Automation or Manual Testing

"Manual Testing is a wonderful way to truly understand software!" – Richard Seidl

The triumph of automation is unstoppable, driven by three major factors. And now, AI is here. Providers of automation tools have dollar signs in their eyes. Companies have well-funded AI budgets, and we are about to witness a new generation of tools that will interact with and understand the test objects in entirely new ways.

1. 1. For 20 years Agile has been fueling the market with the mantra: "We must run our regression tests in every iteration."

2. 2. Alongside the major tool manufacturers, software developers have created many custom automation tools, often open-source, for their daily work.

3. 3. The automated testability of frameworks has improved massively. Locators instead of pixel tweaking, resilience, and dynamic surfaces. It’s happening.

And I believe AI will become even more interesting in another area: test data management.

I know companies today that require large test data departments. These teams work diligently to meet the company's test data needs. The scale is enormous: dozens of completely different systems across multiple development, test, and quality environments, massive data sets, GDPR requirements, and then a test team shows up saying, "We'd like to test historical data and time travel." Yeah. The leverage of AI combined with automation can be huge.

Is the future of manual testing at risk?

Is the future of manual testing at risk?

And the manual testers? Is it over for them? Off to the beach or the job center?

Well, I don’t think so. The role of the manual tester will change – as it always has – but it won’t disappear. At least not right now. Why do I believe this? Three reasons suggest that manual testers will still be needed:

1. Complexity: Business and technical requirements are becoming increasingly complex. Just bringing together the right topics, departments, and people requires someone who can intuitively grasp the connections.

2. Trust: As long as there’s something at stake, we want humans to tell us that everything is okay. An automated test report isn't enough. It will take time before we fully trust machines when it comes to quality.

3. Thinking outside the box: Humans make mistakes – when requesting, developing, and testing. Things can always go wrong. And sometimes just intuitively and exploratively looking at things reveals the unforeseen.

But: If you don't move with the times, you'll be left behind.

Today, it's being decided whether you, as a manual tester, will end up on the sidelines or become a quality influencer in your project or company. Technologies like AI and test automation are already here – and I highly recommend learning and utilizing them for your work. Man and machine as the dream team :-)

So: Let’s make better software!

Yours,
Richie

Richard Seidl is a software testing expert, Agile Quality Coach, and author. He has seen all kinds of software in his career: good and bad, large and small, new and old. Software so beautiful it could make you cry and software so bad it curls your toes. For him, it’s clear: those who want to create excellent software today must think holistically about the development process. People, context, methods, and tools – only when everything works together does a mindset emerge for unleashing potential and driving innovation.