a comprehensive evaluation table for each customer by the comprehensive evaluation table generating means, creating an important customer layer list from the comprehensive evaluation table not only in the present period but also in the future; A customer data analysis method for customer maintenance and training, characterized in that the data can be displayed on the website. 8. When a customer purchases a product at a store, the terminal reads the member data and the purchased product data with a member card of the store in which at least the address and name data of the customer are registered in advance. , Its customer code,
Data consisting of the product code, purchase price, purchase quantity, and purchase date and time is transmitted to the computer system of the center via a communication line, collected at regular intervals, and built by the customer purchase history database. At the time of purchase at the store, at least the number of family members, the age of children, and the data of questionnaire responses on the frequency of use of other stores, targeting the customers who are not more than the average of the purchase amounts of the customers within the predetermined period by extracting the amount Is transmitted to the computer system of the center via a communication line, and a customer data analysis method of calculating the order of importance of customers having potential growth potential by a questionnaire response database collected and constructed every predetermined period, The computer system searches the questionnaire response database for each of the predetermined periods, and within the predetermined period. A family number extraction table generating means for extracting and totaling data of customer family number items for each customer and generating a table, and a child age extraction for extracting and totaling data of customer child age items for each customer and generating a table A table generation unit, and another store use frequency extraction table generation unit that extracts and tabulates other store use frequency item data for each customer to generate a table, wherein the system automatically generates the three units at the end of the predetermined period. A first step of extracting data from the three items in the previous term by using the respective program means and generating each table; and the system increases the contribution of the sales increase of the store for each of the generated data extraction tables in the future. Customer ranking means for generating the sorting tables in which the customers are sorted in the order of the highest possibility. When the first step is completed, the customer ranking means is generated there. A second step of generating a permuted table in which the extracted tables are rearranged by the customer ranking means, respectively, and the system further comprises: Sorting is performed for each table by sorting the n-level evaluation values with the evaluation value of the highest hierarchy being n, and classifying the evaluation values for each customer into each evaluation table. A third step in which, when the second step is completed, the customer evaluation unit generates each evaluation table for each of the seven items based on the rearranged tables generated in the second step; The system comprises: a total sum table generating means for each customer for summing the evaluation values of the respective evaluation tables for each customer; and a total sum table based on the sum result for each customer of the total sum table. m and overall ranking table generating means for the array of customers order in order, the highest hierarchy classified rank in the overall ranking table to multiple hierarchical the m and m
A total evaluation table generating means for each customer for calculating a graded evaluation value. When the third step is completed, the total sum table generating means automatically generates the total evaluation table for a predetermined period based on each evaluation table generated there. A fourth step of generating an overall total value table for each customer within the group, a fifth step of generating an overall ranking table for each customer by the overall total value table means, and then generating the overall evaluation table And a sixth step of generating a comprehensive evaluation table for each customer by the means. A list of important customers that may be most important in the future is created from the comprehensive evaluation table and displayed on the system display unit. A customer data analysis method for customer maintenance and training characterized by being able to do so. 9. The points of service points are entered in advance in accordance with the purchase price for the purpose of having the customer collect them, and at least the address and name of the customer are entered when applying for the point seal directly attached to the product of the specified manufacturer. Let
The data is transmitted to the computer system at the center via a communication line, collected at predetermined intervals, and the purchase amount corresponding to the application seal of the customer is extracted from the constructed application history database. At the point of purchase at the store, the survey data on the total purchase volume, purchase status of similar competitors' products, age, and the number of family members at least at the time of purchase at the store is targeted for customers with an average purchase price of less than A customer data analysis method for calculating the order of importance of customers having potential growth potential by using a questionnaire response database sent to a computer system of a center and collected and constructed at each of the predetermined periods, wherein the computer system includes the questionnaire. The answer database is searched for each of the predetermined periods, and within the predetermined period including a month or a year. Total purchase amount table generation means for extracting and totaling data on total purchase amount for each customer and generating a table, and generating a table for other companies purchase amount for extracting and totaling data on the purchase amount of similar competitor products for each customer Means, a customer age table generating means for extracting and totaling customer age data for each customer and generating a table, and a family number table generating means for generating a table extracting and totaling customer family number data for each customer. A first step in which the system automatically extracts data from the four items by each of the four means and generates each table at the end of each of the predetermined periods; and For each of the data extraction tables, the rearranged tables in which the customers are rearranged in the order in which the contribution of the increase in sales at the store is likely to increase in the future are generated. A second step of, when the first step is completed, generating a reordered table in which the extracted tables generated are rearranged by the customer ordering means. In addition, the system sorts each of the rearrangement tables from the highest to the lowest in customer priority into a plurality of layers, n, and sorts an n-level evaluation value with the evaluation value of the highest layer being n for each table. And a customer evaluation means for generating an evaluation table representing an evaluation value for each customer for each of the tables.
When the second step is completed, a third step of generating each evaluation table by the customer evaluation unit based on each rearrangement table generated therein, and the system further comprises: A total sum table generating means for each customer to be summed for each customer; and a total rank table generating means for arranging the customer ranks in the order of the total value based on the sum result for each customer in the total sum table; The order in the comprehensive order table is classified into m hierarchies, and the highest hierarchy is m.
A total evaluation table generating means for each customer for calculating a graded evaluation value. When the third step is completed, the total sum table generating means automatically generates the total evaluation table for a predetermined period based on each evaluation table generated there. A fourth step of generating an overall total value table for each customer within the group, a fifth step of generating an overall ranking table for each customer by the overall total value table means, and then generating the overall evaluation table And a sixth step of generating a comprehensive evaluation table for each customer by the means. A list of important customers that may be most important in the future is created from the comprehensive evaluation table and displayed on the system display unit. A customer data analysis method for customer maintenance and training characterized by being able to do so.