Customer storyHow Navadhan Capital cut regression testing from 6 hours to 30 minutes
A five-person QA team replaced manual regression with plain-English automation—12 suites across 8 device configs, running daily.
Regression time
6 hours → 30 minutes
First automated suite
2 days → 1 hour
QA time freed per week
20+ hours
Device configurations
8, running daily
Test suites built
12
Navadhan Capital is a fintech lender focused on bringing formal credit access to underserved and rural borrowers across India. Their mobile platform handles loan origination, KYC verification, and repayments—flows that are both business-critical and heavily regulated. For a company at this intersection of finance and compliance, a broken release isn't just a bad user experience. It can mean failed disbursements, incomplete KYC trails, and real money not reaching the people who need it.
Manual regression couldn't scale
Navadhan runs a five-person QA team. Smart people, good instincts—but every release, they'd sit down and manually walk through the same flows. Login, user creation, KYC documents, payments, account states. The full regression took six hours. Every single time.
That's not a team problem. That's a process that was never built to scale. They knew automation was the answer. They'd looked at Appium. The honest estimate was two months to get something meaningful running, and that assumed they had someone who could write and maintain the scripts—which they didn't. So the six hours stayed. And every week, 20-plus hours of the team's time went into re-running tests instead of actually testing anything new.
Six hours of regression per release meant QA was always the last thing holding up a deploy. The team had no room to get ahead of it.
Frameworks like Appium needed Python or Java to build and maintain. Engineers who knew the code were building the product; testers who knew the product couldn't write the scripts.
Because all their time went into regression, exploratory testing barely happened. Nobody had the hours to go looking for problems that weren't already on the checklist.
Plain English tests the whole team could run
Quash's plain English testing meant a QA engineer could describe what the app should do in plain language and just run it. No framework setup, no XPath selectors, no dependency on someone from engineering to write the scripts.
What the team built
In the first two months, Navadhan Capital's QA team built 12 test suites covering the flows that matter most:
User profiling
Creating borrower profiles, field validations, data integrity across states.
Document upload
KYC document submission, format handling, upload failure scenarios.
Sanction field appraisal
The loan assessment section, field-level checks, edge cases specific to rural borrower data.
E-sign (web version)
The digital signing flow, session continuity, fallback handling.
Authentication
Login, session expiry, re-authentication.
Payments
Disbursement flows, repayment tracking, failure handling.
Daily regression runs across 8 device configurations—the phones their borrowers actually carry.
Before Quash
After Quash
“We’re now running multiple test suites regularly — every couple of days, and it’s become a proper part of how the team operates rather than something ad hoc. Smaller things have helped too, like not having to hardcode usernames into prompts every time — we can just manage them in the test data library and pull them in. It doesn’t sound like much but it does make the day-to-day smoother. But honestly, it’s the consistency of execution that’s been the biggest shift for us.”
What changed beyond the numbers
Regression testing is important, but it only catches things that broke. What it can't do is find the problems nobody thought to write a test for—the weird edge case in a borrower's document upload, the session that drops mid-way through an e-sign, the appraisal field that accepts something it shouldn't.
That kind of testing takes human attention. And before Quash, Navadhan's team didn't have much of it left after running regression. Now they do. The automated suite handles the known. The team handles everything else.
For a lending platform where a broken flow means a borrower doesn't get their loan, that shift matters more than the time saved. Every release goes out with the confidence that the critical paths have been checked—not by someone sitting there for six hours, but by a suite that runs every day across every device configuration they care about.
