CN115689128B

CN115689128B - Customer data analysis method and system based on CRM

Info

Publication number: CN115689128B
Application number: CN202211716541.5A
Authority: CN
Inventors: 陈守红
Original assignee: Shenzhen Gelonghui Information Technology Co ltd
Current assignee: Shenzhen Gelonghui Information Technology Co ltd
Priority date: 2022-12-30
Filing date: 2022-12-30
Publication date: 2023-08-22
Anticipated expiration: 2042-12-30
Also published as: CN115689128A

Abstract

The invention relates to the field of big data, and discloses a customer data analysis method and system based on CRM (customer relationship management), which are used for improving the accuracy of customer data analysis. The method comprises the following steps: acquiring historical sales data, analyzing the rate of return of the historical sales data, and determining a target rate of return; matching the intention clients through the target rate of return, and determining an intention client set; generating a client portrait for each intention client in the intention client set to obtain a client portrait corresponding to each intention client in the intention client set; carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; classifying a plurality of intention clients through target association relations among the intention clients to obtain at least one client group; and generating a management policy for at least one customer group to obtain a customer management policy, and transmitting the customer management policy to a preset customer information management terminal.

Description

Customer data analysis method and system based on CRM

Technical Field

The invention relates to the field of big data, in particular to a customer data analysis method and system based on CRM.

Background

With the high-speed development of internet technology, an intelligent and automatic analysis scheme can be provided for application scenes of different client data at present, and the efficiency of client maintenance can be improved when auxiliary personnel realize client maintenance and welfare batch delivery.

The large amount of customer data that is typically collected in existing schemes, but without the ability to analyze the data based on the large data, does not understand the customer requirements and preferences well, and does not optimize customer relationship management based on such information, resulting in lower accuracy in analyzing the customer data.

Disclosure of Invention

The invention provides a customer data analysis method and a customer data analysis system based on CRM, which are used for improving the accuracy of customer data analysis.

The first aspect of the present invention provides a CRM-based customer data analysis method, which comprises: acquiring historical sales data, analyzing the historical sales data in a rate of return, and determining a target rate of return; and carrying out the intention client matching through the target rate of return, and determining an intention client set, wherein the intention client set comprises: a plurality of intent clients; generating a customer portrait for the intention customer set to obtain a customer portrait corresponding to each intention customer in the intention customer set; carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; classifying the clients of the plurality of intention clients through the target association relationship among the intention clients to obtain at least one client group; and generating the management policy of the at least one customer group to obtain a customer management policy, and transmitting the customer management policy to a preset customer information management terminal.

With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the obtaining historical sales data and performing a rate of return analysis on the historical sales data, to determine a target rate of return includes: inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data; respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data; and carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return.

With reference to the first aspect, in a second implementation manner of the first aspect of the present invention, the matching of the intent clients by the target rate of return determines an intent client set, where the intent client set includes: a plurality of intent clients, comprising: performing data tag matching on the target return rate to obtain a plurality of data tags; performing client matching through the plurality of data tags to obtain an intention client set, wherein the intention client set comprises: a plurality of intent clients.

With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the generating a customer portrait for the intent client set to obtain a customer portrait corresponding to each intent client in the intent client set includes: collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client; carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client; and generating the portrait by the preference characteristics corresponding to each intention client to obtain the client portrait corresponding to each intention client in the intention client set.

With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, the performing association analysis on the client portrait corresponding to each intended client, and determining the target association between the intended clients includes: carrying out associated index analysis on the customer portrait corresponding to each intention customer to obtain a corresponding associated characteristic index; and constructing an association relation through the association characteristic index to obtain a target association relation between the intention clients.

With reference to the first aspect, in a fifth implementation manner of the first aspect of the present invention, the generating a management policy for the at least one client group to obtain a client management policy, and transmitting the client management policy to a preset client information management terminal includes: performing cluster analysis on the client data of the at least one client group to obtain target cluster data; performing identification matching through the target cluster data, and determining a strategy identification corresponding to the target cluster data; generating a management policy through the policy identifier to obtain a client management policy; and transmitting the client management policy to a preset client information management terminal.

With reference to the first aspect, in a sixth implementation manner of the first aspect of the present invention, the performing cluster analysis on the client data of the at least one client group to obtain target cluster data includes: performing space point mapping on the client data of the at least one client group to obtain a corresponding space point set; performing cluster point analysis on the space point set to determine target cluster points; and carrying out cluster analysis on the client data of the at least one client group through the target cluster points to obtain target cluster data.

A second aspect of the present invention provides a CRM-based customer data analysis system comprising: the acquisition module is used for acquiring historical sales data, analyzing the return rate of the historical sales data and determining a target return rate; the matching module is used for matching the intention clients through the target rate of return and determining an intention client set, wherein the intention client set comprises: a plurality of intent clients; the generation module is used for generating the customer portrait of the intention customer set to obtain the customer portrait corresponding to each intention customer in the intention customer set; the analysis module is used for carrying out association relation analysis on the customer portraits corresponding to each intention customer and determining target association relation among the intention customers; the classification module is used for classifying the clients of the plurality of intention clients through the target association relation among the intention clients to obtain at least one client group; and the generation module is used for generating the management policy of the at least one customer group to obtain a customer management policy, and transmitting the customer management policy to a preset customer information management terminal.

