TWI692733B - Data analysis method and financial system - Google Patents

Abstract

A data analysis method includes following operations. A referring data is collected. The referring data includes a referrer terminal and a referral terminal. A first user attribute of the referrer terminal and a second user attribute of the referral terminal are obtained by analyzing a feature of recommendation process, a feature of user relation and a financial interaction record between the referrer terminal and the referral terminal. Marketing information is sent to the referrer terminal and the referral terminal respectively according to the first user attribute and the second user attribute.

Description

Data analysis method and financial system

This disclosure document relates to a data analysis method and system for the financial industry, especially a method and system for correlation analysis of recommendation factors.

In today's era of emphasizing the importance of personal finance, various financial institutions have launched different activities, offers or products to attract customers to purchase or use in order to increase the number of customers and promote business.

However, there are many types of financial commodities, and event information is updated. It is not easy for customers to know the commodities, activities or marketing information suitable for them in a short time. Therefore, financial institutions use various user data to analyze the strength of different customers to understand their personality and consumption habits. , Recommend customers to buy or participate in their own financial products and activities.

For financial institutions, how to increase the number of users and accurately analyze different customers to achieve accurate marketing have been the core goals of various financial institutions.

In an embodiment of the present disclosure, a data analysis method includes the following operations. Collect a recommendation, the recommendation contains a recommendation The client and a recommended client. The data analysis method analyzes a recommendation process feature, a user association feature or a financial interaction record between the recommended client and the recommended client to obtain a first client attribute of the recommended client and a second client of the recommended client Attributes. According to the first client attribute and the second client attribute, a plurality of marketing information is sent to the recommended client and the recommended client, respectively.

In another embodiment of the present disclosure, a financial system includes a recommended client, a recommended client, and a financial server. The financial server is in communication with the recommended client and the recommended client, respectively. In order to collect a recommendation data between the recommended client and the recommended client. Analyze a recommendation process feature, a user association feature, or a financial interaction record between the recommended client and the recommended client to obtain a first client attribute of the recommended client and a second client attribute of the recommended client, and according to The first customer attribute and the second customer attribute send a plurality of marketing information to the recommended client and the recommended client, respectively.

In summary, according to the similarity between the attributes of the first and second customers, appropriate marketing information can be sent to achieve more accurate marketing.

100‧‧‧Financial System

120‧‧‧Recommended client

140‧‧‧Recommended client

160‧‧‧Financial Server

162‧‧‧Communication circuit

164‧‧‧ processor

200‧‧‧Data analysis method

300‧‧‧Data analysis method

M1, M2‧‧‧‧ Marketing information

REC‧‧‧Recommended information

D1a, D1b‧‧‧The first attribute analysis chart

D2a, D2b‧‧‧Second attribute analysis chart

S210, S220, S221, S222, S223, S224, S230, S231, S232, S233

Figure 1 shows a functional block diagram of a financial system.

FIG. 2 shows a flowchart of a data analysis method in an embodiment of the disclosed document.

FIG. 3 shows a detailed flowchart of the data analysis method in one embodiment of the disclosed document.

FIGS. 4A and 4B illustrate a first attribute analysis graph and a second attribute analysis graph in an embodiment of the present disclosure.

FIGS. 5A and 5B illustrate a first attribute analysis graph and a second attribute analysis graph in another embodiment of the present disclosure.

The words "including", "having", etc. used in this article are all open terms, which means "including but not limited to". In addition, "and/or" as used in this article includes any one or more of the listed items and all combinations thereof.

In this article, when an element is called "connected" or "coupled", it can be referred to as "electrically connected" or "electrically coupled." "Link" or "Coupling" can also be used to indicate the operation or interaction of two or more components. In addition, although the terms "first", "second", ... are used in this article to describe different components, the terms are only used to distinguish the components or operations described in the same technical terms. Unless the context clearly dictates, the term does not specifically refer to or imply the order or order, nor is it intended to limit the present disclosure.

Please refer to FIG. 1, which illustrates a functional block diagram of the financial system 100 in an embodiment of this disclosure. The financial system 100 includes a recommended client 120, a recommended client 140, and a financial server 160. The recommended client 120 and the recommended client 140 are connected to the financial server 160 via a network. The financial server 160 includes a communication circuit 162 and a processor 164. The communication circuit 162 is electrically connected to the processor 164.

