Retrospective format: “Do or do not”
Learning retrospectives
I’ve written extensively in the past about retrospectives, but this time I’d like to touch on a slightly alternative approach to them — a retrospective format to be used by experienced teams.
My earlier thoughts on the subject came from my perspective as a consultant; I was accustomed to a life of jumping into client sites that were in the midst of an ‘agile transformation’ as part of a team that was there to help deliver as well as coach. In such situations somewhat ‘heavy’ and prescriptive process can be useful.
I still stand by that advice, as per the concept of shuhari it’s essential to first understand a thing completely before you can then detach and transcend it.
Unlearning retrospectives
“You must unlearn what you have learned” — Yoda
For teams that are relatively experienced and in a state of ‘flow’ - e.g. communicating well, stable progress, having regular retrospectives etc, some of the ceremony and the bulk surrounding a retrospective can be safely bypassed. Here at carwow we hold regular retrospectives and the team is in constant communication with both each other, and the various other teams in the company through the day. In this setup, the retrospective is just another opportunity for the team to talk and focus on improvement, but it’s not the only one.
We also started to realise that we had gone a long way towards solving some of the more ‘trivial’ or at least simpler problems that the team faced. We were left with more complex topics that needed a more focussed and in-depth approach than we could provide in a retrospective; hence we decided to change things up.
Retrospective redefined:
We experimented with a new bare bones retrospective; gone were the multi-section star-fish and the humorous whiteboard illustrations (although I am quite fond of ridiculous retrospective backstories). Our new format consists of two parts.
Part one
Start with one simple question…
“What do you want to do?”
That’s it. The entire retrospective is focussed on the changes that the team feel are needed to increase their effectiveness. It forces the individuals to break down what they’re feeling into actionable items* that they believe will have a concrete effect on the team. Everyone writes their answers onto some post-its and throws them up on the board. I’d say give the team between 5–10 minutes for this, and encourage them to talk to each other and share ideas through (it doesn’t have to be 5–10 minutes of silence).
*note: it’s important to remind the team that an acceptable action could be to arrange a deeper discussion on a topic. Remember that a trigger for this format was the elimination of the simpler problems. A suitable post-it could read “find a way to solve [issue]” or “have a bi-weekly catch up on [topic]”.
Part two
Add one last question to the board…
What are we going to do?
Part one gives us a list of all the possible actions that the team feels would be of benefit right now. However this doesn’t automatically mean that the team has the capacity or desire to actually do all of these things.
Step two takes this list and filters it down into the most important actions, as decided by the team.
A typical retrospective usually contains a round of voting after the brainstorming. Each individual in a team is given a certain number of votes, and then they vote on the topics that they would like to discuss; the idea being that that there are usually more topics than can be covered in the allotted time, so voting allows the facilitator to direct the conversation towards the higher priorities.
In this new format, by placing a vote for a topic, you are assigning yourself as an owner for that action. Instead of saying “I would like to discuss this topic further”, you say “I care enough about this topic to dedicate time to it in the upcoming iteration”
No team member has the capacity to pick up too many actions, so the garden naturally weeds itself. Note: you could enforce some kind of limit (i.e. each person can only sign up to X actions) but I wouldn’t start with that restriction by default - most people should already be aware of the dangers of signing up to too much, and we’ve not had to do this so far at carwow).
Part three
There isn’t one. That’s it. the whole thing should take less than half an hour, accounting for some discussion and chatting around the actions (although the real work happens outside of the retrospective).
Health warning
I do not advocate regular or repeated use of this format. This can be a great way to achieve a targeted/action focussed retro; from my experience at carwow the efficiency has been a welcome change of pace. However it does come at a cost.
Stagnation is the simplest, and often most common problem. Keeping the format changing and fresh is important for maintaining engagement. As is listening to your team and adapting it to what works for them.
Retrospectives can offer a chance for creative exploration and can be at their best when they help a team uncover issues that they weren’t even aware of before they walked in the room. This format can also restrict the ability to raise an issue for which you have no solution for yet (although I try to mitigate this by suggesting the ‘action’ is to have a deeper investigation into the problem).
It’s also essential to remember that the state of a team only ever represents the state at a given moment in time. So when I said earlier that we had solved many of the smaller/simpler issues the team was facing — this doesn’t mean that they were solved for good. The longer, more detailed retrospectives should always be present; this format just offers an efficient approach when needed.
Interested in making an Impact? Join the carwow-team!
Feeling social? Connect with us on Twitter and LinkedIn :-)
