Pharmacy high-value customer identification method and device

Abstract

The invention discloses a drugstore high-value customer identification method and device. The method comprises the following steps: 1) acquiring customer historical order data in a target drugstore; 2) performing desensitization processing on the data; 3) calculating a customer value evaluation index according to customer historical order data; 4) storing the obtained index data of the three dimensions into a database; 5) grouping the index data of the three dimensions by using a set box dividing strategy, and setting a label marking value of each box after grouping; circulating each binning strategy in sequence, and performing combination according to a grouping result after binning of the index data of each dimension to obtain a set number of binning combinations; 6) selecting an optimal box separation strategy; and 7) converting the index data of the three dimensions into discrete features, and starting to divide the index data of each dimension of each customer into high and low categories by using a preferable binning strategy to obtain a customer value identification result. According to the invention, an effective drugstore high-value customer identification model is established.

Description

Technical Field

[0001] The present invention relates to pharmaceutical big data technology, and in particular to a method and device for identifying high-value customers of a pharmacy. Background Art

[0002] Customer value identification is one of the cornerstones of pharmacies' customer retention efforts. Reasonable and efficient customer value identification can improve efficiency and accuracy during normal pharmacy operations, making it crucial for pharmacies. Establishing a customer value identification model and corresponding customer value evaluation methods can quantify customer value identification, helping companies identify their most valuable customer groups for targeted marketing campaigns. Developing an effective identification model is a pressing issue. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and device for identifying high-value customers of pharmacies in view of the defects in the existing technology.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a method for identifying high-value customers of a pharmacy, comprising the following steps:

[0005] 1) Obtaining historical order data of customers from the target pharmacy; the order data includes the customer's membership ID, order time, and order amount;

[0006] 2) Desensitize the data;

[0007] 3) Calculating customer value evaluation indicators based on the customer's historical order data; the customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount;

[0008] 4) Storing the obtained indicator data of the three dimensions into the database;

[0009] 5) Group the indicator data of the three dimensions using the set binning strategy and set the label value of each bin after grouping;

[0010] Each binning strategy is looped in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations;

[0011] 6) Perform statistics on each combination to obtain the classification ratio under each combination, and select the binning combination strategy with the most balanced classification ratio as the preferred binning strategy;

[0012] 7) Convert the three-dimensional indicator data into discrete features, and use the optimal binning strategy to begin dividing the indicator data of each dimension of each customer into high and low categories. Evaluate the customer value level based on the combination of high and low of each customer's three dimensions as the customer value identification result.

[0013] According to the above scheme, in step 3), the customer value evaluation index is calculated as follows:

[0014] 3.1) Obtain the maximum value of each customer's order date based on their historical order data, combine it with the current date and time, and subtract the two to extract the number of days as the value of each customer's most recent purchase time;

[0015] 3.2) Based on the customer's historical order data, obtain the total number of purchases of each customer as the customer's consumption frequency;

[0016] 3.3) Based on the customer's historical order data, obtain the total consumption amount of each customer in the last 12 months.

[0017] According to the above scheme, in step 3), the customer value evaluation index is calculated as follows:

[0018] 3.1) Obtain the maximum value of each customer's order date based on their historical order data, combine it with the current date and time, and subtract the two to extract the number of days as the value of each customer's most recent purchase time;

[0019] 3.2) Based on the customer's historical order data, obtain the number of purchases made by each customer in the last 12 months as the customer's consumption frequency;

[0020] 3.3) Based on the customer's historical order data, obtain the total amount of each customer's spending in the last 12 months;

[0021] According to the above scheme, in step 5), the binning strategy is set to sort the data in ascending order, then bin the data according to the set number of quantiles, and set a label value for each box, which is a positive or negative number.

[0022] According to the above scheme, in step 7), a positive tag value is defined as a high tag; a negative tag value is defined as a low tag, and the indicator data of the three dimensions are converted into discrete features. The selected binning strategy is used to divide each dimension into high and low categories, and the customer value level is evaluated based on the combination of the high and low values ​​of the three dimensions of each customer as the customer value identification result.