In an embodiment, the communication circuit 162 may be a network card (wired or wireless) with various protocol interfaces, a wireless communication chip, and/or other components with communication functions, but is not limited to these. In this way, the recommended client 120 and the recommended client 140 are connected via the Internet to achieve various signal transmissions. The processor 164 may be implemented by a processor such as a central processor and/or a microprocessor, but is not limited to these.

It should be noted that the implementation of the devices and components in the financial system 100 is not limited to that disclosed in the above embodiment, and the connection relationship is not limited to the above embodiment. Anything sufficient to enable the financial system 100 to implement the following technical content Both the connection method and the implementation method can be applied to this case.

In one embodiment, the recommendation client 120 may send a recommendation information REC to the recommended client 140, and the recommended client 140 completes the recommendation after making itself a member by the recommendation information REC. After the recommendation is completed, the financial server 160 receives the data corresponding to any combination of the recommendation process feature, the user-related feature, or the financial interaction record between the recommendation client 120 and the recommended client 140 through the communication circuit 162. After analyzing the data received by the communication circuit 162, the processor 164 sends marketing information M1 and M2 to the recommendation client 120 and the recommended client 140 through the communication circuit 162 according to the analyzed results.

The recommendation information REC may be an electronic serial number corresponding to the recommendation client 120, a network link, a bar code, or other identifiers used to express a set of information, but it is not limited to these. The method for recommending the client 120 to send the recommended information REC may be to use wireless network transmission, Bluetooth, infrared, etc. The media can be sent through financial software, social networking sites, social software and other network media with sharing functions, but not limited to these.

Please refer to FIG. 2, which illustrates a flowchart of a data analysis method in an embodiment of this disclosure. In step S210 of the data analysis method 200, the communication circuit 162 in the financial server 160 collects the data of the recommended client 120 and the recommended client 140 after the recommendation is completed. In step S220, the processor 164 analyzes the recommendation process feature, user association feature, or financial interaction record between the recommended client and the recommended client to obtain the first client attribute of the recommended client 120 and the recommended client 140 Second customer attribute. In step S230, according to the first client attribute and the second client attribute in step S220, marketing information suitable for the first client attribute and the second client attribute is sent to the recommended user terminal 120 and the recommended user through the communication circuit 162, respectively端140.

Please refer to FIG. 3, which illustrates a detailed flowchart of the data analysis method in one embodiment of the present disclosure. In the data analysis method 300, as in step S210 described in FIG. 2, the communication circuit 162 in the financial server 160 collects data of the recommended client 120 and the recommended client 140 after the recommendation is completed. Step S220 includes step S221, step S222, step S223, and step S224.

In step S221, the processor 164 in the financial server 160 analyzes the recommendation process characteristics between the recommended user terminal 120 and the recommended user terminal 140. The recommended process characteristics include when the recommended user terminal 120 sends recommendation information REC to the recommended user At the end 140, the recommendation client 140 uses the recommendation information REC to make itself a member after completing the recommendation interval. At this interval The length of time will affect the correlation between the recommended client 120 and the recommended client 140. The processor 164 in the financial server 160 can set a reference time. When the interval time is less than the reference time, the processor 164 determines that the recommended client 120 and the recommended client 140 have a high correlation; when the interval time is greater than or equal to the reference time The processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a low correlation. The reference time may be set or other calculation methods based on the average time of all the recommended client 120 and the recommended client 140 completing the recommendation, but not limited to these.

In step S222, the processor 164 in the financial server 160 analyzes the user-related features between the recommended client 120 and the recommended client 140. The user related features include both the recommended client 120 and the recommended client 140 The separation distance between the corresponding residential address or communication address, the distance between the separation distance will affect the strength of the correlation between the recommended client 120 and the recommended client 140. The processor 164 in the financial server 160 can set a first reference distance. When the separation distance is less than the first reference distance, the processor 164 determines that the recommended client 120 and the recommended client 140 have a high correlation; when the separation distance is greater than or When equal to the first reference distance, the processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a low correlation. The first reference distance may be set according to the average separation distance between the recommended client 120 and the recommended client 140 that have completed the recommendation, and different first reference distances or other calculation methods may be set according to different counties and cities, but are not limited to these.

