First let’ s start with a disclaimer. What follows is mostly for my personal further reference. Because I think it is something that improves the daily life of software developers, I think it’s worth sharing it.
Why do retrospectives? Because we want to improve the process of developing in the current project. Therefore it is important to gain some feedback of the current process. Retrospectives should be done after every iteration and at the end of the projects. And the goal of a retrospective are some actual TODOs.
Every Retrospective consists of five parts:
- Set the Stage (5 - 10′): Introduction, defining the goal
- Gather Data (30 - 40′): this is the base of everything what follows. There are some rules:
- don’t interpret
- be personal
- don’t judge
Some methods: Timeline, Mad Sad Glad etc.
- Generate Insights (30′): what was good, what should change. The important thing here is to find the real reasons and not stay with the symptons.
Some Methods: Five Whys, etc. - Decide what to do (15′): Here so called experiments are defined. Normally there are only a few, only what you can manage in the next iteration. An experiments consists of:
- Goal
- Action
- Key Figure (something measurable)
It is important, that you can decide if the experiment was a success or not.
Close Retrospective (10′): That a short retrospective of the retrospective
The time above is for an iteration retrospective, provided that the iteration is about 4 weeks max.