[0023] A high-value customer identification device for a pharmacy, comprising:

[0024] The data collection module is used to obtain the historical order data of customers in the target pharmacy; the order data includes the customer's membership ID, order consumption time and consumption amount;

[0025] Data processing module, used for desensitizing data;

[0026] A customer value evaluation index calculation module is used to calculate customer value evaluation indicators based on customer historical order data; the customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount;

[0027] The customer value identification module is used to identify customer value results based on the integration of customer value evaluation indicators; the details are as follows:

[0028] 1) Store the obtained indicator data of the three dimensions into the database;

[0029] 2) Group the indicator data of the three dimensions using the set binning strategy and set the label value of each bin after grouping;

[0030] Each binning strategy is looped in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations;

[0031] 3) Perform statistics on each combination to obtain the classification ratio under each combination, and select the binning combination strategy with the most balanced classification ratio as the preferred binning strategy;

[0032] 4) Convert the three-dimensional indicator data into discrete features, and use the optimal binning strategy to begin dividing the indicator data of each dimension of each customer into high and low categories. Evaluate the customer value level based on the combination of high and low values ​​of the three dimensions of each customer as the customer value identification result.

[0033] According to the above solution, in the customer value evaluation index calculation module, the customer value evaluation index is calculated as follows:

[0034] 1) Obtain the maximum value of each customer's order date based on each customer's historical order data, combine it with the current date and time, subtract the two to extract the number of days, and use it as the value of each customer's most recent consumption time;

[0035] 2) Based on the customer's historical order data, obtain the number of purchases made by each customer in the last 12 months as the customer's consumption frequency;

[0036] 3) Based on the customer's historical order data, obtain the total amount of each customer's consumption in the last 12 months;

[0037] According to the above scheme, in step 2), the binning strategy is set to sort the data in ascending order, then bin the data according to the set number of quantiles, and set a label value for each box, which is a positive or negative number.

[0038] According to the above scheme, in step 4), a positive tag value is defined as a high tag; a negative tag value is defined as a low tag, and the indicator data of the three dimensions are converted into discrete features. The selected binning strategy is used to divide each dimension into high and low categories, and the customer value level is evaluated based on the combination of the high and low levels of the three dimensions of each customer as the customer value identification result.

[0039] The beneficial effects produced by the present invention are:

[0040] The present invention establishes an effective high-value customer identification model for pharmacies, which can quickly evaluate the value of pharmacy customers based on pharmacy order data, and improve the data basis for pharmacies to carry out targeted marketing activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0042] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0043] Figure 2 Schematic diagram of the binning strategy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0045] like Figure 1 As shown, a method for identifying high-value customers of a pharmacy includes the following steps:

[0046] 1) Obtaining historical order data of customers from the target pharmacy; the order data includes the customer's membership ID, order time, and order amount;

[0047] 2) Desensitize the data; if the data contains abnormal characters, it needs to be cleaned;

[0048] 3) Calculating customer value evaluation indicators based on the customer's historical order data; the customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount;

[0049] The calculation of customer value evaluation indicators is as follows:

[0050] 3.1) Obtain the maximum value of each customer's order date based on their historical order data, combine it with the current date and time, and subtract the two to extract the number of days as the value of each customer's most recent purchase time;

[0051] 3.2) Based on the customer's historical order data, obtain the total number of purchases of each customer as the customer's consumption frequency;

[0052] 3.3) Based on the customer's historical order data, obtain the total consumption amount of each customer in the last 12 months.

[0053] As a preference, we use the number of times each customer has consumed in the last 12 months as the customer's consumption frequency; the total amount of consumption of each customer in the last 12 months; and replace the above values.

[0054] 4) Storing the obtained indicator data of the three dimensions into the database;

[0055] 5) Group the indicator data of the three dimensions using the set binning strategy and set the label value of each bin after grouping;

[0056] The set binning strategy is to sort the data in ascending order, then bin it according to the set number of quantiles, and set the label value of each box, which can be positive or negative.