With reference to the second aspect, in a first implementation manner of the second aspect of the present invention, the acquiring module is specifically configured to: inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data; respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data; and carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return.

With reference to the second aspect, in a second implementation manner of the second aspect of the present invention, the matching module is specifically configured to: performing data tag matching on the target return rate to obtain a plurality of data tags; performing client matching through the plurality of data tags to obtain an intention client set, wherein the intention client set comprises: a plurality of intent clients.

With reference to the second aspect, in a third implementation manner of the second aspect of the present invention, the generating module is specifically configured to: collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client; carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client; and generating the portrait by the preference characteristics corresponding to each intention client to obtain the client portrait corresponding to each intention client in the intention client set.

With reference to the second aspect, in a fourth implementation manner of the second aspect of the present invention, the analysis module is specifically configured to: carrying out associated index analysis on the customer portrait corresponding to each intention customer to obtain a corresponding associated characteristic index; and constructing an association relation through the association characteristic index to obtain a target association relation between the intention clients.

With reference to the second aspect, in a fifth implementation manner of the second aspect of the present invention, the classification module further includes: the analysis unit is used for carrying out cluster analysis on the client data of the at least one client group to obtain target cluster data; the matching unit is used for carrying out identification matching through the target clustering data and determining a strategy identification corresponding to the target clustering data; the generation unit is used for generating a management policy through the policy identification to obtain a client management policy; and the transmission unit is used for transmitting the client management strategy to a preset client information management terminal.

With reference to the second aspect, in a sixth implementation manner of the second aspect of the present invention, the analysis unit is specifically configured to: performing space point mapping on the client data of the at least one client group to obtain a corresponding space point set; performing cluster point analysis on the space point set to determine target cluster points; and carrying out cluster analysis on the client data of the at least one client group through the target cluster points to obtain target cluster data.

According to the technical scheme provided by the invention, historical sales data are obtained, and the historical sales data are subjected to rate of return analysis to determine the target rate of return; matching the intention clients through the target rate of return, and determining an intention client set; generating a client portrait for each intention client in the intention client set to obtain a client portrait corresponding to each intention client in the intention client set; carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; classifying a plurality of intention clients through target association relations among the intention clients to obtain at least one client group; the invention carries out report rate analysis on historical sales data to further match a plurality of intention clients, then carries out client group classification according to the client portraits of the intention clients to further generate the client management policy of each client group.

Drawings

FIG. 1 is a schematic diagram of one embodiment of a CRM-based customer data analysis method in accordance with an embodiment of the present invention;

FIG. 2 is a flow chart of data tag matching and customer matching in an embodiment of the present invention;

FIG. 3 is a flow chart of preference feature analysis and representation generation in an embodiment of the present invention;

FIG. 4 is a flowchart of management policy generation in an embodiment of the present invention;

FIG. 5 is a schematic diagram of one embodiment of a CRM-based customer data analysis system in accordance with an embodiment of the invention;

FIG. 6 is a schematic diagram of another embodiment of a CRM-based customer data analysis system in accordance with an embodiment of the invention.

Detailed Description

The embodiment of the invention provides a customer data analysis method and a customer data analysis system based on CRM, which are used for improving the accuracy of customer data analysis. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

For ease of understanding, a specific flow of an embodiment of the present invention is described below with reference to fig. 1, where an embodiment of a CRM-based customer data analysis method according to an embodiment of the present invention includes:

s101, acquiring historical sales data, analyzing the rate of return of the historical sales data, and determining a target rate of return;

it will be appreciated that the execution subject of the present invention may be a CRM-based customer data analysis system, or may be a terminal or a server, and is not limited in this particular context. The embodiment of the invention is described by taking a server as an execution main body as an example.

It should be noted that the customer management system (CRM, customer Relationship Management) mainly provides functions of management and analysis of customer basic information, credit analysis and risk monitoring, benefit and business analysis, and personalized service, and may collect sales data generated during the process of the user through the preset application program to perform personalized analysis on the user.

Where the rate of return refers to the economic return of the business from the investment of an investment business, such as the ratio between the sales return and the cost, in this embodiment the rate of return generated by the business during the sales of the financial product. Specifically, the server determines a target rate of return by determining historical sales data of enterprise financial products and analyzing the rate of return of the historical sales data, specifically by acquiring historical conversion rates corresponding to various purchase data in the historical sales data, wherein each purchase data corresponds to a plurality of historical financial products respectively; and determining the target return rates corresponding to the purchase data respectively according to the historical conversion rates of the purchase data, and determining the target financial products corresponding to the purchase data respectively according to the target return rates.

S102, matching the intention clients through the target rate of return, and determining an intention client set, wherein the intention client set comprises: a plurality of intent clients;

specifically, the server acquires webpage browsing information of a client, wherein the webpage browsing information comprises personal information of the client and user demand information; predicting and matching the web browsing information according to a preset matching algorithm to obtain a matching result representing the intention of the client; traversing a preset financial product information base according to the matching result to acquire financial product information meeting the client intent, taking clients corresponding to the financial product information meeting the client intent as intent clients, and performing set conversion on the intent clients to generate an intent client set.

