New Analytics Tool, No Problem… Kind of 🫣🫠

I’ve recently been assigned a series of problem areas for a fintech (personal/business lending, savings, BNPL, pay), and part of the product team’s data war chest is CleverTap. Being accustomed to Mixpanel, I had to quickly overcome my internal bias and learn a new tool doing the same thing (maybe better, maybe worse – flexibility is key).

To overcome this bias, my approach with any new tool is to dive into the features, adapt my old framework, and quickly test it with a real problem.

In this post, I’ll share my new framework for problem discovery using CleverTap. Maybe I’ll include details about my first project too? Probably in the future 👀

Step 1: Become The Problem Expert

  1. What signal/indicator are we detecting?
  2. What user pain(s) does it point to? (How is the signal measured? What sub-signals affect it?)
  3. When in the user journey does it happen?
  4. When the user encounters the pain point(s), what do they do?
  5. What is the underlying business/app logic for the affected journeys and why was it implemented that way?
  6. Finally, Can we generate some customer/stakeholder/ Clevertap™️- data questions, and hypotheses to validate?

Step 2: Where do I look in the data for quick insights, and hypothesis generation/validation?

Problem Type Analytic Tools
User Engagement Drop-off Funnels: to identify where users are dropping off in the conversion flow
Flows: to track user actions before and after the drop-off point to find patterns.
Campaigns: find in-use push notifications or in-app messages to re-engage users at critical drop-off points.
Journeys: investigate efficiency metrics of current automated sequences designed to guide users back on track and analyse the effectiveness. Segments: Are there users with specific attributes whose engagement drop-off could be skewing the metrics or are the most affected
Understanding User Behaviour Events View + Trends + Segments: Monitor user interactions and behaviours, and group them based on user attributes to detect different behaviours across segments
Cohorts: Compare how different cohorts perform over time to understand behavioural shifts.
Improving Conversion Rates w/ Nudges + Improving UX Funnels + Journeys (Automate a sequence of messages to guide users towards a goal) + A/B Testing of content
Retention + Feature Adoption Issues Cohorts (identify usage patterns)+ Segmentation + Events View (Track usage or near usage of feature)+ Journeys + Pivots (master view based on events/segments/cohorts)