Designing remote retrospectives for a productive outcome

On the last Friday of every month, we take some time to reflect and, after our team breakfast, hold a retrospective with the entire Düsseldorf team.

We want to figure out how we can improve. Step by step. We identify problems and tackle them together. We highlight our successes and what we’ve learned, making them visible and tangible for everyone. One for all, all for one.

The role of facilitator usually rotates among a handful of UXIs (that’s the affectionate name we give ourselves). Sometimes someone steps in to facilitate on the spot, and that’s perfectly fine. After all, we’ve been doing this for years: we know and appreciate each other.

But a few weeks ago, a lot changed due to the pandemic. In addition to our daily meetings, workshops, and many other routines, our retrospectives are now also held digitally. And it’s not really possible to facilitate them in a mindful and goal-oriented way on the fly.

A remote retrospective for the entire UX&I team

After holding our first remote retrospective in March with just the Düsseldorf team, we decided in April to host the retrospective with the entire UX&I family, that is, together with our team from Munich. After all, we’re all working from our home offices, and the boundaries between locations are becoming increasingly blurred. The entire UX&I family in one retrospective, possibly 30 participants. You don’t know all of them as well from day-to-day work as you do your colleagues in your own office and on your own project. That’s a real challenge when it comes to designing a productive retrospective. I was eager to design and facilitate this one.

Even though I’ve facilitated many retrospectives in the past, both with teams in the same room and remotely using digital tools, this one felt different. Even before it kicked off on Friday. Everyone was at home. We’d already been working from home for several weeks and were struggling either with being alone or with the daily back-and-forth between childcare and work.

How do I prepare for a retro like this? How do I create a safe space where everyone has a chance to speak and be heard? How do you foster a lively discussion with up to 30 faces on screen? Is just under two hours enough time? How do I ensure we come up with actionable items? These are many questions that can be summed up as: “How do I design a good remote retro for us UXIs?”

I had to prepare them in much greater detail than usual. That means setting many details from the very beginning, defining the rules, finding and adapting a suitable format, setting stricter time frames, and so on.

It wasn’t until afterward that I realized how much time went into this preparation: roughly the same amount of time as the retrospective itself. Today, however, I’m very glad I took the time to do it. Not only did the retrospective yield valuable insights for us as a team, but the feedback I received personally continued to move me long afterward. Thank you once again to the entire team.

If you ever have the opportunity to facilitate a remote retrospective with an exceptionally large, experienced, and diverse team (under special circumstances), the following experiments and insights might be helpful to you.

Preparing for a remote retrospective

I’d like to take this opportunity to briefly describe some of the things I prepared for this retro and how I came to do so. Some of these are probably familiar to you and are considered “standard” or “basic” for a retrospective. However, it’s especially important, particularly with these elements that everyone supposedly knows, to make them explicit, that is, to bring them back into focus, during particularly stressful, hectic, or uncertain times.

Trust as the foundation for what’s to come

There are two things that are always very important to me during a retrospective, both as a facilitator and as a participant: a safe space and an attitude of mutual respect among all participants.

We usually address the former with the so-called “Vegas rule”: “What happens in Vegas, stays in Vegas.” I like to add the following: “Everything we discuss here stays among us, unless we decide together to share a specific topic with the outside world.” We usually assume implicitly that everyone is familiar with the retrospective’s framework. However, in my experience, it’s effective to state this explicitly or post it on the (digital) whiteboard.

The latter, an attitude of mutual respect among all participants, is something we usually take for granted. But we also know that problems and conflicts usually involve people. However, to ensure that we all reach a positive and constructive outcome and actively facilitate learning, it’s important to focus on the action, the process, or similar aspects. I usually express this with a phrase of my own, which I also make visible. “We always assume that everyone on the team makes the best decision for the team and the product at all times, using the resources and information available to them.”

For me, making these two rules, or fundamental attitudes, that I describe here visible is essential for building and strengthening trust.

No matter how much time we set aside, it’s never enough

Our retrospective was scheduled to last 110 minutes. That sounds like a lot of time, but it isn’t.

Anyone who has ever participated in a retrospective knows how difficult it is to use those first few seconds or minutes efficiently to get things going. Doing this remotely makes it a whole lot more complex. We’re not in a room with a whiteboard, bulletin board, or other physical materials. Everyone usually has to get at least two tools up and running: video chat and a digital whiteboard (or a similar collaboration tool). We use Google Meet and Miro.

As a rule, I always set aside about 5% of the time for settling in and another 5% as a buffer. I’ve never been to a retrospective where the participants didn’t want to delve deeper into one topic or another. I’ve included these buffers here as well. Since 5:30 minutes seemed like too much time to me, I decided on exactly three minutes.

That leaves 104 minutes for the actual five phases of a retrospective:

Coming together
Gather data
Gain insights
Decide on actions
Wrap up together

Phases two and three are often the most time-consuming, because this is where the actual learning and exchange take place. But with 24 participants (that’s how many of us there actually were that Friday), everyone needs to be given space to settle in (Phase one).

