Exploratory testing

Also called: exploratory session, session-based testing

Exploratory testing is testing where design and execution happen together: the tester learns the product and decides the next check from what the last one showed.

In exploratory testing the tester is not following a script. They hold a charter, such as what happens to a booking when the connection drops, and follow what they find. It is the fastest way to learn how a changed area behaves, and it finds the problems nobody thought to write a case for.

It is not unstructured clicking. A session has a charter, a timebox and notes, and its findings are written up as defects, questions or new checks.

I spend exploratory time around whatever a release changed, and leave the stable core to repeatable checks. Pairing during a session, with a developer or a newer tester, makes it more useful still.

Notes on this

More on this subject: Release readiness and regression. Written by Marius Ene, senior QA consultant. See how I work with teams or browse all terms.