Effective Retrospectives
As with just about everything I find I’m writing, or being asked these days, there is no one right answer. This is also the case with running an effective retrospective.
But, there are some good practices that will help you get started. As with everything in the Agile world, I hope you’ll use these as starting points, and learn what work for you and what doesn’t. And figure out ways to make it more meaningful, and more valuable, for your teammates.
So first of all, let’s review why we do retrospectives?
We do them to find better ways of getting our work done. That’s pretty much it. This isn’t really the time for us to talk about our work itself, but rather about how we’re doing our work. That may seem odd. But we have demos (or sprint reviews) where we review the work we’ve done. The retrospective is our time to review how we’re working. You’ll see why this is the first point I’m going to make in a moment.
But before we get into a retrospective, it’s important to decide who gets to attend. Often, it’s the people actually doing the work. I tend to recommend that managers not be a part, because it can often cause people to not talk as freely. It’s also important to remember who the retrospective is for. It’s for the team.
Now, the team may need to rely on others to accomplish some of the things that come out of a retrospective. It’s okay to share action plans that come out, but not such a good idea to share individual comments. If you want me to speak freely, it’s got to be a safe environment where I can say what’s on my mind. If I’m concerned that what I say will make it to people outside my team, I’m going to be less likely to say what’s on my mind, and that may prevent us from figuring out what’s really going to help our team improve.
And one more thing. There’s a Retrospective Prime Directive. Even though I am a Star Trek fan, I didn’t come up with it. And unlike Kirk, it’s not something that you can violate without impacting the trust of your teammates. One of the most obvious fears people have when first trying a retrospective is that the ritual will become a negative gripe session, interspersed with blame and counter blame.
Clearly such an event will not contribute to very much learning.
The key to a constructive successful ritual is assuring that all the participants adhere to the Retrospective Prime Directive. It says:
Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, an the situation at hand.
At the end of a project, or period of time, everyone knows so much more. Naturally we will discover decisions and actions we wish we could do over. This is wisdom to be celebrated, not judgement used to embarrass.
This comes from Norm Kerth, and is found in his Project Retrospectives book.
A retrospective is a key part of any improvement cycle. It’s in Scrum (Inspect & Adapt). It’s part of Kanban (Improve & Evolve; Agree to pursue improvement through evolutionary change). It’s in the Agile Manifesto (“At regular intervals, the team reflects on how to become more effective, then tunes and adjusts its behavior accordingly.”). And it’s part of the PDCA-cycle (it’s explicitly in the Check section, but informs everything else in that cycle!).
It’s important. Got it. But how does someone actually run an effective one?
There are lots of options. But all of them tend to follow the following steps:
1. Set the Stage > 2. Review previous actions > 3. Gather Data > 4. Generate Insights > 5. Decide What to Do > 6. Close the Retrospective
Setting the stage is very much like laying out an agenda. Why are you there? How long will you be there? What are you hoping to get out of the time you’re together?
Review action items from the previous retrospective. This is where you close the loop from your previous retro, but reviewing the progress you’ve made and the impact it’s had on your team. It’s a chance to celebrate the improvement you made. It’s a check that what you planned to do got done, and made a difference. If it didn’t make a difference, is it still something you’d like to work on? Don’t just assume that because it was something you wanted to work on a week or two ago that it’s still the thing you want to work on. We hate to leave work unfinished – make sure you’re not falling victim to the sunk cost fallacy. If it’s not done, or didn’t have the impact you hoped, use that as an input to the next steps, to determine what your priority will be.
Gathering data can often start with hard data. What milestones or objectives were achieved, or not. What decision point were made along the way? Depending on your team, you’ll have different metrics you can use, but this is only part of the story. Feelings and emotions make up the other half, or often more than half! It’s important to capture this, and often the “soft” data will drive some amazing insights and conversations. While many of us shy away from the f-word, there are lots of ways to understand the emotions. And they’re important to helping uncover ways for your team to improve their way of working together.
In generating insights, we want to ask “why?”. We can use tools, like 5Ys, Ishikawa diagrams, root-cause analysis… There are lots of ways to use the data that’s been collected to try to hone in on opportunities for improvements. But be careful about jumping to solutions too fast! A solution to solve a problem too early in the process is often like a band-aid – it’s not actually fixing the source of the problem. It’s important to explore how events, behaviours, and circumstances contributed to whatever it is you’ve identified.
When you’re deciding what to do, you’ll want to come up with a number of possible experiments and improvements. We’re finally at the time when we’re going to narrow our scope to one, maybe two items. Any more than that and it’s unlikely you’ll make traction on any of them. Whatever you decide to do should be made nice and visible for any and all to see in a Continuous Improvement section of your visual management board. This allows you to make sure that work isn’t getting lost while you’re delivering your other work, and allows others to see what it is you’re working on improving. It never ceases to amaze me the number of times someone else is walking by a board, sees what another team is working on trying to improve, and either (a) is trying to improve the same thing, or (b) has already solved it, and can help that team.
When it comes time to close the retrospective, make sure you have a clear way to measure if your improvement experiment is successful at your next retrospective. You’ll want to start by there next time. The learning and action items that have been generated belong to the team. Not the manager. Not the coach. It’s up to the team to own the improvement for their own working environment. And sometimes, yes, that will mean getting help from others. Remember, only the improvement experiment, not the information as to how you got there, needs to be shared beyond the participants. And remember, you can (and should) make improvements to your retrospective, too.
A lot of this material comes from Esther Derby and Diana Larsen, and their “Agile Retrospectives” book. Really worth picking up, IMHO.
There are lots of tools available for retrospective formats. I’d encourage you to look at ways to make them different, so they don’t become boring and routine. Here are just a few of the many, many available: