Product recommendation method based on sales clue interaction records

Abstract

The invention provides a product recommendation method based on sales clue interaction records, and the method comprises the following steps: S1, collecting sales clue information of a product, the sales clue being a customer who is not verified and filtered; s2, establishing a sales clue prediction model, and evaluating a purchase intention index of the sales clue; s3, collecting communication information between the sales clue and the salesman; s4, constructing verbal skill database, wherein the salesperson obtains recommended verbal skills in a verbal skill database according to the targetsales clue, and recommends the target product to the sales clue by adopting the obtained recommendation verbal skill. Through the method, the purchase intention of the sales clue can be accurately judged according to the information of the sales clue and the communication information between the sales clue and the salesman, and the conversion rate of converting the sales clue to the actual purchaser is effectively improved, so that the sales performance of an enterprise is improved, and the sales cost of the enterprise is reduced.

Description

technical field [0001] The invention relates to a product recommendation method, in particular to a product recommendation method based on sales lead interaction records. Background technique [0002] With the continuous development of information technology and the Internet, intelligent recommendation systems have gradually penetrated into every industry in our lives. Through the collection and analysis of users' basic attributes and behavioral data, it can help enterprises make more effective decisions and predictions in various fields. [0003] Many existing intelligent recommendation systems use users as the main line to establish data collection index dimensions: purchase history, human attributes, purchase events, etc. However, these data can only serve as an initial guide for users, and are not very helpful for low-frequency and high-value products. For example, in the automobile industry, users generally involve three steps in purchasing a car instead of buying it ...

AI Technical Summary

Problems solved by technology

Ordinary intelligent recommendation systems can only make one-time recommendations based on the user's basic information, which not only results in the call records in the above three links, sales leads to store behavior and other sales lead interaction records are all ignored, but also many lead opportunities are not obtained Well-used, resulting in poor sales performance, high sales costs, and difficulty for users to buy their desired products

Images