The Testing Lean Coffee Experiment
At the start, the facilitator creates a kanban board with a backlog, WIP, and “done” columns on the table. Each participant writes down her desired topics, one per sticky note, to place in the backlog. Each person gets a chance to quickly explain her topics. The group dot-votes to prioritize the backlog. And then, they’re off on a time-boxed discussion of the top priority topic.
When the timer politely dings, the group can elect to continue for another block of time or call that discussion done and switch to the next topic. I was surprised at how the group gives each topic its due then moves seamlessly to the next. I was also surprised how many great ideas can be shared within a short time period. The group decides when it’s time to stop. Participants can use a “fist of five vote” to denote how useful they found the discussions. They can also share “aha” moments and takeaways to wrap up.
Ryan Freckleton and I live within easy driving distance of each other on the Colorado Front Range. We both want to build the agile testing community here and had the idea to try a “testing” version of a Lean Coffee. We chose a coffee shop halfway between our residences and scheduled a time.
The right people always show up at a Lean Coffee. Despite a small group size, we covered some big topics: how to get a team to buy into conducting regular code reviews, pairing for exploratory testing, keeping a sustainable pace, doing specification by example for API commands and end points, and teaching test-driven development. We also brainstormed on ways to market our Testing Lean Coffee idea and build a regular group of participants.
The Lean Coffee format isn’t just for user groups. It could be used as a way to share ideas among teams and practitioners within a company. A development team could try it as a format for a retrospective.
Our first Testing Lean Coffee experiment showed me that this format is an excellent way to discuss a wide range of testing issues and generate many useful experiments to try.