Analytics is supposed to tell you where users drop off, not who they are. Yet in most FinTech audits we run, personal data has already leaked into Google Analytics 4: a phone number in a URL, a PAN in an event parameter, an email address set as the user ID. Nobody meant to send it. It simply rode along.
For a lending app, a neobank or a payments company, that is more than a tidy-up job. Google Analytics terms do not allow personally identifiable information to be sent to it, and personal data in India now sits under the Digital Personal Data Protection Act, 2023. This checklist shows where PII usually hides and how to keep it out for good.
Analytics data is copied widely. It flows into reports, exports, BigQuery tables, dashboards and the inboxes of everyone with access. Once a PAN or an account number lands there, it is very hard to say where it has travelled.
Most leaks come from a handful of predictable places. Check each one in your own property:
?mobile=98xxxxxx or ?email= after a form step or an OTP redirect.1. Scan what you already have.
Run explorations on page location, page title and every custom dimension. Search them for 10-digit mobile numbers, email addresses, 12-digit numbers and the PAN pattern of five letters, four digits and one letter.
2. Strip sensitive query parameters.
Remove them before the hit is sent, in your tag manager or, better, in a server-side container. Never rely on reports filtering them afterwards.
3. Use an opaque user ID.
Send an internal, random customer ID. A hashed phone number can often be reversed, so treat it as personal data too.
4. Clean page titles and errors.
Keep titles generic ("Loan summary") and send error codes, not error messages that echo what the user typed.
5. Exclude search terms you do not need.
If people search for account details, stop recording those parameters.
6. Set a sensible retention period.
GA4 lets standard properties keep event data for 2 or 14 months. Choose the shortest period that still answers your questions.
7. Respect consent.
Implement Consent Mode v2 so tags behave according to what each visitor allowed.
8. Limit who can see raw data.
Give most people Viewer access, keep editing rights to a few, and review the user list every quarter.
9. Know how to delete.
If PII does get in, use GA4 data deletion requests, then fix the source so it cannot happen again.
Every new screen, A/B test or vendor script can reintroduce a leak. Add a PII check to your release checklist, run the scan in step one every month, and give one person clear ownership of analytics governance.
This article is practical guidance, not legal advice. For questions about your obligations under the DPDP Act, speak to your legal or compliance team.
Clean analytics is not only about compliance. When your data contains no personal information, you can share dashboards freely, move faster in audits and trust your numbers. The fixes above are small, but together they make your analytics something you can defend.
Want a second pair of eyes on your funnel? Our FinTech team maps every step from ad click to funded account and prices each leak. Get your FinTech growth audit.