*Names and details in this case study are illustrative examples based on common scenarios RoboQA is built to solve.*
The message came in on a Wednesday afternoon.
Kavya Menon, founder of Stitch & Knot — a small online store selling handmade leather accessories — had a product launch scheduled for Thursday morning. Her developer had gone off sick that day. Her paid ad campaign was already live, driving traffic to a site she hadn't tested on mobile in two days. And two days ago, they'd pushed a round of UI changes.
"I was panicking," she told us afterward. "Nothing huge changed in the push — some layout tweaks on the product pages, some CSS cleanup. But I hadn't actually opened the site on my phone since before the changes went up. I had no idea if anything was broken."
She found RoboQA, signed up in under a minute, and typed her first test prompt:
"Test the full checkout flow on mobile. Check the product page layout, the promo code field, and the checkout button. Use a mobile viewport."
Fifty-four seconds later, she had a report. It had found three bugs she didn't know existed.
Bug #1: The Checkout Button Wasn't Registering Taps on Mobile
The most critical issue: on iPhone Safari at 390px viewport width, the "Add to Cart" button on the product detail page was completely unresponsive. It rendered correctly — right colour, right size, right position on screen — but tapping it did nothing at all. No animation, no cart update, no error. Just silence.
RoboQA's interaction log flagged it within the first scan:
"Element #add-to-cart is present and visible at 390px viewport but does not respond to click events. Possible z-index or pointer-events conflict with an overlapping element detected above the button layer."
Kavya forwarded the report to her developer, who diagnosed the issue remotely in under five minutes: a pointer-events: none rule had been accidentally applied to a parent container in the latest CSS push. One line of CSS in one file had silently broken every single mobile purchase on the site.
If the launch had gone ahead untested, every mobile user who tried to buy would have tapped a button that did nothing. More than 70% of Stitch & Knot's traffic comes from mobile. The entire sale event would have effectively been dead on arrival.
Bug #2: Price and Button Overlapping on Product Pages
The second bug was visual — but bad enough to look broken and confuse customers on smaller phones.
At 375px viewport width (the iPhone SE display, still widely used), the product price label and the "Add to Cart" button were overlapping due to a flexbox layout collapse. The price text rendered on top of the button, making both elements nearly illegible and the button functionally untappable in that region.
RoboQA caught it during its layout validation pass:
"At 375px viewport, .product-price bounding box intersects with .btn-add-to-cart. Elements overlap by 34px vertically. Visual collision detected in product detail layout."
This one had slipped by because the team always tested on their own current-generation iPhones — 390px wide and above. The 375px breakpoint had never been checked. It only broke at that specific size, which meant it could have been there for weeks without anyone noticing.
There are still millions of users on devices at that viewport. One of them was probably about to get served an ad pointing them at a product page that looked like it had been broken by a child.
Bug #3: The Promo Code Field Wasn't Accepting Input on Android
The third bug was the subtlest of the three. The promo code input field on the checkout page looked and worked perfectly on desktop and on iOS. But on Android Chrome at 412px viewport, tapping the field did nothing. The mobile keyboard never appeared. Users could stare at the empty input field indefinitely and type nothing.
The RoboQA interaction log captured exactly what happened:
"Input field #promo-code-input does not receive focus on tap at 412px viewport (Android Chrome simulation). Field appears non-interactive. Mobile keyboard not triggered. JavaScript focus event appears to be failing silently on this platform."
The root cause: an autocomplete="off" attribute combined with a custom focus handler that was throwing a silent JavaScript exception on Android's WebKit implementation. The error never surfaced in a desktop browser console. Desktop worked fine. iOS worked fine. Android Chrome silently swallowed the focus event every single time.
This is exactly the kind of bug that survives code review, passes desktop QA, and only surfaces when a real user on a real Android device tries to apply a discount code at checkout and quietly gives up.
From Report to Fixed: 90 Minutes
All three bugs were diagnosed, fixed, and deployed within 90 minutes of Kavya receiving the RoboQA report. The Thursday morning launch went ahead without incident. The campaign ran clean. No angry customers, no support tickets about dead buttons, no sales lost to a broken checkout.
"We would not have caught this manually," she said. "Even if I'd spent an hour testing on my phone before the launch, I probably wouldn't have checked the 375px breakpoint specifically. And I definitely wouldn't have caught the pointer-events issue — that's completely invisible to a manual tester. You'd just assume your tap didn't register and try again."
That's the core problem with manual mobile testing. You test on the devices you own. You check the screen sizes you happen to think of. You run through the flows you personally use. And you miss things — not because you're careless, but because the surface area is just too large to cover consistently by hand.
A proper manual regression of a checkout flow — covering 4–5 viewport sizes, two browsers, the happy path and a handful of edge cases — takes an experienced QA engineer somewhere between two and four hours. And even a thorough tester might miss the Android focus bug that only shows up on a specific keyboard implementation in a specific browser.
RoboQA found three real, production-affecting bugs in 54 seconds.
What the Prompt Looked Like
In case you want to run a similar test on your own store, here's exactly what Kavya typed:
"Test the full checkout flow on mobile. Check the product page layout, the promo code field, and the checkout button. Use a mobile viewport."
That was the entire input — one sentence describing what to test, and the store URL. RoboQA selected viewport sizes, ran the interactions, checked the layout, and returned a report with the specific elements affected, what the observed behaviour was, and where to start looking to fix each one.
The full list of what gets checked in a run like this — interaction testing, layout validation, mobile compatibility, performance signals — is on the Capabilities page.
If you're about to ship, a 60-second test is cheaper than finding out something is broken from a customer.
Try RoboQA →