Alternative Approaches

In the past few years I have ended with the following approach. We started to call it "Testing with a stranger".

  1. Welcome everybody (5 minutes)
  2. Introduce the concept of touring, i.e. using Michael Kelly's Touring Heuristics. (10 minutes)
  3. Provide the product, ask testers to pair with each other, and tour the application to make yourself familiar with it. (20-30 minutes)
  4. Run a debriefing, collect results on a flipchart or mindmap visible for everyone. (5 minutes)
  5. identify additional sessions. Where would people go deeper if they had the time? (5 minutes)
  6. Switch pairs, ask them to pick a detailed mission, and give it another round. (20-30 minutes)
  7. repeat until energy becomes low.
  8. Do a final debriefing with next steps, i.e. how would you use this stuff at your work? (10-15 minutes)

That usually can be run in sessions between 75 minutes and 3 hours. I wouldn't go shorter than that. Longer is always possible by running additional sessions. Usually the energy starts to sink after three sessions though.

Petteri Lyytinen shared the approach he used in the Finish public dojos:

  1. start with a brief introduction round, followed by a brief description of what exploratory testing is (as it was unclear to some, 15min total)
  2. divide the participants into small groups of 3 people and show them an application to be tested (10min)
  3. tell the groups to come up with testing ideas i.e. what would they test (free discussion between groups allowed, 15min)
  4. tell the groups that due to time constraints they will not be able to test all of their ideas so they will need to prioritize their top-3 testing ideas - what do they consider the three most important aspects to test (5min)
  5. switch the prioritized ideas between groups so that none of the groups gets to test their own ideas (5min)
  6. break, encouraging people to discuss freely (10min)
  7. test the application, one group at a time, based on the testing ideas they were given (15min per group)
  8. review round of the testing performed, by the group that originally came up with the testing ideas that were just tested
  9. repeat steps 7 and 8 until all groups have both tested the ideas they got and reviewed the ones they came up with
  10. feedback, questions, sauna, beer, free conversation about anything and everything (approx. 1 hour)

That took approximately 3 hours with the 1st hour for preparations, 2nd hour for testing and reviewing, and the 3rd hour for the social part. Many of the participants thought that this was a welcome variation to the "usual" testing dojo format and they quite enjoyed it. I liked it too but it would probably not work if it was done too often since the switching missions part is supposed to come as a surprise.

Key learnings: prioritization under time pressure, quick adapting to unexpected changes, reporting results in a clear yet concise manner