Method for bank to carry out personalized marketing by utilizing industry public data

Abstract

The invention discloses a method for a bank to carry out personalized marketing by utilizing industrial public data. The method comprises the following steps: integrating global total data and storing the global total data in an industrial map; selecting an initial node and finding out all candidate enterprises in the industrial map according to the initial node; selecting a filtering model to filter the candidate enterprises; obtaining an optimal reaching path for each candidate enterprise through a reaching model; and obtaining product information and marketing verbal skills by utilizing a recommendation model according to the candidate enterprise features. According to the method, the associated enterprises can be obtained as marketing candidate objects, then the funnel model is used, screening and filtering are continuously carried out, a small number of candidate enterprises meeting requirements are obtained, precise marketing is facilitated, and the conversion rate is greatly increased.

Description

Technical Field

[0001] The present invention belongs to the technical field of financial data processing, and specifically is a method for a bank to use industry public data for personalized marketing. Background Art

[0002] In the past, banks conducted corporate business mainly from the perspective of enterprises, which was individualized, decentralized, and difficult to acquire customers. In fact, changes in economic structure are usually caused by changes in industrial structure. When the industrial structure changes, the economic structure will also adjust accordingly.

[0003] In the financial industry, how to effectively use industry public data for personalized marketing is still an issue that has not been fully resolved. The main difficulties are: 1. the availability, validity and integrity of data; 2. how to mine data associations for marketing.

[0004] As data supervision becomes increasingly strict, many detailed data of enterprises cannot be obtained in the open market, such as enterprise turnover, profit, invoice data, etc. Through websites such as Qichacha and Tianyancha, you can only query the data of a single enterprise by enterprise name or social unified credit code, without relevant industry information and global full data.

[0005] After obtaining the full amount of global data, you also need to be able to analyze, discover, and summarize the correlations for marketing, and you must ensure the marketing conversion rate.

[0006] Traditional banking generally acquires customers through marketing, campus marketing, online planning, offline scenario marketing, etc. After digital transformation, some banks can acquire customers through new enterprise registration information, equity relationships of existing customers, transaction relationships, etc., but they still mainly rely on expert experience to define customer acquisition channels and obtain marketing candidate customers through precise queries, which are usually single-dimensional.

[0007] Traditional methods have no data support or rely on only a small amount of data, with a single dimension for acquiring customers, insufficient accuracy, and low conversion rates. It is difficult to reach new customers and lacks marketing strategies. Summary of the invention

[0008] In view of the above problems, the present invention is proposed to provide a method for banks to use industry public data for personalized marketing, which overcomes the above problems or at least partially solves the above problems.

[0009] To achieve the above object, the present invention adopts the following technical solutions:

[0010] A method for a bank to use industry public data for personalized marketing, the method comprising:

[0011] Integrate all global data and save them in the industry map;

[0012] Select the initial node and find all candidate enterprises in the industry graph based on the initial node;

[0013] Select a filtering model to filter candidate companies;

[0014] Obtain the optimal reach path for each candidate enterprise through the reach model;

[0015] Use recommendation models to obtain product information and marketing strategies based on the characteristics of candidate companies.

[0016] Optionally, the global full data includes full operating enterprise data, industrial chain data, news data and private equity data.

[0017] Optionally, the initial nodes include news, policies, events, private placements, regions, industries and existing customers.

[0018] Optionally, all candidate enterprises in the industry graph are found based on the initial nodes through a graph algorithm.

[0019] Optionally, the filtering model is selected based on successful expert experience, a scoring model or historical statistics.

[0020] Optionally, the filtering model is a funnel model.

[0021] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0022] 1. The present invention widely acquires data of various dimensions and organizes them into an industrial map data network with rich associations, making it easy to find marketing clues. When news, policies, events, etc. occur, related companies are obtained as marketing candidates, and then the funnel model is used to continuously screen and filter to obtain a small number of candidate companies that meet the requirements, which is convenient for precision marketing.