S103, generating client portraits for the intention client set to obtain client portraits corresponding to each intention client in the intention client set;

specifically, personal information and user demand information corresponding to each intention client in the intention client set and historical transaction data corresponding to the intention client are obtained; wherein, the personal information at least comprises mobile phone number, age, sex, etc.; then, according to the personal information and the user demand information corresponding to each intention client, extracting online behavior data corresponding to each intention client in the intention client set from the CRM system; then, according to historical transaction data corresponding to the intention clients, determining offline behavior data corresponding to the intention clients; and finally, combining the offline behavior data and the online behavior data corresponding to the intention clients to form client association data, and constructing client portraits corresponding to the intention clients to obtain the client portraits corresponding to each intention client in the intention client set.

S104, carrying out association relation analysis on the customer portraits corresponding to each intention customer, and determining target association relation among the intention customers;

specifically, the server extracts the association indexes of the customer portraits corresponding to each intention customer to obtain the association indexes corresponding to each intention customer, then constructs an association index system, constructs the mutual connection between the association indexes corresponding to each intention customer, endows the association indexes with quantitative values of each association characteristic, and determines the association degree suffered by the association indexes corresponding to each intention customer according to the quantitative values of each association characteristic to obtain the index association degree corresponding to each intention customer; judging whether the index association degree exceeds a preset target value or not to obtain a judging result; and constructing the association relation of the association characteristic indexes according to the judging result to obtain the target association relation between the intention clients.

S105, classifying the clients of the plurality of intention clients through target association relations among the intention clients to obtain at least one client group;

specifically, the server creates a client grouping model corresponding to a plurality of intention clients according to the target association relationship among the intention clients. Generating a relation cross distribution diagram according to target association relations among a plurality of intention clients; generating a customer group distribution of a plurality of intended customers according to the relational cross distribution map; and carrying out customer group division on the plurality of intention customers based on the customer group distribution to obtain at least one customer group. Further, analyzing feature points of target association relations among the intention clients based on the association model to obtain a plurality of association relation nodes; respectively calculating the distribution weights of a plurality of association nodes to obtain the distribution weight corresponding to each association node; and carrying out client group division on the plurality of intention clients according to the distribution weight corresponding to each association node to obtain at least one client group.

S106, generating management policies of at least one customer group to obtain customer management policies, and transmitting the customer management policies to a preset customer information management terminal.

Specifically, cluster analysis is performed on the client data of at least one client group, specifically, the method includes inputting the client data of at least one client group into a preset cluster model, and performing feature clustering on the client data of at least one client group through the cluster model to obtain target cluster data corresponding to the client data, including: inputting customer data for at least one customer group into a preset clustering model; clustering the client data of at least one client group through a clustering model to obtain a plurality of characteristic data clusters; and obtaining a clustering center according to the plurality of characteristic data clusters, and generating target clustering data corresponding to the client data according to the clustering center. Then, carrying out identification matching on the target clustering data through the target clustering data, and determining a strategy identification corresponding to the target clustering data; generating a management policy through the policy identification to obtain a client management policy; and transmitting the client management policy to a preset client information management terminal. In addition, in the embodiment, the clustering center and the space point of the client data are extracted; respectively calculating Euclidean distances between the space points and the clustering center to obtain a target Euclidean distance corresponding to each space point; generating target cluster points according to the target Euclidean distance corresponding to each space point, and carrying out cluster analysis on the client data of at least one client group through the target cluster points to obtain target cluster data.

In the embodiment of the invention, historical sales data is obtained, and the historical sales data is subjected to rate of return analysis to determine the target rate of return; matching the intention clients through the target rate of return, and determining an intention client set; generating a client portrait for each intention client in the intention client set to obtain a client portrait corresponding to each intention client in the intention client set; carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; classifying a plurality of intention clients through target association relations among the intention clients to obtain at least one client group; the invention carries out report rate analysis on historical sales data to further match a plurality of intention clients, then carries out client group classification according to the client portraits of the intention clients to further generate the client management policy of each client group.

In a specific embodiment, the process of executing step S101 may specifically include the following steps:

(1) Inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data;

(2) Respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data;

(3) And carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return.

Specifically, the server queries historical sales data from a preset CRM system, performs data grouping on the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data, divides the historical total sales time into a plurality of time intervals, then extracts sales data corresponding to each time interval to obtain a plurality of groups of sales data, and then utilizes a pre-established return rate analysis model set which at least comprises a conversion rate model and a return rate calculation model, wherein the conversion rate model and the return rate calculation model are trained deep learning models, the weights of the conversion rate model and the return rate calculation model are determined according to each group of sales data, the return rate corresponding to each group of sales data is calculated according to the weights, and then the return rate corresponding to each group of sales data is weighted and calculated to obtain normalized return rate fusion data, namely target return rate.

In a specific embodiment, as shown in fig. 2, the process of executing step S102 may specifically include the following steps:

s201, carrying out data tag matching on the target return rate to obtain a plurality of data tags;

s202, performing client matching through a plurality of data labels to obtain an intention client set, wherein the intention client set comprises: a plurality of intent clients.