So I tried to break down the time slots even further. For the check-in (Phase One), I set the following guidelines:

90 seconds for everyone: All participants label their note on the board in Miro and place it in the column that best describes how they’re feeling (sun, clouds, thunderstorm)
20 seconds for everyone: Going around in a circle, each person gets to say one sentence about how they’re feeling that morning. It’s not just work: our personal lives also shape our mood.

Loading this sends data to Miro.

I planned and ran through all the phases in this level of detail, both in case we end up with 20 or even 30 people, and in case things get a bit more intense here and there. Both intuitively and mathematically, it worked out.

Rules are for everyone

To ensure we could make efficient use of the time slots and to help the concept of a familiar environment work well remotely, I established a few ground rules for the retrospective. I’d like to briefly explain three of these ground rules here.

Together, we determine the order of speakers. We’ve been doing this for quite some time in nearly all remote meetings and workshops where the number of participants is hard to keep track of, or when (individual) participants cannot use their cameras for specific reasons. In such cases, we position ourselves clockwise on the corresponding frame on the Miro board as we join the session.

If there is a need for discussion during the session or workshop, each participant may contribute to the discussion in accordance with the established order. Anyone who does not wish to speak may pass the floor directly to the next person.

This order of speakers has the advantage that even quieter participants or those who cannot be seen on camera have an adequate opportunity to speak.

This brings us to the second rule: the principle of artful participation. At its core, this principle involves each participant asking themselves the following question: “Is my behavior the best contribution I can make to this collaboration right now?” To ensure we stay on track and make focused progress, it’s important that all participants embrace this personal commitment and act accordingly.

In some remote meetings, participants may not want to or be able to turn on their cameras. There are reasons for this. Even I sometimes don’t feel like it: in those cases, the audio has to suffice. During a remote retrospective, however, it’s important to see everyone you’re interacting with. Gestures and facial expressions are especially important when we want to figure out where the problem might lie or how deeply an event might be affecting a particular person. We can’t do that without seeing their face and eyes. In this context, my colleague Martin recommends the article “Limits of Online Communication: On the Communication Psychology of Virtual Coachings and Meetings.”

Which game makes us all winners?

Have you ever explored the wide variety of well-known retrospective formats? Retromat alone lists more than 100 activities. What might be a suitable activity for us? Which one leads to a good outcome for all participants?

I think it was the third or fourth format I’d come up with. With this one, I immediately felt it offered enough space to gather insights from the past four weeks (Phase Two), followed by an in-depth discussion (Phase Three) and the development of potential action items (Phase Four). It’s the “Lean Coffee” format. Within a set time frame, participants individually gather topics they’d like to discuss and briefly present them. Everyone has a fixed number of votes and can distribute them among the topics. The topics are discussed within a fixed time slot based on the number of votes they receive. When the session time is up, it’s over.

Of course, this format only works with a fair amount of discipline in this kind of setting. So I tried to set some limits here as well. For example, I limited the number of topics each person could bring up to two: this was a deliberate effort to keep the focus sharp. The time allotted for writing down the two topics and presenting them was also strictly limited.

I chose the Lean Coffee format for our remote retrospective for several reasons:

Anyone can and should bring topics to the agenda. In other words, I hope to see a wide range of experiences and events, as well as topics that affect multiple people.
We can group topics right when we introduce them, which saves us a little time.
Everyone gets three votes, which they can distribute among all the cards or clusters after the presentation. Since we address the topics with the most votes first, the issues that affect the most people are automatically addressed first.
Action items can be (pre-)formulated directly during the discussion of a topic. Every participant knows exactly who will be responsible for the topic even after the retro.
If we make good progress, we can easily cover five to seven topics. Every topic we discuss is valuable and helps us move forward.
Next to topics that couldn’t be discussed at the end due to time constraints are names. These individuals can be approached about these topics even after the retrospective.

All in all, I think it’s a very flexible and rewarding format for a retrospective like this, one that we’re hosting for the first time and where we, as a large team, still needed to reconnect with one another. And, as always, my hope was that everyone would walk away from the retrospective with something valuable for themselves.

Preparation… check. Now what?

That’s a glimpse into how I prepared for the first remote retrospective with the entire UXI family. As you can see, as facilitators or Scrum Masters, we have a wide range of options we can explore in the lead-up to a remote retrospective, whether through clear ground rules, establishing desired attitudes, an adapted format, or various trust-building measures.

If you have any questions, feedback, or even more suggestions, feel free to comment on this article on Twitter. I look forward to discussing this topic with you.

I’m sure you’d also like to know how the remote retrospective went, how I did as the facilitator, whether everything went “according to plan,” and how we designed and used our Miro board, for example. I’ll be writing another post about that in the coming days.