Understanding the behavior of customers is crucial for large companies. With the knowledge of the behavior of their customers, companies can provide products more assertively and at the most opportune moment. For this, Datalive proposes the integration between online and offline behavior, in order to gain a richer knowledge about customers.
As primary objectives we can cite the understanding of the reasons that motivate a user to identify themselves in the websites of the companies besides to understand in which moment this occurs. In possession of this knowledge it is possible to request identification of other clients who are under imminent identification.
Given the identification of potential customers the next step is to understand the purchasing process and, similarly, to win potential customers whose characteristics are similar to those observed previously.
As a solution to the initial problems were generated webservices that, given the characteristics of a particular user, return a decision to be made in relation to their navigation (send a message or not). For this, several packages were used in the R:
- plumber: Webservices development
- stringr: String manipulation
- dplyr: Data handling
- data.table: Data handling
- Develop and support web services
- To search for techniques applicable to these problems
- Coding of techniques in R
October 2017 through February 2018.