Leading Tax Filing Company
Personalization-Nudging to prevent drop-off
Problem
The client aimed to enhance the user experience on their tax filing platform, which faced multiple issues causing a significant number of users to abandon the process prematurely.
Solution
An ML-based drop-off propensity model was developed to predict the likelihood of a customer exiting the platform. Based on the model's scores, personalized interventions were provided to guide users in the right direction and keep them engaged.
Outcome
Timely nudges and tailored guidance increased customer engagement by 20% and reduced premature drop-offs by over 18%.
Targeted Interventions to Address User Drop-Off Challenges.
The client faced challenges with users dropping off prematurely during the tax filing process due to issues like information overload, distractions, user fatigue, and connectivity problems. To address this, a drop-off propensity ML model was developed, leveraging past user behavior data to predict the likelihood of drop-offs at each stage. Based on these scores, users received targeted “nudges” such as tips, sticky notes, and potential tax return projections to guide them through the process. A Random Forest classifier was selected for its superior performance, and the model was integrated with the platform via a REST API.
The solution significantly improved the user experience by offering timely guidance and support, increasing customer engagement by over 20%. Additionally, the proactive interventions reduced drop-off rates by more than 18% within three months of deployment, helping users successfully complete their tax filing.
Nudges That Deliver
Explore All
Discover our complete range of high-quality products designed to meet your needs. Explore now and find the perfect fit for your requirements.