In the above step S222, the user-related feature further includes the positioning distance between the recommended user terminal 120 and the recommended user terminal 140. The location of the user terminal 120 and the recommended user terminal 140 can use GPS or He locates the system to obtain, but not limited to these. The distance between the positioning intervals will affect the strength of the correlation between the recommended user terminal 120 and the recommended user terminal 140. The processor 164 in the financial server 160 can set a second reference distance. When the positioning interval distance is less than the second reference distance, the processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a high correlation; when the positioning interval distance When it is greater than or equal to the second reference distance, the processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a low correlation. The second reference distance may be set according to the average positioning interval distance between the recommended client 120 and the recommended client 140 that have completed the recommendation, and different second reference distances or other calculation methods may be set according to different counties and cities, but not limited to these.

In step S223, the processor 164 in the financial server 160 analyzes the financial history record between the recommended client 120 and the recommended client 140, and the financial history record includes between the recommended client 120 and the recommended client 140 The frequency of transfer of account funds. The frequency of transfer of account funds includes the frequency of transfer of funds to the other party using the transfer function of the physical accounts of the recommended client 120 and the recommended client 140 respectively. The frequency of fund transfer in this account will affect the strength of the correlation between the recommended client 120 and the recommended client 140. The processor 164 in the financial server 160 may set a reference transfer frequency. When the fund transfer frequency is higher than the reference transfer frequency, the processor 164 determines that the recommended client 120 and the recommended client 140 have a high correlation; when the funds When the transfer frequency is lower than or equal to the reference transfer frequency, the processor 164 determines that the correlation between the recommended client 120 and the recommended client 140 is low. The reference transfer frequency can be set according to the user occupation, salary, salary structure, income source and other items corresponding to the respective recommendation client 120 and the recommended client 140 Fixed, but not limited to these.

In the above step S223, the frequency of transfer of account funds further includes the frequency of contact between the recommended client 120 and the recommended client 140. The contact frequency includes the frequency of contact between the recommended client 120 and the recommended client 140 through wired communication, wireless communication, shaking induction, or scanning, but is not limited to these methods. The level of this contact frequency will affect the correlation between the recommended client 120 and the recommended client 140. The processor 164 in the financial server 160 may set a reference contact frequency. When the contact frequency is higher than the reference contact frequency, the processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a high correlation; when the contact frequency is lower than or When it is equal to the reference contact frequency, the processor 164 determines that the recommended user terminal 120 and the recommended user terminal 140 have a low correlation. The reference contact frequency may be set according to the user's occupation, salary, salary structure, income source and other items corresponding to the recommendation client 120 and the recommended client 140, but is not limited to these.

Through the analysis results in step S221, step S222, and step S223, the first client attribute of the recommended client 120 and the second client attribute of the recommended client 140 are obtained. The processor 164 can set different weights for the recommendation process characteristics, user-related characteristics, and financial history records in step S221, step S222, and step S223, or further, for the interval time, interval distance, positioning interval distance, and account Different weights are set for the frequency of fund transfer and the frequency of contact, resulting in different strengths of association.

In step S231, the processor 164 compares whether the first client attribute of the recommended client 120 is similar to the second client attribute of the recommended client 140. The method of comparison can be, for example, to determine whether it is greater than a setting Reference value or other comparison methods, but not limited to this. If the comparison results are similar, in step S232, quantitative results will be generated based on the similar first and second customer attributes, respectively, as shown in Figures 4A and 4B, the first attribute analysis graph D1a and the second attribute analysis Figure D2a.

Please refer to FIGS. 4A and 4B. FIGS. 4A and 4B illustrate schematic diagrams of the first attribute analysis graph D1a and the second attribute analysis graph D2a in an embodiment of the present disclosure. The first attribute analysis graph D1a corresponds to the recommendation client 120, and the second attribute analysis graph D2a corresponds to the recommended client 140. The first attribute analysis graph D1a and the second attribute analysis graph D2a have five indicators, which are the amount of assets, automation acceptance, risk tolerance, transaction frequency and connection strength.

In Figures 4A and 4B, the four indicators of asset size, automation acceptance, risk tolerance and transaction frequency are the recommended client 120 and the recommended client 140 in their respective analysis results, and the correlation strength is the recommended client 120 The analysis results of the recommended client 140 through step S221, step S222, and/or step S223.

