Essays on Algorithms for Customer Acquisition and Retention in SaaS Business Model

Abstract

In recent years, software firms have migrated from the traditional licensing business model to the ``Software as a Service'' (SaaS) model, where consumers subscribe to the software on monthly, quarterly, or annual contracts. This change has created new opportunities and challenges in the domain of customer acquisition and retention for firms. SaaS firms have access to an unprecedented amount of data since offering software as a service enables them to capture users' behavioral and contextual data at a granular level. Nevertheless, utilizing this amount of data requires a set of high-dimensional friendly tools and methodologies that may not be available. In this dissertation, I try to address the challenge of high-dimensionality in modeling customer acquisition and retention in the SaaS business model. The first chapter, following the introduction, outlines high-dimensional data algorithms available to design and evaluate optimal free trials, the most commonly used customer acquisition strategy in SaaS. Using data from a field experiment, I showcase how companies can design optimal trial policies and understand the underlying mechanism for the effect of trial length on customer acquisition. In the second chapter, I discuss the problem of modeling customer retention as a dynamic discrete choice model. I offer a new algorithm that let researchers incorporate the high-dimensional usage data when modeling users' subscription decision. I run several simulations using the canonical bus engine replacement problem to show the performance of the proposed algorithm. Then, I discuss the limitations of the new algorithm and explain how researchers can use it to model customer retention in SaaS. The algorithm's source code is available on Github to be used and further developed for other applications.