JP6143930B1 - Marketing support method, program, computer storage medium, and marketing support system - Google Patents

Abstract

Useful marketing data is provided in marketing using product or service usage days. A customer action for starting or stopping the use of a product or service, or a change in the state of the customer related to the use of the product or service is specified and set. Thereafter, the customers are classified into segments according to the presence or absence of an action or state change in the determination period T2. Thereafter, for the segments classified as having an action or a change in state, the utilization data for the total period T3 is totaled to calculate the expected value. Note that the start point of the determination period T2 is the same as the start point of the aggregation period T3 before the start point of the aggregation period T3. [Selection] Figure 3

Description

  The present invention relates to a marketing support method, a program, a computer storage medium, and a marketing support system that are executed by a computer using usage data of customer products or services.

  In general, it is important to perform appropriate marketing when selling products or providing services. Particularly in recent years, consumer society has diversified, individualized, and matured, and differentiated marketing is indispensable in order to efficiently sell products and provide services.

  Conventionally, various methods have been proposed as a method for performing the marketing described above. For example, purchase data of products purchased by customers and usage data using services are analyzed, and the analysis results are used for marketing of products and services.

  Patent Document 1 discloses performing customer tracking analysis (trend analysis) for grasping a time-series change in a purchase pattern of a target customer. In the customer tracking analysis, information registered in a database is analyzed using a database including customer attribute information indicating attributes such as a customer's name, sex, and address, and purchase information indicating products purchased by the customer. At this time, for example, a period and a customer segment are specified, and analysis is performed for each period and each customer segment. Then, based on the time series change of the customer's purchase pattern, the customers to be the target of the action (use promotion activity) are narrowed down and an appropriate action plan is determined.

JP 2004-185598 A

  However, although the customer tracking analysis disclosed in Patent Document 1 can grasp the time-series change of the customer's purchase pattern, it cannot accurately grasp what caused the change in the customer's purchase pattern. In other words, it is impossible to grasp a change factor that can be useful information when marketing a certain product or service. Therefore, it is difficult to say that marketing based on the customer tracking analysis of Patent Document 1 is optimal, and there is room for improvement.

  This invention is made | formed in view of this point, and it aims at providing useful marketing data (marketing information) in the marketing using the utilization day of goods or services.

In order to achieve the above object, the present invention provides a marketing support method executed by a computer using usage data of a customer's product or service, the customer's action for starting or stopping the use of the product or service, Alternatively, a setting step for identifying and setting a customer's state change related to the use of goods or services, a classification step for classifying a customer into segments according to the presence or absence of the action or the state change during a determination period, and a tabulation for each segment includes a calculation step of calculating an aggregate value based on the usage data of the period, the starting point of the determination period, Ri same der with the start point of the start point from the previous or the aggregation period of the counting period, the in calculating step , The segment classified as having the action or the state change in the classification step, Calculate the expected value that is the aggregate value, and the aggregate value is the cumulative value of the usage data in the aggregation period, or the usage data in the aggregation period and the period before the aggregation period and the same length as the aggregation period. It is characterized by the difference value or the rate of increase from usage data .
According to another aspect of the present invention, there is provided a marketing support method executed by a computer using usage data of a customer's product or service, the customer's action for starting or stopping the use of the product or service, or the product or service. A setting step for identifying and setting a customer's state change related to usage, a classification step for classifying a customer into segments according to the presence or absence of the action or the state change in the determination period, and usage data for the aggregation period for each segment And a calculation step of calculating a total value based on the start point of the determination period is the same as the start point of the total period or before the start point of the total period, the calculation step, in the classification step From the aggregate value of the segments classified as having the action or the state change, Subtract the aggregate value of the segments classified as having no action or state change in a similar process to calculate a push-up value, and the aggregate value is a cumulative value of usage data in the aggregation period, or in the aggregation period It is a difference value or an increase rate between the usage data and the usage data in the period of the same length as the aggregation period before the aggregation period.

  According to the present invention, according to the presence or absence of customer action or customer state change in the determination period, the customer is classified into segments, and the aggregate value is calculated based on the usage data of the aggregation period. Know which actions or state changes are useful for marketing. For example, when the total value for one action is higher than the total value for another action, it can be determined that one action is effective. Thus, the total value of the present invention can be useful marketing data. Further, by grasping useful actions or state changes as described above, it is also possible to make appropriate recommendations (recommendation of products, services, etc.) to customers. Therefore, the present invention exerts a very remarkable effect when performing marketing.

  In the marketing support method, in the calculation step, the aggregate value of the segments classified as having been subjected to the action in the classification step, and a ratio of customers who have performed the action in the determination period with respect to all customers. An action rate may be calculated, and an expected value may be further calculated based on the total value and the action rate.

  In the marketing support method, in the calculation step, the action is performed in the determination period for the total value of the segments classified as having undergone the action in the classification step and the state change and for all customers. An action rate indicating the percentage of customers who have performed the action during the determination period, and a state change rate indicating the percentage of customers in which the state change has occurred for the customers who have performed the action during the determination period. The expected value may be further calculated based on the state change rate.

  In the marketing support method, in the classification step, customers are classified into segments according to the presence or absence of the action that responded to the usage promotion activity during the determination period, and in the calculation step, the customer responds to the usage promotion activity in the classification step. Calculating the total value of the segments classified as having taken the action and the response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period for all customers. Then, an expected value may be further calculated based on the total value and the reaction rate.

  In the marketing support method, in the classification step, customers are classified into segments according to the presence or absence of the action and the presence or absence of the state change in response to the use promotion activity during the determination period, and in the calculation step, the classification step includes the step The action is performed in response to the use promotion activity, and the total value of the segment classified as having undergone the state change and the action in response to the use promotion activity in the determination period for all customers. A response rate indicating the percentage of customers who have changed, and a status change rate indicating the percentage of customers who have undergone the status change for a customer who has performed the action in response to the use promotion activity during the determination period. The expected value may be based on the total value, the reaction rate, and the state change rate.

  In the marketing support method, in the calculation step, the aggregated value may be calculated by decomposing the aggregated value into a directly aggregated value directly resulting from the use promotion activity and a spread aggregate value propagated by the use promotion activity. Good.

  In the marketing support method, the aggregation period is a period set for each customer in the segment classified as having the action or the state change in the classification step, and is based on the action or the state change. It may be a period of the predetermined length.

  According to another aspect of the present invention, there is provided a computer program for controlling the marketing support system so that the marketing support method is executed by the marketing support system.

  According to another aspect of the present invention, a readable computer storage medium storing the program is provided.