[0023] 2. The selection of the funnel model of the present invention relies on the experience of successful experts, scoring models or historical statistics; compared with traditional methods, rich association data can provide marketing clues, and the funnel model can screen out the best candidate companies for precision marketing. For candidate marketing companies, it provides contact methods, optimal contact paths, and matching products and words based on corporate feature labels, providing account managers with the best contact paths and marketing words, greatly improving conversion rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a method for banks to use industry public data for personalized marketing provided in an embodiment of the present application;

[0025] Figure 2A flowchart of an application method for a bank to use industry public data for personalized marketing is provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] See also Figure 1 and Figure 2 This embodiment provides a method for a bank to use industry public data for personalized marketing, the method comprising:

[0028] S1. Integrate all global data and save it in the industry map.

[0029] The global full data includes full operating enterprise data, industrial chain data, news data and private equity data, etc.

[0030] S2. Select the initial node and find all candidate enterprises in the industry map based on the initial node.

[0031] The initial nodes include news, policies, events, private placements, regions, industries and existing customers.

[0032] Use graph algorithms to find all candidate companies in the industry graph based on the initial nodes. For example, you can use the k-hop algorithm or the weighted shortest path algorithm. To determine whether there are candidates that meet the marketing conditions, such as small and micro enterprises, the data set obtained will generally be very large.

[0033] S3. Select a filtering model to filter candidate companies.

[0034] The selection of the filtering model is based on successful expert experience, scoring model or historical statistics. The filtering model is a funnel model. Using the funnel model, the filtering model is continuously selected for filtering until the number of candidates meets the requirements. For example, first find those within 5km of the business outlets, then find those related to existing customers, and then find those whose scores are higher than the average of the industry link.

[0035] S4. Obtain the optimal reach path for each candidate enterprise through the reach model.

[0036] S5. Use the recommendation model to obtain product information and marketing tactics based on the characteristics of candidate companies. Sales personnel conduct marketing.

[0037] This embodiment widely obtains data of various dimensions and organizes them into an industrial map data network with rich associations, making it easy to find marketing clues. When news, policies, events, etc. occur, related companies are obtained as marketing candidates, and then the funnel model is used to continuously screen and filter to obtain a small number of candidate companies that meet the requirements, which is convenient for precision marketing.

[0038] The selection of the funnel model in this embodiment depends on the experience of successful experts, scoring models or historical statistics. Compared with traditional methods, rich association data can provide marketing clues, and the funnel model can screen out the best candidate companies for precision marketing. For candidate marketing companies, it provides contact methods, optimal contact paths, and matching products and words based on corporate feature labels, providing account managers with the best contact paths and marketing words, greatly improving conversion rates.

[0039] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for a bank to use industry public data for personalized marketing, characterized in that: The method comprises: Integrate all global data and save them in the industry map; Select the initial node and find all candidate enterprises in the industry graph based on the initial node; Select a filtering model to filter candidate companies; Obtain the optimal reach path for each candidate enterprise through the reach model; Use recommendation models to obtain product information and marketing strategies based on the characteristics of candidate companies.

2. A method for banks to use industry public data for personalized marketing as claimed in claim 1, characterized in that: The global full data includes full operating enterprise data, industrial chain data, news data and private equity data.

3. A method for banks to conduct personalized marketing using industry public data as claimed in claim 1, characterized in that: The initial nodes include news, policies, events, private placements, regions, industries and existing customers.

4. A method for banks to conduct personalized marketing using industry public data as claimed in claim 1, characterized in that: Through graph algorithms, all candidate companies in the industry graph are found based on the initial nodes.

5. A method for banks to conduct personalized marketing using industry public data as claimed in claim 1, characterized in that: The filtering model is selected based on successful expert experience, scoring model or historical statistics.

6. A method for banks to conduct personalized marketing using industry public data as claimed in claim 1, characterized in that: The filtering model is a funnel model.

Images