[0057] Each binning strategy is looped in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations;

[0058] like Figure 2 As shown in the figure, our preset binning strategies include: dividing the sorted data into five equal parts, with the first four parts marked as negative and the last part marked as positive; dividing the sorted data into five equal parts, with the first three parts marked as negative and the last two parts marked as positive; dividing the sorted data into five equal parts, with the first two parts marked as negative and the last three parts marked as positive; dividing the sorted data into three equal parts, with the first two parts marked as negative and the last part marked as positive; dividing the sorted data into two equal parts, with the first part marked as negative and the last part marked as positive. Finally, 125 grouping combinations were obtained.

[0059] 6) Perform statistics on each combination to obtain the classification ratio under each combination, and select the binning combination strategy with the most balanced classification ratio as the preferred binning strategy;

[0060] 7) Define the label value as high if it is a positive number; define the label value as low if it is a negative number. Convert the indicator data of the three dimensions into discrete features, and use the optimal binning strategy to divide the indicator data of each dimension of each customer into high and low categories. Evaluate the level of customer value based on the combination of high and low values ​​of the three dimensions of each customer as the customer value identification result.

[0061] Here, the customer value identification results are divided into 8 categories. Pharmacies can adopt corresponding marketing strategies based on the identification results:

[0062] (High, High, High): Identified as an important value customer; (High, High, Low): An average value customer; (High, Low, High): Identified as an important development customer; (Low, High, High): Identified as an important retention customer; (High, Low, Low): Identified as an average development customer; (Low, High, Low): Identified as an average retention customer; (Low, Low, High): Identified as an important retention customer; (Low, Low, Low): Identified as an average retention customer.

[0063] The present invention also provides a high-value customer identification device for a pharmacy, comprising:

[0064] The data collection module is used to obtain the historical order data of customers in the target pharmacy; the order data includes the customer's membership ID, order consumption time and consumption amount;

[0065] Data processing module, used for desensitizing data;

[0066] A customer value evaluation index calculation module is used to calculate customer value evaluation indicators based on customer historical order data; the customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount;

[0067] The calculation of customer value evaluation indicators is as follows:

[0068] Get the maximum value of each customer's order date based on their historical order data, combine it with the current date and time, and subtract the two to extract the number of days as the value of each customer's most recent consumption time;

[0069] Based on the customer's historical order data, obtain the number of purchases made by each customer in the last 12 months as the customer's consumption frequency;

[0070] Based on the customer's historical order data, obtain the total consumption amount of each customer in the last 12 months;

[0071] The customer value identification module is used to identify customer value results based on the integration of customer value evaluation indicators; the details are as follows:

[0072] 1) Store the obtained indicator data of the three dimensions into the database;

[0073] 2) Group the indicator data of the three dimensions using the set binning strategy and set the label value of each bin after grouping;

[0074] The set binning strategy is to sort the data in ascending order, then bin it according to the set number of quantiles, and set the label value of each box, which can be positive or negative.

[0075] Each binning strategy is looped in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations;

[0076] 3) Perform statistics on each combination to obtain the classification ratio under each combination, and select the binning combination strategy with the most balanced classification ratio as the preferred binning strategy;

[0077] 4) Define the positive label value as high and the negative label value as low. Convert the indicator data of the three dimensions into discrete features. Use the optimal binning strategy to divide the indicator data of each dimension of each customer into high and low categories. Evaluate the customer value level based on the combination of high and low values ​​of the three dimensions of each customer as the customer value identification result.

[0078] Here, the customer value identification results are divided into 8 categories. Pharmacies can adopt corresponding marketing strategies based on the identification results:

[0079] (High, High, High): Identified as an important value customer; (High, High, Low): An average value customer; (High, Low, High): Identified as an important development customer; (Low, High, High): Identified as an important retention customer; (High, Low, Low): Identified as an average development customer; (Low, High, Low): Identified as an average retention customer; (Low, Low, High): Identified as an important retention customer; (Low, Low, Low): Identified as an average retention customer.