According to another aspect of the present invention, there is provided a marketing support system using customer product or service usage data, which is a customer action for starting or stopping the use of a product or service, or a customer related to the use of a product or service. A setting unit that identifies and sets a state change of the user, a classification unit that classifies customers into segments according to the presence or absence of the action or the state change in the determination period, and totals for each segment based on usage data for the total period has a calculation unit for calculating a value, the start point of the determination period, Ri same der with the start point of the start point from the previous or the aggregation period of the counting period, the calculation unit, the action in the classifying portion or For the segment classified as having the state change, the expected value that is the aggregate value is calculated, and the aggregate value is Is characterized in that the accumulated value of the usage data in the aggregation period, or the a the aggregation period before the usage data in the aggregation period is a difference value or rate of increase in the usage data in a period of the same length as the total period Yes.
Another aspect of the present invention is a marketing support system using usage data of a customer's product or service, and the customer's action for starting or stopping the use of the product or service, or the customer's use regarding the use of the product or service. A setting unit that identifies and sets a state change, a classification unit that classifies customers into segments according to the presence or absence of the action or the state change in a determination period, and a total value based on usage data for a total period for each segment And the start point of the determination period is the same as the start point of the aggregation period or the start point of the aggregation period, and the calculation unit is configured to perform the action or the state in the classification unit. From the aggregated value of the segments classified as having a change, the action or the state in the classification unit By subtracting the aggregate value of the segments classified as having no change, the boost value is calculated, and the aggregate value is the accumulated value of the usage data in the aggregation period, or the usage data in the aggregation period and before the aggregation period It is characterized by the difference value or the rate of increase from the usage data in the period having the same length as the total period.

  In the marketing support system, the calculation unit indicates the aggregate value of the segments classified as having undergone the action in the classification unit, and the percentage of customers who have performed the action during the determination period with respect to all customers. An action rate may be calculated, and an expected value may be further calculated based on the total value and the action rate.

  In the marketing support system, the calculation unit performs the action during the determination period for the aggregate value of the segments classified as having undergone the action by the classification unit and having undergone the state change, and all customers. An action rate indicating the percentage of customers who have performed the action during the determination period, and a state change rate indicating the percentage of customers in which the state change has occurred for the customers who have performed the action during the determination period. The expected value may be further calculated based on the state change rate.

  In the marketing support system, the classification unit classifies customers into segments according to the presence or absence of the action that responds to the usage promotion activity during the determination period, and the calculation unit responds to the usage promotion activity in the classification unit. Calculating the total value of the segments classified as having taken the action and the response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period for all customers. The expected value may be further calculated based on the total value and the reaction rate.

  In the marketing support system, the classification unit classifies customers into segments according to the presence / absence of the action and the presence / absence of the state change in response to the use promotion activity during the determination period. The action is performed in response to the use promotion activity, and the total value of the segment classified as having undergone the state change and the action in response to the use promotion activity in the determination period for all customers. A response rate indicating the percentage of customers who have changed, and a status change rate indicating the percentage of customers who have undergone the status change for a customer who has performed the action in response to the use promotion activity during the determination period. The expected value may be further calculated based on the total value, the reaction rate, and the state change rate.

  In the marketing support system, the calculation unit may calculate the aggregated value by decomposing the aggregated value into a directly aggregated value directly resulting from the use promotion activity and a spread aggregate value propagated by the use promotion activity. Good.

  In the marketing support system, the counting period is a period set for each customer in a segment classified as having the action or the state change in the classification unit, and the action or the state change is a starting point. It may be a period of the predetermined length.

  ADVANTAGE OF THE INVENTION According to this invention, useful marketing data can be provided in the marketing using the utilization data of goods or a service.

It is explanatory drawing which shows the outline of a structure of the marketing assistance system concerning this Embodiment. It is explanatory drawing which shows a mode that a customer is classified into a segment. It is explanatory drawing of an estimated value, (a) shows a cumulative value, (b) shows a difference value and a rate of increase, and (c) shows an arrival probability. It is explanatory drawing which shows a total period. An example of credit card usage data in the first embodiment is shown. The expected value and the push-up value (use amount, use frequency) in the first embodiment are shown. The expected value and the push-up value (use amount, use frequency, with narrowing down) in the first embodiment are shown. The expected value and push-up value (utilization increase rate) in the first embodiment are shown. The expected value and the push-up value (customer remaining rate) in the first embodiment are shown. An example of the utilization data of the department store in 2nd Embodiment is shown. The expected expected value and the expected boost value (action rate × total value) in the second embodiment are shown. The expected expected value and the expected boost value (action rate × (direct total value + spread total value)) in the second embodiment are shown. The expected expected value and the expected push-up value (action rate × state change rate × total value) in the second embodiment are shown. The expected expected value and the expected boosted value (action rate × state change rate × (direct total value + spread total value)) in the second embodiment are shown. An example of the utilization data of the fresh supermarket in 3rd Embodiment is shown. The expected expected value and the expected boost value (reaction rate × total value) in the third embodiment are shown. The expected expected value and the expected boosted value in the third embodiment (reaction rate × (direct total value + propagation total value)) are shown. The expected expected value and the expected boosted value (reaction rate × state change rate × total value) in the third embodiment are shown. The expected expected value and the expected boosted value (reaction rate × state change rate × (direct count value + spread count value)) in the third embodiment are shown.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

<1. Configuration of marketing support system>
FIG. 1 shows an outline of the configuration of a marketing support system 1 according to the present embodiment. The marketing support system 1 is connected to the external system 2 via the network N. The network N is not particularly limited as long as it can communicate between the marketing support system 1 and the external system 2. For example, the network N is configured by the Internet, a wired LAN, a wireless LAN, or the like. .

  Then, the marketing support system 1 processes the usage data input from the external system 2 to the marketing support system 1 to generate marketing data (hereinafter referred to as marketing data). In addition, various data are used as utilization data. For example, purchase data for merchandise in retail and manufacturing, transaction data at financial institutions (eg deposits, withdrawals, payments, etc.), eating and drinking data at restaurants, accommodation data at accommodation facilities, music / video services and game play / play data, points Point usage data at the site. That is, the use in the present invention is a broad concept including the purchase of these products and the use of services.

  The marketing support system 1 includes a communication unit 10, a setting unit 11, a classification unit 12, a calculation unit 13, a storage unit 14, and an output unit 15.

  The communication unit 10 is a communication interface that mediates communication with the network N, and performs data communication with the external system 2.

  The setting unit 11 specifies and sets a customer action or a customer state change. This action or state change may be set from the setting unit 11 by an analyst, or may be automatically set from the setting unit 11 by a computer. The action or state change set by the setting unit 11 is stored in the storage unit 14.

