The Testsigma Quality Awards are in their second year. Quality work is invisible when it goes well, and nobody sends a note about the release that didn’t break. Every World Testers Day we fix that for a few teams.
Nominations came in across all four categories this year, from customer QA teams of every size, and the panel judged them on the strength of the story and impact. The Quality Bar Award goes to the change that made releases safer or testing leaner, whether that’s one initiative or a whole function. It went to Spacewell, and to the QA team Arun Subramanian leads.
Spacewell’s QA team knew exactly which part of the release they dreaded. The last sprint of every two-month cycle was regression, two weeks to retest everything, and two weeks was never enough.
Table Of Contents
Nobody Knew What Still Worked
Spacewell makes the software that organisations run their buildings and workplaces on. Web, mobile, API, and mobile stretches to kiosk devices. The legacy products had Selenium automation. The newer mobile and web apps had none at all, just functional testing by hand.
So regression became risk-based out of necessity. Run a build, log the bugs, get fixes, retest, move on. Build one might cover 400 test cases and build two only 100, and the other 300 went untested into a release that shipped anyway.
Four or five people carried all of it. Three or four critical bugs reached production every release, and 100 to 115 support tickets came back each cycle. The team priced the obvious fix, building it out in Selenium or Playwright, and the honest estimate was close to a three-year project. The bugs weren’t going to wait three years.
Two Specialists Became a Team of Seven
Spacewell’s fix started with people, not scripts. Automation had been sitting with two people, in their own lane, separate from the five doing functional testing. He merged the lanes.
Seven people now do functional testing, automation, API testing, execution reporting and maintenance as one job. That was the actual goal: getting automation out of being a specialist skill that only two people in the building could touch.
Arun was clear about what he didn’t want measured. Not script count. He wanted the scripts to mean something as feedback.
Critical Bugs, down to about One a Year
The automated regression suite runs at about 1,500 test cases. Mobile runs against every new build. A full run takes 16 to 20 hours.
Out of those 1,500 cases, a typical run turns up one or two failures on critical or high-priority tests. Maintenance sits under 10 to 15% most years.
Critical production bugs have gone from three or four a release to about one a year. Support tickets are down 50 to 60%. Both numbers moved together, which is the tell that quality actually improved rather than the reporting changing.
Arun’s own summary: the team just has to worry about quality now, not automation.
“Winning ‘The Quality Bar’ award is a proud moment for Spacewell. Working with Testsigma, our QA team has scaled automation across releases and cut our testing cycles significantly. This recognition belongs to them for their consistency and relentless effort. Congratulations to the entire QA team.”
Ammar Mannan, CTO, Spacewell
Why Spacewell Won the Quality Bar Award
What tipped it for the panel was how the Spacewell team solved it. The obvious version of this project is a bigger automation team. They went the other way and made automation everyone’s job, which is why the numbers held after the first year instead of drifting back.
The Part Worth Stealing
Count how many people on your team can write an automated test. If the answer is two, that number is your real coverage ceiling, whatever your dashboard says.
Congratulations to the QA team at Spacewell and Arun. Three or four critical bugs a release down to about one a year is the kind of number the rest of the business feels without ever knowing who to thank.
