Scaling a fintech app in India with a two-step CPI → CPA model
An illustrative walkthrough of how a personal-finance app could move from install volume to funded accounts.
This case study is a composite example written to show how a campaign is structured. It does not describe a real client or real results.
Challenge
A personal-finance app launching in India needs volume to learn, but its business depends on funded accounts — users who complete KYC and make a first deposit. Paying purely for installs risks buying users who never verify their identity.
Strategy
Start on CPI across a small set of Android-first sources to learn which placements produce users who complete KYC. Define the value event as "KYC completed and first deposit made", measured by the MMP and sent as a postback. Plan to move proven sources to CPA on that event once enough data exists.
Execution
- Localized store listing in English and Hindi
- Tracking links with sub-IDs per placement and creative
- Traffic rules: no incentivized installs, no misleading claims about returns
- Daily caps per source during the learning phase
- Weekly review of KYC completion by sub-ID
Optimization
Sources are compared on KYC completion and funded-account rate, not install cost. Placements with high install volume but low verification are paused. Sources that consistently produce funded accounts are offered a CPA on the deposit event, with a payout set so their EPC stays competitive.
Result
The campaign structure shifts spend from the top of the funnel to the event that matters to the business, and gives partners a clear, measurable target. This is an illustrative scenario; no results are reported.