Analytics & Trade Conversion
Problem
Event tracking on an investment app was unreliable, so downstream product decisions ran on bad data.
Impact
38.5% more trade conversions via funnel analysis and nudges, and 40% better tracking accuracy after the rebuild.
The legacy app, serving lakhs of active investors, was tracking events, but the setup was messy enough that the funnel charts didn't match what users were actually doing. I had to fix the tracking before I could trust a single chart with the new app.
Fixing the foundation
I owned event tracking end to end and rebuilt it on a clean taxonomy. Events that were firing inconsistently, some duplicated, some missing, some mislabelled, I restructured them, the whole event schema, and parameters and worked with the dev and qa team to fix how and when each one fired. That cut tracking discrepancies by around 40% and made the funnel data something you could actually decide on.
Nobody claps for clean analytics. But every decision after it runs on whether you got it right.
Then, the conversions
Once the tracking held up, I ran funnel analysis to find where users dropped off on the way to a trade, then built targeted nudges at those exact moments. Each nudge went where a specific drop-off was happening, not across the whole flow.
more trade conversions
via funnel analysis and nudgesbetter tracking accuracy
fewer tracking discrepancies after the rebuildThe conversions came later, once the data was clean enough to show where people were actually dropping.
Fix the measurement first. Everything you build after inherits whether you did or not.
The event taxonomy, the funnel data, and the specifics of where users were dropping off on the way to a trade are under NDA.
