Yesterday you caught a total lying while the subtotal on the same screen told the truth. Today nobody tells you where to look.
Exploratory testing gets a bad name because people confuse it with clicking around until they get bored. It is nothing of the sort. It is skilled investigation: time-boxed, focused, note-taking work where you use judgment to decide what to try next based on what you just saw. The two bugs you have found so far were guided hunts. This time the map is blank.
Human curiosity is still one of the strongest testing tools we have. Written cases prove the promises you already knew about. Exploration finds the problems nobody wrote down, and every product has them.
The shape of a real session
- Charter. One sentence, written before you start: what you are exploring and why. “Explore the Tumble Kitchen cart and discount flow for surprises around emptying, re-adding, and re-applying, thirty minutes.”
- Timer. Thirty minutes. When it rings, you stop. The box forces focus.
- Notes as you go. What you tried, what you saw, what surprised you. Ugly is fine; a trail is the point.
- Debrief. Two lines when the timer ends: what you covered, what you found, what you would explore next. File any bugs. Name what you skipped.
That debrief is what turns “I clicked around” into evidence a team can act on.
Workflow
- Write charter
- Set timer
- Explore + note
- Debrief
What to vary when you feel stuck
The difference between a rich session and a shallow one is knowing which levers to pull:
- The data. Search for an empty string, an emoji, “PATTY” in capitals, a word with a space before it.
- The sequence. Apply the discount before adding anything. Place an order with an empty cart. Walk the whole checkout with a pretend card and read the confirmation slowly against what you ordered. Sign in after filling the cart and see what survives.
- The timing. Search twice quickly. Refresh the page mid-load. What happens to your cart after a refresh, and is that what a shopper would expect?
- The aftermath. After every interesting action, check yesterday’s lesson: does every number on the screen still agree with every other number?
- Your own reactions. If you hesitated, a user will hesitate. If a label made you re-read it, write that down. “This surprised me” findings are often the most valuable thing in the debrief, because nobody else is positioned to notice them.
Do this today
Run the full session against Tumble Kitchen: charter, thirty minutes, notes, debrief. I am not telling you what is there to find this time. Some of what you notice will be bugs, some will be questions worth raising, and learning to tell those apart is the skill.
When the timer ends, run my Visual QA Checklist against the store as a second pass; it is 67 checks for what only a human eye catches, and comparing its findings with your session notes shows you what your eyes skipped.
Tomorrow: where all of this work sits inside a real team’s week, meeting by meeting.