Specifically, the server acquires a data tag list, calculates an analysis coefficient of each candidate data tag in the data tag list, determines that the candidate data tag is a data tag required by the embodiment if the analysis coefficient is greater than or equal to a preset analysis coefficient threshold value, traverses the data tag list to obtain a plurality of data tags, and sends the plurality of data tags to the cloud data platform; if the analysis coefficient is smaller than a preset analysis coefficient threshold value, determining that the candidate data tag is not the data tag required by the embodiment, finally matching a plurality of data tags generated after traversal with the candidate client cluster, taking the matched clients as the intention clients, obtaining a plurality of intention clients, and generating an intention client set according to the intention clients.

In a specific embodiment, as shown in fig. 3, the process of executing step S103 may specifically include the following steps:

S301, collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client;

s302, carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client;

s303, generating portraits through preference characteristics corresponding to each intention client to obtain client portraits corresponding to each intention client in the intention client set.

Specifically, the server acquires personal information and user demand information corresponding to each intention client in the intention client set and historical transaction data corresponding to the intention client; wherein, the personal information at least comprises mobile phone number, age, sex, etc.; then, according to the personal information and the user demand information corresponding to each intention client, extracting online behavior data corresponding to each intention client in the intention client set from the CRM system; then, according to historical transaction data corresponding to the intention clients, determining offline behavior data corresponding to the intention clients; finally, combining the offline behavior data and the online behavior data corresponding to the intention clients to form client association data, and constructing client portraits corresponding to the intention clients to obtain client portraits corresponding to each intention client in the intention client set; the method comprises the steps of carrying out preference feature analysis on client-associated data corresponding to each intention client to obtain preference features corresponding to each intention client, specifically, carrying out interest vector calculation on the client-associated data corresponding to each intention client by a server to obtain interest vectors corresponding to each intention client, carrying out distance value calculation on the interest vectors corresponding to each intention client and each standard preference feature in a preset database to obtain a plurality of distance values, carrying out calculation through a preset cosine similarity calculation formula to obtain the plurality of distance values, and taking preference corresponding to the standard preference feature meeting preset conditions as the preference feature corresponding to each intention client according to each distance value after the distance values are obtained by calculation; the method comprises the steps that through the preference characteristics corresponding to each intention client, client portraits corresponding to each intention client in an intention client set are obtained, specifically, a server carries out standard client portraits matching on the preference characteristics corresponding to each intention client, standard client portraits corresponding to each intention client are obtained, the standard client portraits obtained through matching are used as the client portraits corresponding to each intention client, and the client portraits corresponding to each intention client in the intention client set are obtained respectively.

In a specific embodiment, the process of executing step S104 may specifically include the following steps:

(1) Carrying out associated index analysis on the customer portrait corresponding to each intention customer to obtain a corresponding associated characteristic index;

(2) And constructing an association relation through the association characteristic indexes to obtain a target association relation between the intention clients.

Specifically, the server analyzes the association indexes of the customer portraits corresponding to each intended customer to obtain corresponding association characteristic indexes, specifically extracts the association indexes of the customer portraits corresponding to each intended customer to obtain the association indexes corresponding to each intended customer, then constructs an association index system, constructs the mutual association between the association indexes corresponding to each intended customer, and endows the quantitative value of each association characteristic, and determines the association degree suffered by the association indexes corresponding to each intended customer according to the quantitative value of each association characteristic to obtain the index association degree corresponding to each intended customer; judging whether the index association degree exceeds a preset target value or not to obtain a judging result; if the judgment result is exceeded, carrying out association relation construction through the association characteristic index to obtain the target association relation between the intention clients. Obtaining a dynamic segmentation result corresponding to the associated feature indexes, and constructing a mapping relation between the associated feature indexes; and on the basis of each dynamic segmentation result and the mapping relation, respectively carrying out association relation construction on the association characteristic indexes by a preset storage mode and a preset integer list storage mode to obtain target association relation among the intention clients.

In a specific embodiment, as shown in fig. 4, the process of performing step S105 may specifically include the following steps:

s401, carrying out cluster analysis on the client data of at least one client group to obtain target cluster data;

s402, performing identification matching through target cluster data, and determining a strategy identification corresponding to the target cluster data;

s403, generating a management policy through a policy identifier to obtain a client management policy;

s404, transmitting the client management policy to a preset client information management terminal.

Specifically, the server performs cluster analysis on the client data of at least one client group by inputting the client data of at least one client group into a preset cluster model; clustering the client data of at least one client group through a clustering model to obtain a plurality of characteristic data clusters; and obtaining a clustering center according to the plurality of characteristic data clusters, and generating target clustering data corresponding to the client data according to the clustering center. Then, carrying out identification matching on the target clustering data through the target clustering data, and determining a strategy identification corresponding to the target clustering data; generating a management policy through the policy identification to obtain a client management policy; and transmitting the client management policy to a preset client information management terminal. In addition, in the embodiment, the clustering center and the space point of the client data are extracted; respectively calculating Euclidean distances between the space points and the clustering center to obtain a target Euclidean distance corresponding to each space point; generating target cluster points according to the target Euclidean distance corresponding to each space point, and carrying out cluster analysis on the client data of at least one client group through the target cluster points to obtain target cluster data. And performing identification matching through the target cluster data, determining a strategy identification corresponding to the target cluster data, wherein a mapping corresponding relation is constructed in advance between the target cluster data and the strategy, the strategy identification is adopted for one-to-one correspondence, management strategy generation is performed through the strategy identification, a client management strategy is obtained, and finally the client management strategy is transmitted to a preset client information management terminal.

