Feature usage correlation with customer retention in the SaaS platform

Abstract

This study examines the customer base operational data of the B2B SaaS company to find out the insights on improving the onboarding process, sticky features, behavioral differences between retained and churned customers, as well as the features correlating the most with customer retention. The analysis involves examining the correlation to define the association between feature usage and customer retention. It investigates the touching points between these two variables and derives the driving factors of engagement. Studying the patterns of feature usage is crucial for the case company to understand the right approach towards involving and acquiring customers.

Quantitative analysis combines methods such as correlation analysis, group-difference testing, co-adoption metrics, retention lift, and breadth. By using this, the merged feature usage and customer lifecycle data were studied accordingly. To identify the feature usage patterns, Jupyter notebooks along with Python were used throughout the data preparation and analysis phases. The results from analytical methods were explored; however, the analysis itself didn’t provide any reliable outcomes, indicating limitations in the ratio of the studied dataset. Nevertheless, some features indeed highlight descriptive activity, which can be studied more deeply with a more balanced sample, a wider observation window, and detailed customer feedback.