In Figures 4A and 4B, the amount of assets includes deposits, bank loans, debts, or other assets known to financial institutions in the physical accounts or online accounts of the respective users of the recommended client 120 and the recommended client 140. Information, but not limited to these. The amount of assets may be set by the processor 164 according to the average assets of users of the same age corresponding to the recommended client 120 and the recommended client 140 or other calculation methods, but not limited to these.

In Figures 4A and 4B, automation acceptance includes The acceptance of the respective users of the recommended client 120 and the recommended client 140 to the automation services of financial institutions, including the transition from traditional counter-office financial services to the use of network devices or mobile devices to perform digital banking or perform financial business functions, automation The acceptance criteria can be calculated by the processor 164 for the frequency with which the user uses the financial business functions of the digital bank, the number of times the digital bank is used over a period of time, or other calculation methods, but not limited to these.

In FIGS. 4A and 4B, the risk tolerance includes the risk acceptance of investment financial products by the users of the corresponding recommended client 120 and the recommended client 140. The risk tolerance can be calculated by the processor 164 Investment records such as the type, frequency, amount and holding time of financial commodities or the user's existing assets, income, loans and debts and other asset records, but not limited to these.

In FIGS. 4A and 4B, the transaction frequency includes the number of times that the respective users of the recommended client 120 and the recommended client 140 transfer account deposits to other users. The transaction frequency can be calculated by the processor 164 over a period of time Number of transfers, average number of transfers, or other calculation methods, but not limited to these.

In FIGS. 4A and 4B, the correlation strength includes the correlation strength between the corresponding recommended client 120 and the recommended client 140 calculated through the analysis results of step S221, step S222, and/or step S223, detailed The calculation method and steps are as described above and will not be repeated here.

As shown in FIG. 4A and FIG. 4B, since the correlation strength is calculated based on each correlation item between the recommended client 120 and the recommended client 140, from the calculation standard of the correlation strength in the foregoing embodiment, The higher the connection strength (closer to the outermost frame), it means that the recommendation client 120 and the recommended client 140 are more common, contact or interact, have a closer residence, or have a higher frequency of funds transfer. From the above, it may also mean that the relationship between the users corresponding to the recommended client 120 and the recommended client 140 is more intimate, the personality is more similar, the growing environment is closer, or the values are closer, or they may be neighbors, friends, lovers, couples. , Family members, relatives or colleagues rather than strangers. Therefore, the higher the correlation strength, the higher the probability that the first customer attribute is similar to the second customer attribute, and the more similar the first attribute analysis graph D1a and the second attribute analysis graph D2a are.

In step S232, based on the analysis results of the similar first client attributes and second client attributes, the processor 164 simultaneously transmits the same or similar marketing information M1 and M2 to the recommendation client 120 and the recommended client through the communication circuit 162端140. In an embodiment, in step S232, the first attribute analysis graph D1a and the second attribute analysis graph D2a shown in FIGS. 4A and 4B may be generated according to the similar first client attributes and the second client attributes, and then processed According to the analysis results of the first client attribute and the second client attribute and the first attribute analysis graph D1a and the second attribute analysis graph D2a, that is, according to the user characteristics of the corresponding recommendation client 120 and the recommended client 140, The same or similar marketing information M1 and M2 are simultaneously transmitted to the recommendation client 120 and the recommended client 140 through the communication circuit 162. The recommendation client 120 can view the marketing information M1 that meets their needs, and the recommended client 140 can view Marketing information M2 that meets your needs.

For example, when the processor 164 is expected to send information about life insurance products to the recommendation client 120, the financial system 100 then According to the analysis, it is judged that the recommended client 120 and the recommended client 140 are highly similar. Therefore, the same life insurance product information or the same type of insurance product information can also be sent to the recommended client 140. As another example, if it is expected to send the preferential loan interest rate advertisement to the recommended client 140, the financial system 100 may also send the same preferential loan interest rate advertisement or the same type of loan advertisement to the recommending client 120.

In step S231, if the comparison results are different (for example, less than or equal to the set reference value), in step S233, quantized results will be generated according to the different first and second customer attributes, as shown in Figure 5A and The first attribute analysis graph D1b and the second attribute analysis graph D2b shown in FIG. 5B.

