顧客間のヒューマンネットワークを活用したキーパーソン抽出方式

抄録

In the case of outcall sales at financial institutions, how to acquire potential customers that lead to new contracts has a direct bearing on the performance of sales representatives. Unfortunately, to acquire prospective customers in an efficient manner is, in many cases, an intuitive skill that is learned by sales representatives through experience, and is not part of the institutional knowledge of the company.

In this study, we propose a customer key person extraction method using customer data based on the premise of Business style "Referral Sales", which will introduce new customers from existing customers by utilizing organizational intelligence to enable customers to be able to anticipate new contracts. Specifically, in order to extract the customer key person, the chain property and diversity of the relationship between the customers, by the various edge of the Human Relations Network, to express the following contents.

(1) A character in an edge is represented as peculiar coefficients.

(2) Each node value is summation of characters in edges that are connected to the node.

(3) A key person value in one node is calculated from its node value and node values of the descendant nodes multiplied by a distance coefficient.

Customer key person is selected in order high key person value of a node. We call this above method HuRAT; Human Relational Analysis Technique.

In addition, as a result of applying this method to the actual business operations of the three insurance companies, and HuRAT improved the customer key person extraction rate compared with PageRank and the other conventional centrality indicator methods.