Background
I built and launched a co-branded low-cost EMI card on a large consumer fintech app in partnership with an NBFC partner. The card targets a large eligible user base within the app ecosystem. The card works differently from a standard credit card: it activates only when the cardholder makes their first offline purchase at an affiliated partner store, at which point a consumer-durable loan is initiated with the NBFC partner.
This mechanic creates a specific product challenge. Issuing a card is step one. Getting that card used — activated — at a physical store is an entirely separate conversion problem, and one that no amount of digital onboarding can fully solve on its own.
The Problem
Post-launch, card issuance scaled well. Activation did not keep pace.
The fundamental issue: after a user received their card, the product journey essentially went quiet. There was no structured, personalized experience guiding users toward their first store visit and purchase. Users who didn't immediately know where to use the card, what offers were available, or which stores were nearby simply didn't activate — and stayed dormant.
Month-1 activation rates were low, and Month-11 activation — a proxy for long-term card utility — was even more concerning.
The Approach
I framed the post-carding journey as a separate product surface that needed its own segmentation and personalization logic, not just a marketing campaign.
Step 1: User segmentation. I segmented cardholders across multiple dimensions — engagement patterns during the carding journey, credit profile, credit limit received, proximity to affiliated partner stores, and eligibility for specific consumer-durable loan categories. These weren't demographic segments — they were behavioral and financial signals that predicted what kind of activation nudge each user needed.
Step 2: AI/ML recommendation model. I fed these persona insights into an AI/ML recommendation model that personalized the post-carding product journey for each user — surfacing relevant consumer product deals, partner cashback offers, nearest affiliated stores by location, and the best available OEM offers for that user's profile and credit limit.
Step 3: A/B experimentation. I ran multiple A/B experiments across marketing campaigns and in-product journeys, testing personalized card detail surfaces and card-benefit awareness touchpoints. Inputs included competitor research, user VOCs, and industry insights combined with the NBFC partner's existing deals and offers data.
Step 4: UX decisions grounded in data. Every design decision in the post-carding experience was driven by experimentation data — not by assumption about what a typical cardholder needed.
Results
| Metric | Improvement |
|---|---|
| Month-1 card activation rate | 2x improvement |
| Month-11 card activation rate | ~50% improvement |
| Consumer-durable loans driven | Significant GMV from NBFC-affiliated partner stores |
| Card issuance at peak | Scaled to peak monthly issuance within one year of MVP launch |
Key Takeaways
Issuance and activation are two different products. The mistake most card programs make is treating post-issuance as a marketing problem. It's a product problem — one that requires its own user segmentation, journey design, and experimentation loop.
Proximity matters more than most PMs think. Knowing that a user is within a short distance of a partner store is a more powerful personalization signal than knowing their credit score alone — because it changes what action you ask them to take and when. Location-aware personalization was one of the highest-lift levers in this case.
Partner data unlocks personalization that internal data alone can't. The NBFC partner's deals and offers data, combined with our own user signals, created a richer personalization surface than either side could have built independently. The partnership integration wasn't just a compliance requirement — it was a product asset.