The customer action is an action for starting or stopping the use of a product or service. In particular,
・ Purchase products and products of categories,
・ Start using the service, or stop using the service,
Etc. The category is a plurality of product groups of the same type. The action includes a case where the customer actively takes an action and a case where the customer passively takes an action through the use promotion activity. The usage promotion activities include, for example, sales promotion activities such as events related to products and advertisements for products, usage promotion activities such as events related to services and advertisements for products.

A customer status change is a customer status change related to the use of goods or services. In particular,
-Changing from a state where there is no purchase of a product or category to a state where it is present, or changing from a state where purchase of a product or category is present to a state where there is no purchase,
・ Change from a state where the service is not used to a state where the service is used, or change from a state where the service is used to a state where the service is not used,
・ Changes in customer segments categorized by usage amount, usage frequency, number of usage categories, etc.
Etc. The state change includes a case where the customer actively causes a state change and a case where the customer passively changes the state as a result of taking an action.

  In the setting unit 11, what is calculated as marketing data in the calculation unit 13 described later is set, and various conditions for calculating the marketing data are also set.

  The classification unit 12 classifies the customers into segments based on the usage data depending on whether the customer's action or the customer's state has changed during a predetermined determination period. The customer segments classified by the classification unit 12 are stored in the storage unit 14. The determination period is a period for determining whether or not there is an action or a state change.

  The calculation unit 13 calculates marketing data by totaling usage data for the total period for each customer segment. The above-described determination period is before or within the counting period. The marketing data includes, for example, an expected value (aggregated value of the present invention), a boosted value (aggregated value of the present invention), an expected expected value (expected value of the present invention), and an expected increased value (expected value of the present invention). Can be mentioned. The marketing data calculated by the calculation unit 13 is stored in the storage unit 14. The four contents of the marketing data described above will be described later.

  The storage unit 14 stores various data processed by the marketing support system 1. Specifically, usage data input to the marketing support system 1, data regarding actions or state changes set by the setting unit 11, data regarding customer segments classified by the classification unit 12, calculated by the calculation unit 13 Marketing data is stored. The storage unit 14 stores various programs for calculating marketing data.

  The output unit 15 outputs the marketing data stored in the database of the storage unit 14 to the outside of the marketing support system 1.

  The components of the marketing support system 1 shown in FIG. 1 can be composed of a circuit (hardware) or a central processing unit such as a CPU and a program (software) for causing them to function. And this program controls each part 10-15, and performs the data processing mentioned later. In this case, the program may be stored in, for example, the storage unit 14, or may be a computer-readable hard disk (HD), flexible disk (FD), compact disk (CD), magnet optical desk (MO), It may be stored in a computer-readable storage medium such as various memories. The program can be stored in the storage medium or the like by downloading it via a communication line network such as the Internet.

<2. Marketing data>
Next, marketing data calculated by the calculation unit 13 described above will be described. The marketing data includes the following (1) to (4). In order to calculate the marketing data in the calculation unit 13, the customer action or the customer state change is specified and set in the setting unit 11, and the customer action or the customer state in the determination period is further set in the classification unit 12. Classify customers into segments based on whether or not there are changes.
(1) Expected value (2) Push-up value (3) Expected expected value (4) Expected push-up value

(1) Expected value The estimated value is calculated by aggregating the usage data of the aggregation period for the segment of the customer who took action during the determination period or the segment of the customer whose state change occurred during the determination period. Note that, as described above, the customer segment in which the state change has occurred includes the case where the state change has occurred as a result of the customer taking an action. That is, the segment when calculating the expected value includes the segment when the customer actively takes action, the segment when the customer's state change actively occurs, and the state change as a result of the customer taking action 3 patterns of segments are included when passively occurs.

  FIG. 2 is an explanatory diagram showing how customers are classified into segments. The graph in FIG. 2 shows changes over time in usage data for each customer. In FIG. 2, the comparison period T1, the determination period T2, and the counting period T3 are continuous. The comparison period T1 is a period for calculating a difference value / increase rate as a prospective value, which will be described later, and has the same length as the total period T3. The determination period T2 is a period for the classification unit 12 to determine whether there is an action or a state change. The aggregation period T3 is a period for the calculation unit 13 to calculate the expected value. In the present embodiment, the case where the determination period T2 is before the aggregation period T3 will be described. However, the determination period T2 may be within the aggregation period T3, that is, the determination period T2 and the aggregation period T3 coincide. It may be. In other words, the start point of the determination period T2 may be the same as the start point of the aggregation period T3 before the start point of the aggregation period T3.

  The usage data includes a plurality of customer data as shown in FIG. 2 (a). As shown in FIG. 2 (b), the classification unit 12 determines the customer according to the action or state change in the determination period T2. Classify into segments. In FIG. 2B, a thick line indicates a customer who has taken an action during the determination period T2 or a customer whose state has changed during the determination period T2. A dotted line indicates a customer who has not taken an action during the determination period T2 or a customer whose state has not changed during the determination period T2.

  In the calculation unit 13, the thick line customer segment in FIG. 2B (the customer segment that caused an action in the determination period T2 or the customer segment in which the state change occurred in the determination period T2) is calculated for the total period T3. Aggregate usage data to calculate the expected value. For example, when the usage data is POS data, the transaction data in the aggregation period T3 is aggregated to calculate the expected value.

  The expected value is divided into three patterns of “cumulative value”, “difference value / increase rate”, and “reaching probability”. FIG. 3 is an explanatory diagram of expected values.

  The accumulated value as the expected value is a value X obtained by accumulating usage data for the total period T3 as shown in FIG. Examples of the accumulated value include the amount of money used, the number of times used, the number of used products, the number of used categories, the number of used stores, and profits in the total period T3.

  As shown in FIG. 3B, the difference value / increase rate as the expected value is the difference value Y between the usage data in the aggregation period T3 and the usage data in the comparison period T1, and is the increase rate Y (%). As the difference value, for example, the amount of increase in the usage amount from the comparison period T1 to the aggregation period T3, the increase in the number of usages, the increase in the number of usage products, the increase in the number of usage categories, the increase in the number of usage stores, and the increase in profit Forehead. Further, as the rate of increase, for example, the rate of increase in the usage amount, the rate of increase in the number of usages, the rate of increase in the number of used products, the rate of increase in the number of used categories, the rate of increase in the number of used stores, and profit The rate of growth of

  The comparison period T1 may be before the aggregation period T3, and the comparison period T1 and the aggregation period T3 may be continuous.

