Customer Health Indicator

Document Type

Abstract

Publication Date

Fall 12-1-2017

Abstract

Cloud-based SaaS platforms becoming more and more popular with today’s technology companies, many who are pursing either SaaS B2B or B2C have found themselves struggling to retain customers from one payment period to the next. This has become increasingly problematic for companies as more and more competition enter the markets and potentially take away valuable clientele. In order to help combat this issue, many companies have turned to customer success in order to retain customers and keep them from canceling subscriptions. Customer Health Indicator, a web-based application, works to reduce churn (customers notrenewing a subscription), by predicting if a customer is likely to churn in thenext payment period in order to help users be more proactive with theirclientele. However, Customer Health Indicator is not solely useful forpredicting if a customer will churn but also allows users insight into whichcustomers should be upsold in the next payment period. Customer HealthIndicator will utilize python to pre-process data and run regressions such asgradient boosting on data provided by Telcom in order to predict if a customerwill churn in the next payment period or not. Through simplistic user interface, Customer Health Indicator will allowusers to easily and accurately determine whether or not any given customer willchurn based on a few of the most highly correlated customer data inputs inregard to the churn attribute of the dataset. All the user must do in order topredict if a customer will churn in the next payment period is enter theircustomer data, click predict, and see the prediction generated all from the home page.

Comments

Completed as part of the Computer Science Senior Capstone Project.

This document is currently not available here.

Share

COinS