In a specific embodiment, the process of executing step S401 may specifically include the following steps:

(1) Performing space point mapping on the client data of at least one client group to obtain a corresponding space point set;

(2) Performing cluster point analysis on the space point set to determine target cluster points;

(3) And carrying out cluster analysis on the client data of at least one client group through the target cluster points to obtain target cluster data.

Specifically, the server acquires client data of each client group to be clustered, vectorizes each data information in the client data of at least one client group to form a vector data set, selects a plurality of vectors from the vector data set to serve as initial clustering centers respectively, clusters users according to the initial clustering centers to obtain a plurality of characteristic data clusters, updates the clustering centers according to clustered results or clustered results, and continuously clusters each user according to the clustering centers to generate target clustering data corresponding to the characteristic data. Specifically, a server extracts a clustering center and space points of client data; respectively calculating Euclidean distances between the space points and the clustering center to obtain a target Euclidean distance corresponding to each space point; generating a target cluster point according to the target Euclidean distance corresponding to each space point; judging whether the Euclidean distance of the target is larger than the target clustering point or not; if yes, determining the judgment result as a determination target cluster point. Carrying out cluster analysis on the client data of at least one client group through a target cluster point to obtain target cluster data, carrying out numerical preprocessing on the client data of at least one client group by a server according to the target cluster point to obtain data samples, carrying out dimension reduction and feature extraction on the data samples through an automatic encoder, obtaining a clustering result through the data processed by the automatic encoder, calculating the weight of attribute features of the data samples processed by the automatic encoder by a variation coefficient method, calculating the distance between the samples by a weighted Euclidean distance formula, calculating the average distance between all the data samples, traversing the data samples, searching for neighbor points of which the distance between each sample point and the neighbor point is smaller than the average distance, judging whether the Euclidean distance of the target is larger than the target cluster point, and outputting the target cluster data if the Euclidean distance is larger than the target cluster point.

The foregoing describes a method for analyzing customer data based on CRM in an embodiment of the present invention, and the following describes a customer data analysis system based on CRM in an embodiment of the present invention, referring to fig. 5, an embodiment of a customer data analysis system based on CRM in an embodiment of the present invention includes:

the acquiring module 501 is configured to acquire historical sales data, analyze a rate of return of the historical sales data, and determine a target rate of return;

the matching module 502 is configured to perform intent client matching according to the target rate of return, and determine an intent client set, where the intent client set includes: a plurality of intent clients;

a generating module 503, configured to generate a customer portrait for the intent client set, so as to obtain a customer portrait corresponding to each intent client in the intent client set;

the analysis module 504 is configured to perform association analysis on the customer portraits corresponding to each of the intention customers, and determine a target association relationship between the intention customers;

the classification module 505 is configured to classify the clients according to the target association relationship between the clients to obtain at least one client group;

and the generating module 506 is configured to generate a management policy for the at least one customer group, obtain a customer management policy, and transmit the customer management policy to a preset customer information management terminal.

Through the cooperation of the components, historical sales data is obtained, and the historical sales data is subjected to rate of return analysis to determine a target rate of return; matching the intention clients through the target rate of return, and determining an intention client set; generating a client portrait for each intention client in the intention client set to obtain a client portrait corresponding to each intention client in the intention client set; carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; classifying a plurality of intention clients through target association relations among the intention clients to obtain at least one client group; the invention carries out report rate analysis on historical sales data to further match a plurality of intention clients, then carries out client group classification according to the client portraits of the intention clients to further generate the client management policy of each client group.

Referring to FIG. 6, another embodiment of a CRM-based customer data analysis system in accordance with an embodiment of the present invention includes:

Optionally, the obtaining module 501 is specifically configured to:

inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data; respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data; and carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return.

Optionally, the matching module 502 is specifically configured to:

performing data tag matching on the target return rate to obtain a plurality of data tags; performing client matching through the plurality of data tags to obtain an intention client set, wherein the intention client set comprises: a plurality of intent clients.

Optionally, the generating module 503 is specifically configured to:

collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client; carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client; and generating the portrait by the preference characteristics corresponding to each intention client to obtain the client portrait corresponding to each intention client in the intention client set.

Optionally, the analysis module 504 is specifically configured to:

carrying out associated index analysis on the customer portrait corresponding to each intention customer to obtain a corresponding associated characteristic index; and constructing an association relation through the association characteristic index to obtain a target association relation between the intention clients.

Optionally, the classification module 506 further includes:

an analysis unit 5061, configured to perform cluster analysis on the client data of the at least one client group to obtain target cluster data;

a matching unit 5062, configured to perform identification matching through the target cluster data, and determine a policy identifier corresponding to the target cluster data;

a generating unit 5063, configured to generate a management policy according to the policy identifier, so as to obtain a client management policy;

and a transmission unit 5064 for transmitting the client management policy to a preset client information management terminal.

