So, I recently had a thing pop up in my LinkedIn notifications that said something along the lines of, “You’re one of a few experts invited to answer some questions about: Agile Retrospectives” While I usually ignore these types of things, if there is one area of the Agile process that I feel like I’m uniquely qualified to talk about, it’s going to be the good ol’ fashion Retro. And while I have no idea if my responses to that particular advice article will ever be seen or visible, I can at least write about them here on my blog, where they certainly won’t be seen or visible.
Why am I qualified to talk about Retros?
Despite the fact that Agile development was founded in 2001, when I was a mere 10-year-old boy concerned more with how I was going to fight off the imaginary orcs in my bedroom than sound software development, I spent much of my early career in 10’s knowing nothing about it. Until one day, as an insignificant little Jr. Engineer on the team, somebody said, “We’re going to start doing this thing called Scrum, and it’s going to come with some extra meetings.”
Now, I could spend a whole post talking about the good and (mostly) bad parts of Scrum. And I could also talk forever about how Agile and Scrum are completely different things that too often get conflated. But that’s not what I’m here to talk about.
What I am here to talk about is a certain one of those meetings that captivated my imagination from the beginning. The Retrospective. This weird little break from my normal job that let me, even as an inexperienced little Jr. Engineer, have a voice on my team treated with as much respect and dignity as even our most senior member on the team. That feeling would not be forgotten.
As I progressed in my career, I always held the Retro to be the best of all of the Scrum ceremonies, though I found most to be lacking in substance after that first one. I felt compelled throughout my career to better define, better perfect, and better make use of this odd little meeting that once captivated me so. I became the “Retro Guy” at multiple companies I worked for, someone trusted to get actual results or break a team out of the funk of the usual dismalness that this meeting can sometimes take. I credit it for launching me into an eventual leadership position, and I credit it even more for my success in those leadership positions.
So what qualifies me to talk about Retrospectives? I’ve been thinking about them for over a decade, I’ve since ran hundreds of them, and I’ve discovered some important things that work and some important things that don’t.
On Boundaries and Tone
By and large, the most important thing to consider when setting up your retrospectives is to set some boundaries and establish the tone of the meeting. This shouldn’t just be another ceremony meeting that you run your team through for the sake of doing it. Retros are personal, sometimes confrontational, meant to truly provide room for discussion and debate. It’s important to make that known up-front, and to ensure your team knows that this is a space where they are free to speak their minds and their hearts.
If you are struggling with Retros or just getting started, I would encourage running a “Retro of Retro’s” type meeting to establish these boundaries. Ask your team things such as what they want to gain from this process, what they want the “Rules of Engagement” to be when it comes to problems with other teammates, how long do they want to spend on talking about this stuff? Absolutely central to the Retrospective is the team involved in it. If there is little to no interest in this meeting from the expected participants, you should scrap it.
Tone, specifically, is something that I’ve found to be most effective when running a Retro. This is a meeting that should be treated as a group coming together around a restaurant or bar table to voice their grievances and concerns in good company with people willing to hear them out. As such, I always encourage joking, playfulness, and off-topic subjects during my Retros. These things, while seemingly unproductive, can help diffuse the tension in the room and get people more open to speaking about the true problems they are facing.
The Role of Facilitator
There’s no title or role that is exclusively responsible for running a Retro. I know from experience it could be just a seemingly insignificant Jr. Engineer, or it could be a Senior VP looking for feedback. Regardless of who’s running and organizing it, the role of the facilitator doesn’t change.
You are there to listen and collect the feedback. Your participation should be mostly limited to devising the questions that need to be asked of the team, asking follow-up questions, taking notes from the conversation that takes place, and broadly staying out of the way unless you’re needed. This is often the one and only place for the team to express themselves and get things off their chest. Let them have the space to do that.
Too many Retros that I have been in become dominated by the facilitator. An uncomfortable silence while the team is thinking or processing through something should be welcome, not immediately taken as an opportunity to interject your own thoughts or move on to the next topic. Most problems can’t be solved effectively by the first off-the-cuff thoughts the team might have, and some team members may be slower than others to churn on the topic at hand. Give them the space and time needed to truly think through the situation.