  The arrival probability as the expected value is, for example, a probability Z (%) of reaching the specific condition after the total period T3 as shown in FIG. Specific conditions include, for example, the probability of reaching the category of excellent customers after the total period T3, the probability that the usage amount is a predetermined amount or more, the probability of being a member (that is, not leaving), the probability of purchasing a specific product, and the specification This includes the probability of using goods. Note that the arrival probability may be determined after the aggregation period T3, may be at the end of the aggregation period T3, or may be a point at which a predetermined period is opened after the aggregation period T3.

  According to such a prospective value, it is possible to estimate how much the usage after the action or state change will be caused by the customer taking action during the judgment period or when the customer's state change occurs during the judgment period. .

(2) Push-up value The push-up value is the aggregate value of the usage data of the customer segment that had no action or status change during the judgment period, from the aggregate value of the customer segment that had an action or status change during the judgment period. Is calculated by subtracting.

  In calculating the push-up value, the aggregate value in the previous section, that is, the aggregate value of the usage data of the segment of the customer whose action or state has changed during the determination period corresponds to the above-described expected value. The aggregate value in the previous section is the aggregate period for the thick line customer segment in FIG. 2B (the customer segment that caused an action during the determination period T2 or the customer segment whose state change occurred during the determination period T2). T3 usage data is aggregated and calculated.

  The aggregate value of the latter term (hereinafter referred to as comparison value), that is, the aggregate value of the usage data of the customer segment for which no action or state change occurred during the determination period is the dotted customer segment (determination period T2 in FIG. 2B). The usage data of the aggregation period T3 is aggregated and calculated for the customer segment that has not taken any action or the customer segment for which no state change has occurred during the determination period T2.

  The aggregate value (expected value) of the previous item and the aggregate value (comparison value) of the latter item are divided into three patterns of “cumulative value”, “difference value / increase rate”, and “arrival probability” as shown in FIG. The push-up value is also divided into three patterns of “cumulative value”, “difference value / increase rate”, and “reaching probability”.

  According to the boost value, the action or state change is caused by the customer taking action during the determination period or the change of the customer state during the determination period, compared to the customer who did not change the action or state. You can estimate how much later use will be boosted.

(3) Expected expected value The expected expected value is a value calculated by the following eight patterns.
-Expected Expected Value = Action Rate x Expected Value-Expected Expected Value = Action Rate x State Change Rate x Expected Value-Expected Expected Value = Action Rate x (Direct Expected Value + Expected Ripple Value)
-Expected expected value = action rate x state change rate x (direct expected value + estimated ripple value)
-Expected Expected Value = Response Rate x Expected Value-Expected Expected Value = Response Rate x State Change Rate x Expected Value-Expected Expected Value = Response Rate x (Direct Expected Value + Expected Ripple Value)
-Expected expected value = reaction rate x state change rate x (direct expected value + estimated ripple value)

  The expected value in the above formula is the expected value of (1) above.

  The action rate indicates the percentage of customers who took action during the determination period with respect to all customers. As described above, actions include cases in which a customer actively takes action and cases in which a customer takes an action passively through use promotion activities such as events and advertisements. Absent.

  The response rate indicates the percentage of customers who took action in response to the usage promotion activity during the determination period for all customers. That is, in calculating the reaction rate, the case where the customer actively takes action is not included.

  The state change rate indicates the ratio of customers who have made an action during the determination period to customers who have taken action during the determination period. As described above, the state change includes a case where the customer actively causes a state change and a case where the customer passively changes the state as a result of taking an action.

  The direct likelihood value is a prospective value of a customer whose action promotion activity coincides with the action object in the segment of the customer who took action during the determination period. For example, if the usage promotion activity is a milk advertisement and the action purchases milk, the prospective value of the customer is the direct prospective value.

  The expected spread value is the expected value of a customer whose action promotion activity and action target do not match in the segment of the customer who took action during the determination period. For example, when the use promotion activity is an advertisement for milk and the action purchases a product other than milk, the action is spread from the use promotion activity, and the expected value of the customer is the estimated spread value.

  Note that the sum of the direct expected value and the expected spread value is the expected value of (1) above.

  According to the expected expected value, the expected value of how much the usage after the action or state change will occur when the customer takes action during the judgment period or when the customer's state change occurs during the judgment period. It can be estimated in consideration of the probability of causing an action or state change.

(4) Expected push-up value The push-up expectation value is a value calculated by the following eight patterns.
-Expected push-up value = action rate x push-up value-Expected push-up value = action rate x state change rate x push-up value-Expected push-up value = action rate x (direct push-up value + ripple push-up value)
-Expected push-up value = action rate x state change rate x (direct push-up value + ripple push-up value)
-Expected push-up value = reaction rate x push-up value-Expected push-up value = reaction rate x state change rate x push-up value-Expected push-up value = reaction rate x (direct push-up value + ripple push-up value)
-Expected push-up value = reaction rate x state change rate x (direct push-up value + ripple push-up value)

  The expected push-up value is obtained by replacing the expected value in the expected expectation value in (3) above with the push-up value in (2) above. Since the description of each term in the above formula is as described above, the detailed description is omitted.

  According to the expected increase in value, the action or state is compared to the customer who did not change the action or the state due to the customer taking action during the determination period or the change in the state of the customer during the determination period. An expected value of how much the usage after the change is pushed up can be estimated in consideration of the probability of causing an action or a state change.

  In the above description, the aggregation period T3 when calculating the expected value, the boosted value, the expected expected value, and the expected boosted value is a period common to a plurality of customers in the segment. On the other hand, as shown in FIG. 4, the total period T3 may be set for each customer.

  FIG. 4 shows a segment of a customer who took an action during the determination period T2, or a customer segment whose state change occurred during the determination period T2. The triangle mark in FIG. 4 is the time when each customer takes action or when the customer's state change occurs. The aggregation period T3 is a period set for each customer in the segment, and is a period of a predetermined length starting from an action or state change. The comparison period T1 is a period with an action or state change as an end point, and has the same length as the total period T3.

  For the aggregation period T3 and the comparison period T1 set in this way, the expected value, the boosted value, the expected expected value, and the expected boosted value can be calculated by the same method as described above.

<3. Marketing support method>
Next, data processing performed using the marketing support system 1 configured as described above will be described. In the marketing support method of the present invention, data processing is performed to calculate marketing data. The data processing of the present invention will be described below using an embodiment including three specific cases.

<3-1. First Embodiment>
In the first embodiment, the use of a credit card will be described as an example. Specifically, the external system 2 is a database system provided in a credit card company, for example, and the marketing support system 1 processes credit card usage data acquired from the database system. In the present embodiment, a prospective value and a push-up value are calculated as marketing data.