Optionally, the analysis unit 5061 is specifically configured to:

performing space point mapping on the client data of the at least one client group to obtain a corresponding space point set; performing cluster point analysis on the space point set to determine target cluster points; and carrying out cluster analysis on the client data of the at least one client group through the target cluster points to obtain target cluster data.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A CRM-based customer data analysis method, wherein the CRM-based customer data analysis method comprises:

acquiring historical sales data, analyzing the historical sales data in a rate of return, and determining a target rate of return, wherein the method specifically comprises the following steps: inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data; respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data; carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return; specifically, the server queries historical sales data from a preset CRM system, performs data grouping on the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data, divides the historical total sales time into a plurality of time intervals, extracts sales data corresponding to each time interval to obtain a plurality of groups of sales data, utilizes a pre-established return rate analysis model set, wherein the return rate analysis model set at least comprises a conversion rate model and a return rate calculation model, the conversion rate model and the return rate calculation model are both trained deep learning models, determines weights of the conversion rate model and the return rate calculation model according to each group of sales data, calculates return rates corresponding to each group of sales data according to the weights, and performs weighted calculation on return rates corresponding to each group of sales data to obtain normalized return rate fusion data, namely target return rate;

And carrying out the intention client matching through the target rate of return, and determining an intention client set, wherein the intention client set comprises: the plurality of intention clients specifically include: performing data tag matching on the target return rate to obtain a plurality of data tags; performing client matching through the plurality of data tags to obtain an intent client set; specifically, the server acquires a data tag list, calculates an analysis coefficient of each candidate data tag in the data tag list, determines that the candidate data tag is a required data tag if the analysis coefficient is greater than or equal to a preset analysis coefficient threshold value, traverses the data tag list to obtain a plurality of data tags, and sends the data tags to the cloud data platform; if the analysis coefficient is smaller than a preset analysis coefficient threshold value, determining that the candidate data tag is not a required data tag, matching a plurality of data tags generated after traversal is finished with the candidate client cluster, taking the matched clients as the intention clients, obtaining a plurality of intention clients, and generating an intention client set according to the intention clients;

generating a customer portrait for the intention customer set to obtain a customer portrait corresponding to each intention customer in the intention customer set; the method comprises the steps of collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client; carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client; generating a portrait by the preference characteristics corresponding to each intention client to obtain a client portrait corresponding to each intention client in the intention client set; the method comprises the steps of obtaining personal information and user demand information corresponding to each intention client in an intention client set and historical transaction data corresponding to the intention clients; the personal information at least comprises a mobile phone number, an age and a sex; according to the personal information and the user demand information corresponding to each intention client, extracting online behavior data corresponding to each intention client in the intention client set from the CRM system; according to historical transaction data corresponding to the intention clients, determining offline behavior data corresponding to the intention clients; combining the offline behavior data and the online behavior data corresponding to the intention clients to form client association data, and constructing client portraits corresponding to the intention clients to obtain client portraits corresponding to each intention client in the intention client set; performing preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client, performing interest vector calculation on the client associated data corresponding to each intention client to obtain interest vectors corresponding to each intention client, performing distance value calculation on the interest vectors corresponding to each intention client and each standard preference feature in a preset database to obtain a plurality of distance values, wherein the plurality of distance values are obtained through calculation according to a preset cosine similarity calculation formula, and taking preferences corresponding to standard preference features meeting preset conditions as the preference features corresponding to each intention client according to each distance value after the distance values are obtained through calculation; generating a customer portrait corresponding to each intention customer in the intention customer set by using the preference characteristics corresponding to each intention customer, performing standard customer portrait matching on the preference characteristics corresponding to each intention customer to obtain standard customer portrait corresponding to each intention customer, taking the standard customer portrait obtained by matching as the customer portrait corresponding to each intention customer, and respectively obtaining the customer portrait corresponding to each intention customer in the intention customer set;

Carrying out association relation analysis on the client images corresponding to each intention client, and determining target association relation among the intention clients; carrying out association index analysis on the customer portraits corresponding to each intended customer to obtain corresponding association characteristic indexes; building an association relation through the association characteristic indexes to obtain a target association relation between intention clients; extracting association indexes of the customer portraits corresponding to each intended customer to obtain association indexes corresponding to each intended customer, then constructing an association index system, constructing the interconnection among the association indexes corresponding to each intended customer, giving a quantification value of each association characteristic, and determining the association degree of the association indexes corresponding to each intended customer according to the quantification value of each association characteristic to obtain the index association degree corresponding to each intended customer; judging whether the index association degree exceeds a preset target value or not to obtain a judging result; if the judgment result is exceeded, carrying out association relation construction through the association characteristic index to obtain a target association relation between the intention clients; the method comprises the steps of obtaining a dynamic segmentation result corresponding to the associated characteristic indexes, and constructing a mapping relation between the associated characteristic indexes; based on each dynamic segmentation result and the mapping relation, respectively carrying out association relation construction on the association characteristic indexes by a preset storage mode and a preset integer list storage mode to obtain target association relation between intention clients;

