Joined by hundreds of QA experts around the globe, Quality Horizon was truly a remarkable event, where a lot of thought-provoking conversations were sparked. We believe that participants of the Summit have taken home many actionable insights to improve their day-to-day work.
If you haven't had the chance to join the summit, here's a quick recap. There were 5 sessions in total:
- The state of QA in 2025
- AI in QA: The Hype vs Reality
- Scaling QA in Enterprise: A practical playbook
- Achieving ROI: The business case for QA investments
- The future of QA
Let's dive in!
📹 You can also watch the recording here.
1. The state of QA in 2025
Speakers:
Despite modern tooling, only 11% of QA teams consider themselves "mature", and 82% still rely on manual testing. Why? Interestingly, most aren’t lacking tech. They’re lacking alignment, strategy, and time to grow.
Too many teams jump into automation without fixing foundational issues: poor testability, lack of CI/CD, no vision for maintainability. Leaders should not budget not just for tools, but for the process and people behind them. As Hemalatha put it:
“Leaders think by putting into automation, it’s faster release… without bringing the various components that will help it. It could be in terms of the right automation strategy, right skill set, infrastructure… legacy systems… fear of change."
Both speakers emphasized intentional planning and realistic evolution, not rushing:
“It’s not like, you were going in a bullock cart and suddenly you want to go to Mars. That’s not gonna happen.”
– Hemalatha Murugesan
In other words, adopting automation should happens synergistically with team up-skilling, as per Sergio put it:
“Testers are not just people that execute tests… They are quality enablers. They should build skills around communication, leadership, awareness of risk. [...] I’ve seen testers looking at pull requests as a way to learn about the risks being embedded into the software.”
- Sergio Freire
2. AI in QA: Hype vs. reality
Speakers:
Both speakers emphasized that AI is far beyond the buzzword phase. It's now a productivity multiplier. Michael Giacometti shared real-world applications:
“We’ve piloted AI-based tools that can generate test cases directly from user stories… instead of running every test suite after a code change, AI can suggest what is the most impacted test — saving us hours, sometimes days.”
He advised starting small:
“Pick a small project, do a pilot, measure the impact, then scale.”
Lyndon D’Souza added that AI is finally helping QA realize its long-promised potential:
“Redundant manual test cases are now created by AI… That’s giving quality engineers the perspective of being customer-first.”
He explained how his team reduced 2–3 days of manual test case writing to a few hours using AI, noting:
“AI tutored them along the way… it mentored us, it told us where we were going wrong.”
On whether team size affects AI success, both speakers said absolutely not. Lyndon explained:
“We started with just three testers using Katalon with AI… it does not matter the size of the team [...] We did not use the big bang approach. So I I think, irrespective of the size of the teams, I think each and everyone should be able to use that as a leverage.
Michael added:
“The learning curve to become productive is about two days… the tools are easy, approachable, and drive introspection into your processes.”
Finally, Lyndon emphasized:
“Maybe you need to be friends with AI rather than treating it as an enemy.”
Michael was candid:
“AI is powerful, but it’s not empathetic… It still needs domain knowledge, human judgment, and soft skills to guide it.”
Michael also shared his company’s internal training strategy:
-
Weekly peer storytelling sessions
-
Bi-weekly innovation demos
-
Tiered training (AI 101 → 301 via platforms like Coursera)
-
Certification as part of promotion criteria
3. Scaling QA in enterprise
Speakers:
The session kicked off by tackling a harsh truth from the 2025 State of Software Quality Report:
- 55% of QA teams say they don’t have enough time to complete their testing
- 41% struggle with skill gaps.
Despite rising investments in automation, 43% still hit a wall at the implementation stage. So why scaling QA still fails? Priyanka Halder explained that many teams delay investing in quality until hypergrowth forces their hand, by which time technical debt is entrenched and testing is reactive, not strategic.
Jason Lee reinforced this, saying that automation is still viewed transactionally, rather than as a strategic investment.
“A lot of the measurement of automation is around the coverage... but at times, the failure of making the business case is measuring the ROI in a more business and delivery context — like reduction of timeline, improve time to value, and risk conversation through repeated automated testing.”
Next, Priyanka spotlighted two root causes:
-
Lack of strategic vision: Teams don’t plan for what quality should look like in 1–5 years.
-
Misconceptions about tooling and skills: Many assume automation = any test script, or tools = magic fixes.
Jason then emphasized the importance of maturity assessments—not just in QA, but across the entire SDLC. He urged leaders to consider how well development, data, and environments support testing from day one.
Finally, Priyanka’s QA Playbook for Enterpirse includes:
-
Maturity assessments across SDLC
-
Hotspot analysis to target bug-prone modules
-
Tooling decisions based on talent + goals
-
Prioritized automation on high-risk, repetitive areas
-
A culture of upskilling + AI enablement
4. Achieving ROI: The business case for QA Investments
Speakers:
The session kicked off with a challenge from the 2025 State of Software Quality Report:
- 26% of QA teams say they still don’t know how to measure ROI from their automation efforts
- 45% are dissatisfied with their vendors
So why are some QA teams struggling to prove value while others are seeing real business impact? Suraj Jadhav argued that the fundamental issue is mindset. Many still treat QA as an end-of-cycle checkpoint, rather than an enabler of speed, experience, and revenue.
“QA is your first end user. If you treat them like a cost center, you’re missing the biggest lever to improve customer experience and retention.”
Andrew Hammond echoed this and pointed out a simple but powerful metric:
Manual hours saved through automation = real ROI
His team has helped reduce weeks of regression testing to overnight cycles. And when PMs start asking for automation proactively? “That’s when you know you’re doing it right.” Suraj also broke down what separates high-ROI teams from the rest: They measure what matters.
Instead of only tracking coverage or defect counts, they tie QA to:
- Number of successful product launches
- Reduction in production support tickets
- Faster time to market and customer adoption
However, Suraj cautioned against jumping into tools without a plan:
“No-code tools can empower teams—but only if there’s a roadmap. Tools are just enablers. You still need people, vision, and strategy.”
Andrew added that many partnerships with vendors fail not because of the tool, but because of mismatched expectations:
“If you're finishing each other’s sentences, that’s a partnership. If you're asking what went wrong every sprint, it’s not.”
5. The future of QA
Speakers:
Danielle Forier kicked things off with a lesson from experience:
“The earlier a QA person is brought into discussions, the better the outcome. We ask questions no one else does. And when we’re included in the build and design phase, we help expose risks long before testing even begins.”
Rosie Sherry echoed this from a community perspective, emphasizing the role of systems thinking:
“Testing touches every part of the business—product, customers, ecosystem. QA professionals should learn to see the bigger picture and advocate for themselves. That’s how we step up and earn a seat at the table.”
Joe Colantonio backed this with a real-world example from GE:
“QA or testers used to go visit customers… watch them in the workflow… and go, oh my gosh, these UX designs… this is not what the radiologist really wants.”
Alex Martins extended the idea of alignment, calling for QA to gather metrics from both internal systems and end-user behavior:
“Correlating what matters to the users… to what we are testing in the release cycle itself is huge.”
When asked about how high-performing teams align QA with product and support, Danielle described UAT cycles and QA-led knowledge transfer to support teams:
“Even before production, we get another look—did we truly deliver what the customer is expecting?”
Rosie brought in a powerful analogy between community and quality:
“Community is care and, like, quality is care as well.”
Joe advocated for internal QA communities to break silos:
“We used to have a weekly QA or testers call… someone would say ‘Hey, we solved this problem,’ and another group would say, ‘Oh, that’s great, let me implement it.’”
Alex recalled brown-bag sessions that organically fostered inter-departmental collaboration:
“It started organically just by giving each other visibility… bring your laptop, hook it up, and share.”
And here's the final advice:
- Danielle Forier: Fight for your seat at the table. And don’t give up.
- Rosie Sherry: Trust your gut. If something isn’t working in your testing practice, change it. Use AI to solve real problems, not just follow trends.
- Joe Colantonio: Don’t outsource your thinking to AI. It’s your peer, not your replacement.
- Alex Martins: QA isn’t just a checkpoint—it’s a catalyst for transformation. And in the age of AI, your human insight is more essential than ever.
.svg)