  FIG. 5 shows an example of credit card usage data. Usage data is obtained by extracting data necessary for calculating marketing data in the setting unit 11 based on data input from the external system 2 to the marketing support system 1. The period from December 2014 to February 2015 is the comparison period T1, the period March 2015 is the determination period T2, and the period April to June 2015 is the aggregation period T3.

  The usage data includes the setting status of the mail magazines (hereinafter referred to as mail magazines) of the customers A to W, the card usage amount, and the card usage count in the comparison period T1 and the aggregation period T3. In addition, the usage data includes a setting change of the mail magazines of the customer A to the customer W in the determination period T2. This e-mail magazine setting state indicates the states of the customers A to W, and the e-mail magazine setting change corresponds to a state change. Further, the usage data includes the card validity status as of March 2016 after the total period T3.

  Next, an example in which the calculation unit 13 calculates marketing data (expected value, push-up value) based on the credit card usage data shown in FIG. 5 will be described. As described above, the expected value and the pushed-up value are divided into three patterns of “cumulative value”, “difference value / increase rate”, and “reach probability”.

  FIG. 6 shows cumulative values as expected values and push-up values. In calculating the expected value a, first, customers who have changed from the absence of mail magazine setting status to the presence status in the determination period T2 are classified into segments. Then, as the expected value a, the card usage amount and the card usage count in the total period T3 of this segment are totaled, and the total value is divided by the number of customers to calculate an average value per customer.

  In calculating the push-up value c, first, customers who have no mail magazine setting status in the determination period T2 are classified into segments. The amount of card usage and the number of card usages in the total period T3 of this segment are totaled, and the total value is divided by the number of customers to calculate an average value per customer (hereinafter referred to as comparison value b). The push-up value c is calculated by subtracting the comparison value b from the expected value a.

  FIG. 7 shows cumulative values as expected values and push-up values calculated by adding a narrowing condition. As a narrowing-down condition, customers whose usage amount in the comparison period T1 is 40,000 yen or more and 100,000 yen or less are targeted. Then, as shown in FIG. 6, when calculating the expected value a and the pushed-up value c, the customers targeted by this narrowing condition are classified into segments. The calculation method of the expected value a and the push-up value c is the same as the calculation method described with reference to FIG.

  FIG. 8 shows the rate of increase as the expected value and the pushed-up value. When calculating the expected value a and the push-up value c, first, as shown in FIG. 8A, the rate of increase in the usage amount of each customer from the comparison period T1 to the totaling period T3 (hereinafter referred to as the usage amount increase rate). Is calculated. Then, as shown in FIG. 8 (b), as the expected value a, the usage rate increase rate is totaled for the customer segment that has been changed from the absence of the mail magazine setting to the presence in the determination period T2, and the totaled value is further calculated. Divide by the number of customers to calculate the average value per customer.

  In calculating the push-up value c, first of all, the usage rate increase rate is totaled for the segment of the customer who has no e-mail magazine setting status in the determination period T2, and the total value is divided by the number of customers. An average value per customer (hereinafter referred to as comparison value b) is calculated. The push-up value c is calculated by subtracting the comparison value b from the expected value a.

  FIG. 9 shows the arrival probability as the expected value and the push-up value. The specific condition serving as a reference for the arrival probability is the card situation as of March 2016. If the card is valid at that time, it is determined that the specific condition has been reached. In other words, the arrival probability is the remaining rate of customers. Then, as the expected value a, the ratio of the number of remaining customers to the total number of customers is calculated in the customer segment that has been changed from having no e-mail magazine setting to being present in the determination period T2.

  In calculating the push-up value c, first, the ratio of the number of remaining customers to the total number of customers (hereinafter referred to as a comparison value b) is calculated in the customer segment in which the mail magazine setting state is not present in the determination period T2. The push-up value c is calculated by subtracting the comparison value b from the expected value a.

  According to the present embodiment, the effect of mail magazine setting (state change) can be grasped. That is, by grasping the expected value, it is possible to estimate how much the usage after the mail magazine setting will be. In addition, by grasping the push-up value, it is possible to estimate how much the use after the mail magazine setting is pushed up when the mail magazine setting is performed and when the mail magazine setting is not performed.

  In addition, by grasping the effect of e-mail magazine setting in this way, it becomes possible to distribute effective e-mail magazines to customers.

<3-2. Second Embodiment>
In the second embodiment, the use of a department store will be described as an example. The external system 2 is a database system provided in, for example, a department store, and the marketing support system 1 processes customer usage data (product purchase data) acquired from the database system. In this embodiment, the expected expected value and the expected boost value using the action rate are calculated as marketing data.

  FIG. 10 shows an example of department store usage data. Usage data is obtained by extracting data necessary for calculating marketing data in the setting unit 11 based on data input from the external system 2 to the marketing support system 1. December 2015 is a determination period T2 of whether or not there is an action, and January to June 2016 is a total period T3. The aggregation period T3 is also a determination period for the presence or absence of a state change, as will be described later.

  The usage data includes a category (hereinafter referred to as a target category) of products that have not been purchased by customers A to X until November 2015, which is before the determination period T2. In addition, the usage data includes the presence / absence of new purchases of products in the target category of customers A to X in the determination period T2. A new purchase of a product of this target category corresponds to a customer action. Further, the usage data includes whether or not the products of the target categories of the customers A to X are purchased again, the purchase price of the products of the target category, the purchase price of the products other than the target category, the products of all categories in the aggregation period T3. Includes total purchase price. Among these, the presence / absence of purchase of the product of the target category indicates whether or not the purchase of the customer has been established, that is, the presence / absence of a state change.

  Next, an example in which the calculation unit 13 calculates marketing data (expected expected value, expected increase value) based on the department store usage data shown in FIG. 10 will be described.

FIG. 11 shows expected expected values and boosted values calculated by the following formula.
-Expected expected value = action rate x expected value-Expected push-up value = action rate x push-up value

  First, for each target category, customers who newly purchased products of the target category in the determination period T2 are classified into segments, and the number of customers a is totaled. In addition, the customers who did not newly purchase the products in the target category in the determination period T2 are classified into segments, and the number of customers c is totaled. Then, an action rate e indicating the percentage of customers who took action during the determination period T2 is calculated. Specifically, the action rate e is the ratio of the number of customers a who has taken action with respect to the total number of customers (a + c).

  In calculating the expected expected value f, the expected value b, which is the average purchase price of the products of all categories, is calculated for the segment of the customer who newly purchased the products of the target category in the determination period T2. The method for calculating the expected value b is the same as the method for calculating the expected value in the first embodiment, and a description thereof will be omitted. The expected expected value f is calculated by multiplying the action rate e by the expected value b.