The clients are classified according to the target association relationship among the clients to obtain at least one client group, which comprises the following steps: the server creates a client grouping model corresponding to a plurality of intention clients according to the target association relationship among the intention clients, and generates a relationship cross distribution diagram according to the target association relationship among the intention clients; generating a customer group distribution of a plurality of intended customers according to the relational cross distribution map; dividing a plurality of intention clients into client groups based on client group distribution to obtain at least one client group, and analyzing characteristic points of target association relations among the intention clients based on an association model to obtain a plurality of association relation nodes; respectively calculating the distribution weights of a plurality of association nodes to obtain the distribution weight corresponding to each association node; carrying out client group division on a plurality of intention clients according to the distribution weight corresponding to each association node to obtain at least one client group;

generating a management policy for the at least one customer group to obtain a customer management policy, and transmitting the customer management policy to a preset customer information management terminal, wherein the method specifically comprises the following steps: performing cluster analysis on the client data of the at least one client group to obtain target cluster data; performing identification matching through the target cluster data, and determining a strategy identification corresponding to the target cluster data; generating a management policy through the policy identifier to obtain a client management policy; transmitting the client management policy to a preset client information management terminal; the clustering analysis is performed on the client data of the at least one client group to obtain target clustering data, which specifically comprises the following steps: performing space point mapping on the client data of the at least one client group to obtain a corresponding space point set; performing cluster point analysis on the space point set to determine target cluster points; performing cluster analysis on the client data of the at least one client group through the target cluster points to obtain target cluster data; specifically, the server performs cluster analysis on the client data of at least one client group by inputting the client data of at least one client group into a preset cluster model; clustering the client data of at least one client group through a clustering model to obtain a plurality of characteristic data clusters; acquiring a clustering center according to the plurality of characteristic data clusters, and generating target clustering data corresponding to the client data according to the clustering center; performing identification matching on the target clustering data through the target clustering data, and determining a strategy identification corresponding to the target clustering data; generating a management policy through the policy identification to obtain a client management policy; transmitting the client management policy to a preset client information management terminal; wherein, the clustering center and the space point of the client data are extracted; respectively calculating Euclidean distances between the space points and the clustering center to obtain a target Euclidean distance corresponding to each space point; generating target cluster points according to the target Euclidean distance corresponding to each space point, carrying out cluster analysis on client data of at least one client group through the target cluster points to obtain target cluster data, carrying out identification matching through the target cluster data, and determining strategy identification corresponding to the target cluster data, wherein a mapping corresponding relation is built in advance between the target cluster data and the strategies, carrying out one-to-one correspondence through the strategy identification, carrying out management strategy generation through the strategy identification to obtain client management strategies, and finally transmitting the client management strategies to a preset client information management terminal.

2. A CRM-based customer data analysis system, the CRM-based customer data analysis system comprising:

the acquisition module is used for acquiring historical sales data, analyzing the historical sales data in a rate of return and determining a target rate of return, and specifically comprises the following steps: inquiring historical sales data from a preset CRM system, and grouping the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data; respectively carrying out the return rate calculation on each group of sales data to obtain the return rate corresponding to each group of sales data; carrying out rate of return fusion on the rates of return corresponding to each group of sales data to obtain a target rate of return; specifically, the server queries historical sales data from a preset CRM system, performs data grouping on the historical sales data according to a plurality of preset time intervals to obtain a plurality of groups of sales data, divides the historical total sales time into a plurality of time intervals, extracts sales data corresponding to each time interval to obtain a plurality of groups of sales data, utilizes a pre-established return rate analysis model set, wherein the return rate analysis model set at least comprises a conversion rate model and a return rate calculation model, the conversion rate model and the return rate calculation model are both trained deep learning models, determines weights of the conversion rate model and the return rate calculation model according to each group of sales data, calculates return rates corresponding to each group of sales data according to the weights, and performs weighted calculation on return rates corresponding to each group of sales data to obtain normalized return rate fusion data, namely target return rate;

The matching module is used for matching the intention clients through the target rate of return and determining an intention client set, wherein the intention client set comprises: the plurality of intention clients specifically include: performing data tag matching on the target return rate to obtain a plurality of data tags; performing client matching through the plurality of data tags to obtain an intent client set; specifically, the server acquires a data tag list, calculates an analysis coefficient of each candidate data tag in the data tag list, determines that the candidate data tag is a required data tag if the analysis coefficient is greater than or equal to a preset analysis coefficient threshold value, traverses the data tag list to obtain a plurality of data tags, and sends the data tags to the cloud data platform; if the analysis coefficient is smaller than a preset analysis coefficient threshold value, determining that the candidate data tag is not a required data tag, matching a plurality of data tags generated after traversal is finished with the candidate client cluster, taking the matched clients as the intention clients, obtaining a plurality of intention clients, and generating an intention client set according to the intention clients;

