01How a session works
- Charter — declare what you're going to test and why. "Explore the new payment flow on Safari, looking for broken validation."
- Time box — usually 30 to 90 minutes. Long enough to dig in, short enough to stay focused.
- Test as you learn — try things, observe behaviour, follow hunches.
- Record findings — bugs, observations, questions, "this seemed slow." Not all findings are bugs; all of them are useful.
- Debrief — what did you learn, what should the team do about it.
02Why it earns its keep
Scripted test cases catch regressions in known behaviour. Exploratory testing catches what nobody thought to script — edge cases, integration weirdness, "this looks broken on Firefox," accessibility gaps.
For new features especially, scripted coverage doesn't exist yet. Exploratory testing is the only way to actually exercise the feature before you decide what's worth automating.
03Related
04Frequently asked questions
Is exploratory testing just "playing around with the app"?
No. Mature exploratory testing has a charter (what to test and why), a time box, and recorded findings. The unscripted part is the path through the product, not the discipline.
Does exploratory testing replace scripted test cases?
No. Scripted cases cover known behaviours that need consistent verification. Exploratory testing catches what scripted cases didn't think to look for. Both belong in your process.