  In calculating the expected increase value h, a comparison value d, which is an average purchase price of products of all categories, is calculated for a segment of a customer who has not newly purchased a product of the target category in the determination period T2, and further the boost value g (= expected value b−comparison value d) is calculated. The method for calculating the push-up value g is the same as the method for calculating the push-up value in the first embodiment, and thus the description thereof is omitted. The expected increase value h is calculated by multiplying the action rate e by the increase value g.

FIG. 12 shows expected expected values and boosted values calculated by the following formula.
-Expected expected value = action rate x (direct expected value + estimated ripple value)
-Expected push-up value = action rate x (direct push-up value + ripple push-up value)

  First, the number of customers a who newly purchased products of the target category during the determination period T2 and the number of customers d who did not newly purchase the products of the target category during the determination period T2 are totaled to calculate the action rate g.

  In calculating the expected expected value h, the direct expected value b, which is the average purchase price of the product in the target category, is calculated for the segment of the customer who newly purchased the product in the target category in the determination period T2. In addition, for a segment of a customer who newly purchases a product in the target category during the determination period T2, an estimated ripple value c that is an average purchase price of a product other than the target category is calculated. The expected expected value h is calculated by multiplying the action rate e by the sum of the expected value b and the expected spread value c directly.

  In calculating the expected push-up value k, a direct comparison value e, which is the average purchase price of the product in the target category, is calculated for the segment of the customer who did not newly purchase the product in the target category in the determination period T2, and the determination period A ripple comparison value f, which is an average purchase price of products other than the target category, is calculated for a segment of a customer who has not newly purchased products of the target category at T2. Further, a direct push-up value i (direct expectation value b-direct comparison value e) and a ripple push-up value j (spread expectation value c-propagation comparison value f) are calculated. The expected push-up value k is calculated by multiplying the action rate e by the sum of the push-up value i and the ripple push-up value j.

FIG. 13 shows expected expected values and boosted values calculated by the following formula.
-Expected expected value = Action rate x State change rate x Expected value-Expected push-up value = Action rate x State change rate x Push-up value

  First, the number of customers a that have newly purchased a product of the target category in the determination period T2 and purchased again in the total period T3 is totaled. In addition, the number of customers c who newly purchased a product in the target category during the determination period T2 and did not purchase again during the total period T3 is totaled. Further, the number d of customers who did not newly purchase the product of the target category during the determination period T2 is totaled. The action is a new purchase of a product of the target category in the determination period T2, and the action rate f is calculated by dividing the number of customers who have caused the action (a + c) by the total number of customers (a + c + d). The state change is that a customer who newly purchases a product in the target category in the determination period T2 purchases the product in the target category again in the aggregation period T3. The state change rate is calculated by dividing the number of customers c in which the state change has occurred by the number of newly purchased customers (a + c).

  In calculating the expected expected value h, the expected value b, which is the average purchase price of the products of all categories, is calculated for the segment of the customer who newly purchased the products of the target category in the determination period T2. The expected expected value h is calculated by multiplying the action rate f, the state change rate g, and the expected value b.

  In calculating the expected increase value j, a comparison value e, which is the average purchase price of the products in all categories, is calculated for the segment of the customer who did not newly purchase the products in the target category in the determination period T2, and the increase value i (= expected value b−comparison value e) is calculated. The expected push-up value h is calculated by multiplying the action rate f, the state change rate g, and the push-up value i.

FIG. 14 shows the expected expected value and the boosted value calculated by the following formula.
-Expected expected value = action rate x state change rate x (direct expected value + estimated ripple value)
-Expected push-up value = action rate x state change rate x (direct push-up value + ripple push-up value)

  The expected expected value j is obtained by decomposing the expected value b in the expected expected value h in FIG. 13 directly into the expected value b and the estimated expected value c. The expected value decomposition method is the same as the expected value decomposition method shown in FIG.

  The expected push-up value n is obtained by dividing the push-up value i into the push-up value k and the ripple push-up value m in the push-up expected value j in FIG. The method for decomposing the push-up value is the same as the method for decomposing the push-up value shown in FIG.

  According to the present embodiment, it is possible to grasp the effect of a new purchase (action) of a product in the target category and the effect of purchasing the product in the target category again (state change).

  That is, by grasping the expected expected value (action rate × expected value), it is possible to estimate how much the product in the target category will be used after new purchase. Further, by grasping the expected expected value (action rate × state change rate × expected value), it is possible to estimate how much the product in the target category will be used after purchase again. Furthermore, by directly decomposing the expected value into the expected value and the estimated value, it is possible to grasp the direct effect of the customer purchasing the product in the target category, or the effect of purchasing the product other than the target category. The effect can be grasped.

  In addition, by grasping the expected increase value (action rate x increase value), when a product of the target category is newly purchased, compared to the case where the product is not newly purchased, which is used after the new purchase of this product? You can estimate how much it will be pushed up. In addition, by grasping the expected push-up value (action rate x state change rate x push-up value), when the product of the target category is purchased again, the product is purchased again after purchasing the product again compared to when the product is not purchased again. It is possible to estimate how much use will be. Furthermore, by decomposing the push-up value directly into the push-up value and the ripple push-up value, it is possible to grasp the direct effect of the customer purchasing the product of the target category, or the ripple effect of purchasing the product other than the target category. The effect can be grasped.

  Then, by grasping the effect of the new purchase (action) of the product in the target category and the effect of purchasing the product in the target category again (status change), it is possible to examine which category of the product should be strengthened by the department store. it can.

<3-3. Third Embodiment>
In the third embodiment, the use of a fresh supermarket will be described as an example. The external system 2 is, for example, a database system provided at a fresh supermarket, and the marketing support system 1 processes customer usage data (product purchase data) acquired from the database system. In this embodiment, an expected expected value and a boosted expected value using the reaction rate are calculated as marketing data.

  FIG. 15 shows an example of fresh supermarket usage data. Usage data is obtained by extracting data necessary for calculating marketing data in the setting unit 11 based on data input from the external system 2 to the marketing support system 1. March 31, 2016 is a determination period T2 of whether or not there is an action, and April to June 2016 is a total period T3. The aggregation period T3 is also a determination period for the presence or absence of a state change, as will be described later.

  In the present embodiment, coupons are distributed to customers A to W during the determination period T2. Coupons for different products for each customer are distributed to the customers A to W. This coupon distribution corresponds to a use promotion activity.