Please refer to FIGS. 5A and 5B. FIGS. 5A and 5B illustrate the first attribute analysis graph D2a and the second attribute analysis graph D2b in another embodiment of the present disclosure. The five indicators are the same as shown in Figures 4A and 4B.

As shown in FIG. 5A and FIG. 5B, according to the calculation criterion of the correlation strength in the foregoing embodiment, when the correlation strength is lower (the closer to the innermost frame), it indicates that the recommended client 120 and the recommended client 140 are relatively Rarely meet, contact or interact, have a long distance from the place of residence, or have a low frequency of fund transfer. From the above, it may also mean that the more alienated the relationship between the users corresponding to the recommended client 120 and the recommended client 140, the more different the personality, the more different the growth environment or the more different values, each other is a neighbor, friend, lover, couple, The lower the probability of family members, relatives or colleagues, and the higher the probability of being strangers to each other. Therefore, the lower the correlation strength, the lower the probability that the first customer attribute is similar to the second customer attribute, and the more different the first attribute analysis graph D2a and the second attribute analysis graph D2b are.

In step S233, after generating the first attribute analysis graph D2a and the second attribute analysis graph D2b according to the different first client attributes and second client attributes, the processor 164 will determine the first client attribute and the second client attribute and the second The analysis results of an attribute analysis graph D2a and a second attribute analysis graph D2b, that is, according to the user characteristics of the corresponding recommendation client 120 and the recommended client 140, different or completely different marketing information M1 and M2 are communicated by communication The circuit 162 transmits to the recommendation client 120 and the recommended client 140 respectively. The recommendation client 120 can view the marketing information M1 that meets their needs, and the recommended client 140 can view the marketing information M2 that meets their needs.

For example, when the processor 164 is expected to send information about life insurance products to the recommendation client 120, the financial system 100 judges that the similarity between the recommendation client 120 and the recommended client 140 is low through the above analysis. Of life insurance products or insurance products of the same type are sent to the recommended client 140. At this time, the financial system 100 may additionally recommend different financial commodities, such as linked bonds, according to the characteristics of the recommended client 140.

For example, as shown in Figures 5A and 5B, the first attribute analysis graph D2a and the second attribute analysis graph D2b differ greatly in the amount of assets and risk tolerance. When the processor 164 is expected to send high-risk fund commodity information (For example, a stock fund) to the recommended client 120. At this time, the financial system 100 determines that the recommended client 140 has a low risk-taking ability through the above analysis, and does not need or avoid sending high-risk fund-related commodity information to the recommended client 140.

The above marketing information M1 and M2 may be in accordance with the current marketing activities information of the financial institution, member bonus point activity information, financial commodity information, advertising information, conference information, exhibition information or other information, but not limited to these.

Through the operation of the above embodiment, more accurate customer analysis results can be achieved according to the correlation characteristics between different users, so that users can obtain the information they really need and are interested in real time, saving many users from searching Checking the time also allows users to avoid errors caused by non-financial professional judgment.

Although this disclosure document has been disclosed as above with examples, it is not intended to limit this disclosure document. Anyone who is familiar with this skill can make various changes and retouching without departing from the spirit and scope of this disclosure document. The scope of protection of the disclosure document shall be deemed as defined by the scope of the attached patent application.

300‧‧‧Data analysis method

S210, S220, S221, S222, S223, S224, S230, S231, S232, S233

Claims (8)