[0080] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for identifying high-value customers in a pharmacy, characterized in that: The following steps are involved: 1) Obtaining historical order data of customers in the target pharmacy; the order data includes the customer's membership ID, order consumption time and consumption amount; 2) Desensitize the data; 3) Calculate customer value evaluation indicators based on customer historical order data; The customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount; 4) The obtained indicator data of the three dimensions are stored in the database; 5) Group the indicator data of the three dimensions using the set binning strategy, and set the label value of each bin after grouping; Each binning strategy is cycled in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations; 6) Perform statistics on each combination to obtain the classification ratio under each combination, and take the binning combination strategy with the most average classification ratio as the preferred binning strategy; 7) Convert the three-dimensional indicator data into discrete features, and use the optimal binning strategy to begin dividing the indicator data of each dimension of each customer into high and low categories. Evaluate the level of customer value based on the combination of the three dimensions of each customer as the customer value identification result.

2. The method for identifying high-value customers of drugstores according to claim 1, characterized in that: In step 3), the calculation of the customer value evaluation index is as follows: 3.1) Obtain the maximum value of each customer's order date based on each customer's historical order data, combine it with the current date and time, subtract the two to extract the number of days, and use it as the value of each customer's most recent consumption time; 3.2) Based on the customer's historical order data, obtain the total number of consumption times of each customer as the customer's consumption frequency; 3.3) Based on the customer's historical order data, obtain the total consumption amount of each customer in the last 12 months.

3. The method for identifying high-value customers of drugstores according to claim 1, characterized in that: In step 3), the calculation of the customer value evaluation index is as follows: 3.1) Obtain the maximum value of each customer's order date based on each customer's historical order data, combine it with the current date and time, subtract the two to extract the number of days, and use it as the value of each customer's most recent consumption time; 3.2) Based on the customer's historical order data, obtain the number of times each customer has consumed in the last 12 months as the customer's consumption frequency; 3.3) Based on the customer's historical order data, obtain the total consumption amount of each customer in the last 12 months.

4. The method for identifying high-value customers of drug stores according to claim 1, characterized in that: In the step 5), the set binning strategy is to sort the data in ascending order, then bin the data according to the set number of quantiles, and set a label value for each box, where the label value is a positive or negative number.

5. The method for identifying high-value customers of drug stores according to claim 4, characterized in that: In step 5), in step 7), a positive label value is defined as a label high; Define the label value as negative as low, transform the indicator data of the three dimensions into discrete features, and use the selected binning strategy to divide each dimension into high and low categories. Evaluate the level of customer value based on the combination of high and low values ​​of the three dimensions of each customer as the customer value identification result.

6. A high-value customer identification device for a pharmacy, characterized in that: include: A data collection module is used to obtain historical order data of customers in the target pharmacy; the order data includes the customer's member ID, order consumption time and consumption amount; Data processing module, used for desensitizing data; A customer value evaluation index calculation module is used to calculate the customer value evaluation index based on the customer's historical order data; The customer value evaluation indicators include: the customer's most recent consumption time, the customer's consumption frequency, and the customer's consumption amount; The customer value identification module is used to identify the customer value results based on the integration of customer value evaluation indicators; the details are as follows: 1) Store the obtained indicator data of the three dimensions into the database; 2) Group the indicator data of the three dimensions using the set binning strategy, and set the label value of each bin after grouping; Each binning strategy is cycled in turn, and the grouping results after binning the indicator data of each dimension are combined to obtain a set number of binning combinations; 3) Perform statistics on each combination to obtain the classification ratio under each combination, and take the binning combination strategy with the most average classification ratio as the preferred binning strategy; 4) Convert the indicator data of the three dimensions into discrete features, and use the optimal binning strategy to begin dividing the indicator data of each dimension of each customer into high and low categories. Evaluate the level of customer value based on the combination of the high and low levels of the three dimensions of each customer as the customer value identification result.

7. The high-value customer identification device for drug stores according to claim 6, characterized in that: The set binning strategy is to sort the data in ascending order, then bin it according to the set number of quantiles, and set the label value of each box, which can be positive or negative.

8. The high-value customer identification device for drug stores according to claim 7, characterized in that: Define the label value as high if it is a positive number; define the label value as low if it is a negative number. Convert the indicator data of the three dimensions into discrete features, and use the selected binning strategy to divide each dimension into high and low categories. Evaluate the level of customer value based on the combination of the three dimensions of each customer as the customer value identification result.

9. An electronic device, characterized in that: include: one or more processors; as well as a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the method according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images