  The usage data includes the product of a coupon (hereinafter referred to as coupon distribution product) distributed to the customer A to the customer W in the determination period T2, and whether or not the coupon distribution product is purchased. The purchase of the coupon distribution product corresponds to a customer action. In addition, the usage data includes whether or not the coupon delivery products of the customers A to W are purchased again, the purchase price of the coupon delivery product, the purchase price of the product other than the coupon delivery product, and the total purchase price of all the products during the counting period T3. Is included. Of these, whether or not the coupon distribution product is purchased again indicates whether or not the purchase of the customer has been established, that is, corresponds to the presence or absence of a state change.

  Next, an example in which the calculation unit 13 calculates marketing data (expected expected value, expected increase value) based on the usage data of the fresh supermarket shown in FIG. 15 will be described.

FIG. 16 shows the expected expected value and the boosted value calculated by the following formula.
-Expected expected value = reaction rate x expected value-Expected push-up value = reaction rate x push-up value

  First, the customers who purchased the coupon distribution product during the determination period T2 are classified into segments, and the number of customers a is counted. Further, the customers who did not purchase the coupon distribution product in the determination period T2 are classified into segments, and the number of customers c is totaled. Then, a response rate e indicating the proportion of customers who have taken action in response to the usage promotion activity in the determination period T2 is calculated. Specifically, the response rate e is the ratio of the number of customers a who took action in response to the use promotion activity to all customers (a + c).

  The expected expected value f can be calculated by a method similar to the method of calculating the expected expected value f in FIG. 11, and the action rate e in the expected expected value f in FIG. 11 may be replaced with the reaction rate e. Further, the expected increase value h can be calculated by the same method as the expected increase value h in FIG. 11, and the action rate e in the expected increase value h in FIG. 11 may be replaced with the reaction rate e.

FIG. 17 shows expected expected values and boosted values calculated by the following formula.
-Expected expected value = reaction rate x (direct expected value + estimated ripple value)
-Expected push-up value = reaction rate x (direct push-up value + ripple push-up value)

  The expected expected value h can be calculated by the same method as the expected expected value h in FIG. 12, and the action rate g in the expected expected value f in FIG. 12 may be replaced with the reaction rate g. Further, the expected increase value k can be calculated by the same method as the expected increase value k shown in FIG. 12, and the action rate g at the expected increase value k shown in FIG. 12 may be replaced with the reaction rate g.

FIG. 18 shows the expected expected value and the boosted value calculated by the following formula.
-Expected expected value = reaction rate x state change rate x expected value-expected push-up value = reaction rate x state change rate x push-up value

  The expected expected value h can be calculated by a method similar to the method of calculating the expected expected value h in FIG. 13, and the action rate f at the expected expected value h in FIG. Further, the expected increase value j can be calculated by the same method as the expected increase value j shown in FIG. 13, and the action rate f in the expected increase value j shown in FIG.

FIG. 19 shows the expected expected value and the boosted value calculated by the following formula.
-Expected expected value = reaction rate x state change rate x (direct expected value + estimated ripple value)
-Expected push-up value = reaction rate x state change rate x (direct push-up value + ripple push-up value)

  The expected expected value j can be calculated by a method similar to the method of calculating the expected expected value j in FIG. 14, and the action rate h in the expected expected value j in FIG. Further, the expected increase value n can be calculated by the same method as the expected increase value n shown in FIG. 14, and the action rate h at the expected increase value n in FIG. 14 may be replaced with the reaction rate h.

  According to the present embodiment, it is possible to grasp the effect of coupon distribution (use promotion activity).

  That is, by grasping the expected expected value (reaction rate × expected value), it is possible to estimate how much the coupon distribution product will be used after purchase. In addition, by grasping the expected expected value (reaction rate × state change rate × expected value), it is possible to estimate how much the coupon distribution product will be used after purchase again. Furthermore, by directly decomposing the expected value into the expected value and the estimated value, it is possible to grasp the direct effect of the customer purchasing the coupon distribution, or the ripple effect of purchasing a product other than the coupon distributed product. Can be grasped.

  Also, by grasping the expected push-up value (reaction rate x push-up value), when purchasing a coupon distribution product, compared to the case where the coupon distribution product is not purchased, what is the use of the coupon distribution product after purchase? You can estimate how much it will be pushed up. Also, by grasping the expected push-up value (reaction rate × state change rate × push-up value), when the coupon delivery product is purchased again, the coupon delivery product's You can estimate how much you will use after purchasing again. Furthermore, by decomposing the boost value directly into a boost value and a ripple boost value, the customer can grasp the direct effect of purchasing coupon distribution, or the ripple effect of purchasing a product other than the coupon distribution product. Can be grasped.

  And an effective coupon can be distributed with respect to a customer by grasping the effect of coupon distribution in this way.

<3-4. Effect of marketing support method>
As described above, as shown in the first to third embodiments, according to the present invention, various types of marketing data (marketing information) can be calculated using the use data of goods or services. it can. In particular, the usage data of goods or services is so-called big data, and the marketing data calculated therefrom shows a good tendency. Therefore, the marketing data of the present invention is novel and unprecedented and is extremely useful for marketing.

  In the present invention, various marketing data can be calculated, and the marketing data can be appropriately selected according to what kind of marketing is performed. And by grasping useful marketing data like this invention, it also becomes possible to perform an appropriate recommendation (recommendation of goods and services) to a customer. Various specific recommendation methods can be considered. For example, products and services can be recommended to customers in descending order of marketing data.

  The preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious for those skilled in the art that various modifications or modifications can be conceived within the scope of the idea described in the claims, and these naturally belong to the technical scope of the present invention. It is understood.

  The present invention is useful for marketing using, for example, customer product or service usage data.

DESCRIPTION OF SYMBOLS 1 Marketing support system 10 Communication part 11 Setting part 12 Classification part 13 Calculation part 14 Storage part 15 Output part N Network

Claims (18)