the generation module is used for generating the customer portrait of the intention customer set to obtain the customer portrait corresponding to each intention customer in the intention customer set; the method comprises the steps of collecting client data of each intention client in the intention client set to obtain client association data corresponding to each intention client; carrying out preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client; generating a portrait by the preference characteristics corresponding to each intention client to obtain a client portrait corresponding to each intention client in the intention client set; the method comprises the steps of obtaining personal information and user demand information corresponding to each intention client in an intention client set and historical transaction data corresponding to the intention clients; the personal information at least comprises a mobile phone number, an age and a sex; according to the personal information and the user demand information corresponding to each intention client, extracting online behavior data corresponding to each intention client in the intention client set from the CRM system; according to historical transaction data corresponding to the intention clients, determining offline behavior data corresponding to the intention clients; combining the offline behavior data and the online behavior data corresponding to the intention clients to form client association data, and constructing client portraits corresponding to the intention clients to obtain client portraits corresponding to each intention client in the intention client set; performing preference feature analysis on the client associated data corresponding to each intention client to obtain preference features corresponding to each intention client, performing interest vector calculation on the client associated data corresponding to each intention client to obtain interest vectors corresponding to each intention client, performing distance value calculation on the interest vectors corresponding to each intention client and each standard preference feature in a preset database to obtain a plurality of distance values, wherein the plurality of distance values are obtained through calculation according to a preset cosine similarity calculation formula, and taking preferences corresponding to standard preference features meeting preset conditions as the preference features corresponding to each intention client according to each distance value after the distance values are obtained through calculation; generating a customer portrait corresponding to each intention customer in the intention customer set by using the preference characteristics corresponding to each intention customer, performing standard customer portrait matching on the preference characteristics corresponding to each intention customer to obtain standard customer portrait corresponding to each intention customer, taking the standard customer portrait obtained by matching as the customer portrait corresponding to each intention customer, and respectively obtaining the customer portrait corresponding to each intention customer in the intention customer set;

The analysis module is used for carrying out association relation analysis on the customer portraits corresponding to each intention customer and determining target association relation among the intention customers; carrying out association index analysis on the customer portraits corresponding to each intended customer to obtain corresponding association characteristic indexes; building an association relation through the association characteristic indexes to obtain a target association relation between intention clients; extracting association indexes of the customer portraits corresponding to each intended customer to obtain association indexes corresponding to each intended customer, then constructing an association index system, constructing the interconnection among the association indexes corresponding to each intended customer, giving a quantification value of each association characteristic, and determining the association degree of the association indexes corresponding to each intended customer according to the quantification value of each association characteristic to obtain the index association degree corresponding to each intended customer; judging whether the index association degree exceeds a preset target value or not to obtain a judging result; if the judgment result is exceeded, carrying out association relation construction through the association characteristic index to obtain a target association relation between the intention clients; the method comprises the steps of obtaining a dynamic segmentation result corresponding to the associated characteristic indexes, and constructing a mapping relation between the associated characteristic indexes; based on each dynamic segmentation result and the mapping relation, respectively carrying out association relation construction on the association characteristic indexes by a preset storage mode and a preset integer list storage mode to obtain target association relation between intention clients;

The classification module is configured to classify the plurality of intention clients according to the target association relationship between the intention clients, so as to obtain at least one client group, and specifically includes: the server creates a client grouping model corresponding to a plurality of intention clients according to the target association relationship among the intention clients, and generates a relationship cross distribution diagram according to the target association relationship among the intention clients; generating a customer group distribution of a plurality of intended customers according to the relational cross distribution map; dividing a plurality of intention clients into client groups based on client group distribution to obtain at least one client group, and analyzing characteristic points of target association relations among the intention clients based on an association model to obtain a plurality of association relation nodes; respectively calculating the distribution weights of a plurality of association nodes to obtain the distribution weight corresponding to each association node; carrying out client group division on a plurality of intention clients according to the distribution weight corresponding to each association node to obtain at least one client group;

the generation module is configured to generate a management policy for the at least one customer group, obtain a customer management policy, and transmit the customer management policy to a preset customer information management terminal, and specifically includes: performing cluster analysis on the client data of the at least one client group to obtain target cluster data; performing identification matching through the target cluster data, and determining a strategy identification corresponding to the target cluster data; generating a management policy through the policy identifier to obtain a client management policy; transmitting the client management policy to a preset client information management terminal; the clustering analysis is performed on the client data of the at least one client group to obtain target clustering data, which specifically comprises the following steps: performing space point mapping on the client data of the at least one client group to obtain a corresponding space point set; performing cluster point analysis on the space point set to determine target cluster points; performing cluster analysis on the client data of the at least one client group through the target cluster points to obtain target cluster data; specifically, the server performs cluster analysis on the client data of at least one client group by inputting the client data of at least one client group into a preset cluster model; clustering the client data of at least one client group through a clustering model to obtain a plurality of characteristic data clusters; acquiring a clustering center according to the plurality of characteristic data clusters, and generating target clustering data corresponding to the client data according to the clustering center; performing identification matching on the target clustering data through the target clustering data, and determining a strategy identification corresponding to the target clustering data; generating a management policy through the policy identification to obtain a client management policy; transmitting the client management policy to a preset client information management terminal; wherein, the clustering center and the space point of the client data are extracted; respectively calculating Euclidean distances between the space points and the clustering center to obtain a target Euclidean distance corresponding to each space point; generating target cluster points according to the target Euclidean distance corresponding to each space point, carrying out cluster analysis on client data of at least one client group through the target cluster points to obtain target cluster data, carrying out identification matching through the target cluster data, and determining strategy identification corresponding to the target cluster data, wherein a mapping corresponding relation is built in advance between the target cluster data and the strategies, carrying out one-to-one correspondence through the strategy identification, carrying out management strategy generation through the strategy identification to obtain client management strategies, and finally transmitting the client management strategies to a preset client information management terminal.