And speaking of those team members who may be a bit more quiet than others, you as a facilitator should be acutely conscious throughout the entire Retro of who is speaking a lot and who is speaking very little. For every Retro dominated by a facilitator, there’s many more dominated by one or two team members. As a facilitator, it’s your job to ensure ALL opinions are heard and taken into consideration, not just the quickest and loudest. You shouldn’t hesitate to cut the more domineering personalities off and ask the quieter people directly if they have any input. Being on the shy-er side myself, it’s often the case that being quiet during a meeting doesn’t mean you have nothing to contribute or add to the conversation, but instead can just be a case of feeling like you can’t get a word in between those who speak more quickly and more forcefully than you.
Understanding the personality types and the dynamics of the people you are running the Retro for is absolutely key to running successful and productive Retros. This takes practice and conscious effort to improve upon. You won’t always get it right, especially if you’re facilitating for a group you have little experience with, and that’s ok. But if you keep this in mind and continue to improve on it, each of your subsequent Retros will get better and better.
The Retro Structure
There are an absolute myriad of ways to run a Retrospective. Many teams I’ve been on have gotten stuck in the “What Went Well?”/”What didn’t go Well?” style. While this can certainly identify and help with some types of issues, it doesn’t always lead to the best or most relevant conversations. And it usually ends up with a dwindling version of the same topics and same conversations happening over and over again from week to week. AKA, boredom and disinterest.
If you’re looking to run an engaging and productive Retro, you need to tailor the format to what is relevant and what can lead to the most discussion. The questions and the prompts you ask of the team should mostly change between every meeting (with repeats allowed, but almost never back-to-back) and be custom tailored to try to address the needs of your team. This will help to keep things feeling fresh, ensure you are covering a multitude of topics and concerns, and prevent the Retro from just becoming another pointless waste of time that your team doesn’t want to show up to.
While I linked previously to a whole host of pre-pared Retro styles you might run, you should also not hesitate to be creative and roll your own Retro style or Retro-adjacent meetings to talk about, honestly, really anything. The Retro format is a great breeding ground for conversation and input, if ran properly, and can be used for a whole host of problem solving in a team-focused business environment.
Some of my favorite, more unique Retros that I’ve run, which have produced fantastic results:
- The one word Retro – Describe your feelings about work in 1 word. We’ll then go around in turn and talk about why you chose that word. A good chance to surface feelings and thoughts that might otherwise not be captured through more specific questions.
- The Team Values Retro – What do you value being on a Software Engineering team? Does the rest of the team agree? I intend to write a separate article on this one, as it has been one of the greatest drivers of team cohesion and comradery that I’ve ever seen.
- The Knowledge Gap Retro – What areas of the code do you feel strongest with? What areas of the code do you feel weakest with? A great Retro that can help identify and start breaking down Silos.
- The Retro of Retros – Are you getting what you want out of our Retros? Do you feel heard during our Retros? Do you feel like the things you’re bringing up are being addressed?
One majorly important thing to whatever Retro structure you run is that the people involved are allowed to add their thoughts and opinions to your questions asynchronously and potentially even anonymously. I’ve been involved in many Retros and have NEVER seen work, the “Let’s just get together and talk it out,” style of Retro. You should have a board of some kind, you should have a submission process of some kind, and you most certainly should have a period of quiet time where the team can truly think about the questions being asked and to contemplate the other submissions.
Conclusion
I love Retro meetings, at least the ones that I have been involved in running. I know there is a ton of value that can be extracted when they are run well, run with purpose, and run by the right people. However, Retros are not a requirement or an end-all-be-all solution to every Software Engineering team problem. You should listen to your own Team on the effectiveness and usefulness these types of meetings.
The real success of the Retro meeting is recognizing the humanity and dignity behind your teammates, and that comes with also respecting, “We don’t want to do this.”
Soucie AI License
Use of this work for generative AI research or models comes at cost of $1,000,000,000, paid in full to the author Brandon Soucie upon use or upon discovery of use. For all other uses, this work is licensed as follows:
Let’s Talk About Retros © 2024 by Brandon Soucie is licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International 