A marketing support method executed by a computer using usage data of a customer's product or service,
A setting step for identifying and setting a customer's action for starting or stopping the use of the product or service, or a change in the state of the customer regarding the use of the product or service;
A classification step of classifying customers into segments according to the presence or absence of the action or state change in the determination period;
A calculation step for calculating a total value based on usage data for a total period for each segment;
The starting point of the determination period, Ri same der with the start point of the previous or the aggregation period from the start of the counting period,
In the calculation step, for the segment classified as having the action or the state change in the classification step, the expected value that is the aggregate value is calculated,
The aggregate value is a cumulative value of utilization data in the aggregation period, or a difference value or an increase rate between the utilization data in the aggregation period and the utilization data in the period of the same length as the aggregation period before the aggregation period. A marketing support method characterized by being. A marketing support method executed by a computer using usage data of a customer's product or service,
A setting step for identifying and setting a customer's action for starting or stopping the use of the product or service, or a change in the state of the customer regarding the use of the product or service;
A classification step of classifying customers into segments according to the presence or absence of the action or state change in the determination period;
A calculation step for calculating a total value based on usage data for a total period for each segment;
The starting point of the determination period is the same as the starting point of the counting period before or the starting point of the counting period,
In the calculation step, from the aggregate value of the segment classified as having the action or the state change in the classification step, the aggregate value of the segment classified as having no action or the state change in the classification step. Subtract to calculate the boost value,
The aggregate value is a cumulative value of utilization data in the aggregation period, or a difference value or an increase rate between the utilization data in the aggregation period and the utilization data in the period of the same length as the aggregation period before the aggregation period. A marketing support method characterized by being. In the calculation step,
The aggregate value of the segments classified as having undergone the action in the classification step;
For all customers, calculate an action rate indicating the percentage of customers who performed the action during the determination period, and
Based on the total value and the action rate, further characterized by calculating the expected value, marketing support method according to claim 1 or 2. In the calculation step,
The aggregated values of the segments classified as having undergone the action and having undergone the state change in the classification step;
For all customers, an action rate indicating the percentage of customers who performed the action during the determination period;
For a customer who has performed the action during the determination period, a state change rate indicating a percentage of customers in which the state change has occurred is calculated,
The aggregate value, based on the action rate and the state change rate, further characterized by calculating the expected value, marketing support method according to claim 1 or 2. In the classification step, customers are classified into segments based on the presence or absence of the action that has responded to the use promotion activity during the determination period.
In the calculation step,
The aggregated values of the segments classified as having been subjected to the action in response to the use promotion activity in the classification step;
For all customers, calculate a response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period, and
On the basis of the reaction rate and the total value, further characterized by calculating the expected value, marketing support method according to claim 1 or 2. In the classification step, the customer is classified into segments according to the presence or absence of the action and the presence or absence of the state change in response to the use promotion activity in the determination period,
In the calculation step,
In the classification step, the aggregate value of the segments classified as having undergone the action in response to the use promotion activity and having undergone the state change;
For all customers, a response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period;
For a customer who has performed the action in response to the use promotion activity during the determination period, a state change rate indicating a ratio of customers in which the state change has occurred is calculated,
The total value, on the basis of the reaction rate and the state change rate, further characterized by calculating the expected value, marketing support method according to claim 1 or 2. The calculation step is performed by decomposing and calculating the total value into a direct total value directly resulting from the use promotion activity and a spread total value propagated by the use promotion activity. The marketing support method according to any one of 1 to 6 . The aggregation period is a period set for each customer in the segment classified as having the action or the state change in the classification step, and having a predetermined length starting from the action or the state change. and characterized in that, marketing support method according to any one of claims 1-7. The program which operate | moves on the computer which controls the said marketing support system so that the marketing support method as described in any one of Claims 1-8 may be performed by a marketing support system. A readable computer storage medium storing the program according to claim 9 . A marketing support system using usage data of customer products or services,
A setting unit for specifying and setting a customer's action for starting or stopping the use of a product or service, or a change in the state of the customer regarding the use of the product or service;
A classification unit that classifies customers into segments according to the presence or absence of the action or the state change in the determination period;
A calculation unit that calculates a total value based on usage data for a total period for each of the segments;
The starting point of the determination period, Ri same der with the start point of the previous or the aggregation period from the start of the counting period,
The calculation unit calculates an expected value that is the aggregate value for a segment classified as having the action or the state change in the classification unit,
The aggregate value is a cumulative value of utilization data in the aggregation period, or a difference value or an increase rate between the utilization data in the aggregation period and the utilization data in the period of the same length as the aggregation period before the aggregation period. A marketing support system characterized by being. A marketing support system using usage data of customer products or services,
A setting unit for specifying and setting a customer's action for starting or stopping the use of a product or service, or a change in the state of the customer regarding the use of the product or service;
A classification unit that classifies customers into segments according to the presence or absence of the action or the state change in the determination period;
A calculation unit that calculates a total value based on usage data for a total period for each of the segments;
The starting point of the determination period is the same as the starting point of the counting period before or the starting point of the counting period,
The calculation unit obtains the total value of the segment classified as having no action or the state change in the classification unit from the total value of the segment classified as having the action or the state change in the classification unit. Subtract to calculate the boost value,
The aggregate value is a cumulative value of utilization data in the aggregation period, or a difference value or an increase rate between the utilization data in the aggregation period and the utilization data in the period of the same length as the aggregation period before the aggregation period. A marketing support system characterized by being. The calculation unit includes:
The aggregated values of the segments classified as having undergone the action in the classification unit;
For all customers, calculate an action rate indicating the percentage of customers who performed the action during the determination period, and
The marketing support system according to claim 11 or 12 , wherein an expected value is further calculated based on the total value and the action rate. The calculation unit includes:
The aggregated values of the segments classified as having undergone the action and having undergone the state change in the classification unit;
For all customers, an action rate indicating the percentage of customers who performed the action during the determination period;
For a customer who has performed the action during the determination period, a state change rate indicating a percentage of customers in which the state change has occurred is calculated,
The marketing support system according to claim 11 or 12 , wherein an expected value is further calculated based on the total value, the action rate, and the state change rate. The classification unit classifies customers into segments according to the presence or absence of the action that has reacted to the use promotion activity during the determination period,
The calculation unit includes:
The aggregate value of the segments classified as having been subjected to the action in response to the use promotion activity in the classification unit;
For all customers, calculate a response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period, and
The marketing support system according to claim 11 or 12 , wherein an expected value is further calculated based on the total value and the reaction rate. The classification unit classifies customers into segments according to the presence or absence of the action and the presence or absence of the state change in response to the use promotion activity during the determination period,
In the calculation unit,
The aggregate value of the segments classified as having undergone the action and having undergone the state change in response to the use promotion activity in the classification unit;
For all customers, a response rate indicating the percentage of customers who took the action in response to the use promotion activity during the determination period;
For a customer who has performed the action in response to the use promotion activity during the determination period, a state change rate indicating a ratio of customers in which the state change has occurred is calculated,
The marketing support system according to claim 11 or 12 , wherein an expected value is further calculated based on the total value, the reaction rate, and the state change rate. The calculation unit is configured to calculate the aggregated value by decomposing the aggregated value into a directly aggregated value directly resulting from the use promotion activity and a spread aggregate value propagated by the use promotion activity. The marketing support system according to any one of 11 to 16 . The aggregation period is a period set for each customer in the segment classified as having the action or the state change in the classification unit, and having a predetermined length starting from the action or the state change. The marketing support system according to any one of claims 11 to 17 , wherein the marketing support system is any one of the above.

Images