Why manual QA still matters when AI writes the tests.

The short answer

Manual QA testing still matters because the bugs that frustrate users most are usually experience bugs, and automated tests rarely catch them. AI and automated testing are valuable, and we use both. But every piece of CedarLabs work also goes through a human functional QA team that uses your application the way your customers will.

What does AI and automated testing miss?

It has become increasingly standard to let AI handle testing, and to a point it does that well. Automated tests can check thousands of conditions in seconds, run on every change and never get tired.

But automated tests only check what someone thought to ask. They confirm that a button submits a form. They do not notice that the button is hidden behind the keyboard on a small phone, that the success message disappears before anyone can read it, or that a five-step process feels like fifteen.

After more than ten years in technology, I have seen the same pattern again and again. The first bugs to reach customers, and the ones that cause the most frustration, are purely experience-related. They pass every programmatic check and still fail the person using the product.

What is the three-test QA standard?

Quality assurance is the one thing CedarLabs refuses to compromise on. All work we deliver is reviewed through three main tests.

1. A test for cleanliness

We make sure your systems are not cluttered and that each solution is optimised, evidence-backed and performant. Clean code is easier to test, cheaper to change and less likely to hide problems. This matters even more when AI tools help write it, because generated code can work while still being heavier or more complicated than it needs to be.

2. Programmatic and automated testing

We always write test cases for programmatic testing, along with automated tests for technical QA. These protect the logic of your application on every change and catch regressions early.

3. Manual functional QA

This is what sets us apart. We run a dedicated functional QA team that crawls your application by hand. It is a person, not a script, judging whether the software actually works for the people using it.

How does manual functional QA work?

Our functional QA happens in two passes:

• Targeted user journeys. First, QA walks through every user journey related to the recent change. If we updated checkout, they check out: with valid details, invalid details, slow connections and the unexpected paths real users take.
• Regression across the application. Second, QA moves across popular, and sometimes all, user journeys to make sure the change did not quietly break something elsewhere.

When something needs fixing, QA records the steps taken, what happened and what was expected, so the issue can be repeated and resolved quickly. Once a change passes all three tests, we agree on the release with you and manage the deployment.

Why will CedarLabs not compromise on QA?

We build fast. Our software factory and AI-assisted development mean ideas become working software in days. That makes quality assurance more important, not less. Speed without verification just means problems reach your customers sooner.

Your users do not see your test coverage. They see whether the app works when they need it. That is the standard we hold every release to, and it is why a person will always check the experience before we call anything done.

Frequently asked questions

Is manual QA testing still necessary with automated and AI testing?

Yes. Automated and AI-driven tests are excellent at checking that code behaves as specified, but they often miss experience problems such as confusing flows, broken layouts on real devices and steps that technically work but frustrate users. Manual functional QA catches those issues before customers do.

What is the difference between functional QA and automated testing?

Automated testing runs scripted checks against code to confirm expected results. Functional QA is a person using the application the way a customer would, following real user journeys and judging whether the experience actually works.

What is regression testing?

Regression testing checks that a new change has not broken something that previously worked. CedarLabs combines automated regression tests with manual walkthroughs of popular, and sometimes all, user journeys after a change.

What quality checks does every CedarLabs release go through?

Three: a cleanliness review to keep systems uncluttered, optimised and performant; programmatic test cases and automated testing for technical QA; and manual functional QA that covers the changed user journeys first and then looks for regressions across the application.