A data analysis method includes: collecting recommended data including a recommended user terminal and a recommended user terminal; analyzing a recommendation process feature and a user association between the recommended user terminal and the recommended user terminal Characteristics or a financial interaction record, obtain a first client attribute of the recommended client and a second client attribute of the recommended client; and send a plurality of marketing information based on the first client attribute and the second client attribute, respectively To the recommended client and the recommended client, wherein the recommendation process feature includes an interval between sending a recommendation invitation from the recommended client to the recommended client accepting the recommendation invitation; if the interval is less than a reference time , The data analysis method determines that the first client attribute is similar to the second client attribute, and sends the same marketing information to the recommended client and the recommended client at the same time. If the interval is greater than or equal to the reference time, the The data analysis method determines that the attributes of the first client and the attributes of the second client are different, and sends differentiated marketing information to the recommended client and the recommended client, respectively. The data analysis method according to claim 1, wherein the user-related feature includes a separation distance between the residential address or the correspondence address corresponding to the recommended client and the recommended client, wherein when the separation distance is less than A first reference distance, the data analysis method determines that the first customer attribute is similar to the second customer attribute, and the same marketing information Is sent to the recommended client and the recommended client; when the separation distance is greater than or equal to the first reference distance, the data analysis method determines that the attributes of the first client are different from the attributes of the second client. Marketing information is sent to the recommended client and the recommended client respectively. The data analysis method according to claim 1, wherein the user-related feature includes a positioning interval distance between the recommended user terminal and the recommended user terminal, wherein when the positioning interval distance is less than a second reference distance, the data The analysis method determines that the first client attribute is similar to the second client attribute, and sends the same marketing information to the recommended client and the recommended client at the same time; when the positioning interval distance is greater than or equal to the second reference distance, the The data analysis method determines that the attributes of the first client and the attributes of the second client are different, and sends differentiated marketing information to the recommended client and the recommended client, respectively. The data analysis method according to claim 1, wherein the financial history record includes an account fund transfer frequency between the recommended client and the recommended client, wherein the account fund transfer frequency is higher than a reference transfer Frequency, the data analysis method determines that the first client attribute is similar to the second client attribute, and sends the same marketing information to the recommended client and the recommended client at the same time; when the account fund transfer frequency is less than or equal to The reference transfer frequency, the data analysis method determines that the first client attribute is related to the second client attribute Different, different marketing information is sent to the recommended client and the recommended client. The data analysis method according to claim 1, wherein the financial history record includes a contact frequency between the recommended client and the recommended client, wherein the contact frequency includes the recommended client and the recommended client by Wired communication, wireless communication, shaking induction, or scanning contact frequency, wherein when the contact frequency is higher than a reference contact frequency, the data analysis method determines that the first customer attribute is similar to the second customer attribute, and the same marketing information At the same time, it is sent to the recommended client and the recommended client; when the contact frequency is lower than or equal to the reference contact frequency, the data analysis method judges that the first client attribute is different from the second client attribute, Marketing information is sent to the recommended client and the recommended client respectively. A financial system includes: a recommended client; a recommended client; and a financial server, which are respectively in communication with the recommended client and the recommended client. The financial server is used to: collect the recommended client A recommendation profile with the recommended client; analyze a recommendation process feature, a user association feature, or a financial interaction history between the recommended client and the recommended client Record, obtain a first client attribute of the recommended client and a second client attribute of the recommended client; and send a plurality of marketing information to the recommended client according to the first client attribute and the second client attribute, respectively With the recommended client, wherein the recommendation process feature includes an interval from when the recommended client sends a recommendation invitation to when the recommended client accepts the recommended invitation; if the interval is less than a reference time, the financial server It is determined that the first client attribute is similar to the second client attribute, and the same marketing information is sent to the recommended client and the recommended client at the same time. If the interval time is greater than or equal to the reference time, the financial server determines the The first client attribute is different from the second client attribute, and different marketing information is sent to the recommended client and the recommended client, respectively. The financial system according to claim 6, wherein the user-related feature includes a separation distance between the residential address or correspondence address corresponding to the recommended client and the recommended client, or the user-related feature includes the A positioning interval distance between the recommended client and the recommended client, wherein when the interval distance or the positioning interval distance is less than a reference distance, the financial server determines the first client attribute and the second client attribute Similarly, the same marketing information is sent to the recommended client and the recommended client at the same time; when the separation distance or the positioning separation distance is greater than or equal to the reference distance, the financial server judges the first client attribute and the first The two customers have different attributes, and send the different marketing information to the recommended client and the recommended client respectively. The financial system according to claim 6, wherein the financial history record includes an account fund transfer frequency between the recommended client and the recommended client, or the financial history record includes the recommended client and the recommended A contact frequency between users, wherein when the account funds transfer frequency is higher than a reference transfer frequency or the contact frequency is higher than a reference contact frequency, the financial server determines the first client attribute and the second client attribute Similarly, send the same marketing information to the recommended client and the recommended client at the same time; when the account transfer frequency is lower than or equal to the reference transfer frequency or the contact frequency is lower than or equal to a reference contact frequency, The financial server determines that the first client attribute is different from the second client attribute, and sends the differentiated marketing information to the recommended client and the recommended client, respectively.

Images