Simple, yet quite profound 

Instead of forcing tests to fit rigid technical constraints, it’s about understanding how a user naturally moves through an app and letting that guide automation.

We all know automated tests can become less reliable over time. As projects grow and deadlines loom, test maintenance often slips down the priority list. This leads to fragile tests that demand constant fixing, sapping energy and confidence. The session introduced a refreshing alternative: customer-centric testing approaches that truly put the user first.

Rather than relying on static element identifiers—those input fields, buttons, lists that can change and break tests—this approach uses computer vision to detect UI elements dynamically. And it doesn’t stop there. The system continually retrains itself on-the-fly, adapting to changes and learning domain-specific behaviours, such as the tricky “select one from many” interactions that often cause headaches in traditional testing.

What really makes this stand out is that these automated tests interact with the app exactly like a human would: through the screen, keyboard, and mouse. Even more impressively, they take into account the psychology behind user interactions, considering how people think and behave when using the software.

This isn’t just a technical improvement. It’s a more empathetic, user-focused way to automate—making tests that are not only smarter and more resilient but kinder to the teams maintaining them. By aligning test strategies with real-world usage, organisations can expect more reliable and maintainable automation in the long run.