← All work
Analytics Growth

Credit Line on UPI: building a new revenue engine

When the RBI enabled credit lines on UPI, I led the full implementation at BharatPe, from digging into payment data to designing the activation funnel to coordinating with lending partners. The result was $12K in monthly revenue and a steady 12% week-over-week growth rate.

Company BharatPe
Role Product Intern, Payments
Timeline July — Dec 2024
Impact $12K MoM, 12% WoW growth

The context

UPI processes billions of transactions every month in India. It's the backbone of digital payments for BharatPe's merchant network. But the revenue model was mostly transaction fees, and not much room to grow without growing volume.

Then the RBI enabled "Credit Line on UPI," letting users access pre-approved credit lines directly through UPI payments. For a payments platform like BharatPe, this was a big deal: suddenly every UPI transaction could potentially generate interest income and processing fees on top.

The opportunity: Credit Line on UPI could turn every transaction into a lending moment, adding a revenue layer to payment volume that already existed.

The problem

The regulatory green light was there, but actually building this was messy:

🔒

Get the full breakdown

The detailed process, metrics, and learnings from this project. Drop your info below and it's yours instantly.

No login needed. No spam. Just the content.

What I did

1. Dug into the payment data

I started with SQL queries on our data warehouse and Mixpanel event data, trying to understand where credit could add the most value. I was looking for:

  • Transaction patterns that indicated credit demand: high-frequency, mid-ticket-size payments
  • Which merchant categories were most likely to see credit usage
  • Time-of-day and day-of-week patterns for when to surface credit options

This wasn't glamorous work, but it was the foundation for everything else. The data told us exactly where to focus.

Data analytics dashboard on laptop
Behavioral analysis: mapping transaction patterns to identify credit demand signals

2. Designed the activation funnel

Based on the data, I designed the end-to-end funnel for credit line activation:

  • Discovery: Surfacing credit availability at the right moment in the payment flow (not before, not after)
  • Activation: Minimizing the steps from "oh this exists" to first credit transaction
  • Retention: Notification triggers and usage nudges to build the habit

I applied the same behavioral analysis to Credit Card on UPI placement and increased that revenue 21% just through better funnel positioning. Same product, better placement, more money.

3. Coordinated across teams

This touched engineering (API integration), compliance (regulatory sign-off), business development (lending partner agreements), and the merchant-facing team (education and onboarding). I also led a team of 8 on the parallel Multi-cycle Settlement initiative, which cut payout costs by $20K monthly.

Results

$12K

Monthly revenue

12%

Week-over-week growth

21%

Revenue lift (CC on UPI)

Credit Line on UPI hit steady compound growth from day one. And the 21% revenue lift on Credit Card on UPI came purely from better placement: same product, better understanding of when users were ready to see it.

What I learned

  • Look at the actual data before building hypotheses. The payment patterns revealed credit demand signals that weren't obvious from aggregate metrics. Spending time in the data warehouse paid off more than any brainstorming session.
  • Placement is a product lever. The 21% revenue lift from Credit Card on UPI was purely a funnel positioning change. No new features, no redesign. Just showing the right thing at the right time.
  • 12% weekly growth compounds fast. It sounds incremental, but 12% week-over-week means you roughly double every 6 weeks. That kind of compounding is what makes a feature go from "promising" to "important."