Facilitating a Retro en Masse

About a month ago, My current client asked me to help with an interesting challenge. They wanted to me to facilitate a retro for them… for the entire department. That would mean trying to coordinate about 24 people, in the same room, in what might be a tense conversation. As the magnitude of the request sank in, I started planning.

In a group that size, a couple problems come to mind immediately. How do we avoid having the softer voices in the group from being drowned out? How can we keep the group focused on the topic, rather than just griping? And of course, if the entire department would be there, that includes management. How can we encourage people to talk, despite the cooling effect of having management in the same room?

Oh! and just for good measure, I had to take the department’s emotional condition into consideration as well. The release had been frustrating, at best. The teams hadn’t had a real retro in at least 6 weeks. People were tired. How can we best mitigate the potential negative outputs? After all a retro isn’t much use if it doesn’t suggest places for change and improvement.

Here my gut reaction offers some insight. ‘There’s no way I can effectively encourage conversation with 24 people, by myself’. I didn’t have to do this alone… In fact I couldn’t. If I wanted to truly serve the client I would need help. So I got to work.

The first draft of the plan, called for 6 groups , each with a facilitator. These groups would discuss a common topic individually, and then share any interesting findings with the whole. Then the groups gets a chance to iterate on any interesting ideas from the others and share again. Effectively, I would be facilitating 6 distributed retros on the same topic. Then I’d have them cross-pollinate at a given interval.

After I got approval from the client to make the attend, I started selecting my facilitators. Now, I am a consultant. While I am on one of the delivery teams, I can still come across as an outsider. As a result, I needed to pull from and train client employees to be facilitators. The fact that they were coworkers would address the possible trust gaps. And it should provide the feeling that the client truly owned the retro. The last thing we needed was for people to feel like it was some activity forced on them by the consultants.

I’ll discuss how I trained the facilitators in a later post. But I want to highlight an important interaction we had. I needed the facilitators to buy-in to the plan. I needed them to believe it would work, so that those they led might believe too. Aside from providing them a wide execution latitude, I had them help me refine the plan.

Originally, my plan called for 3 phases of retro-discuss-iterate-share. And the original time was about 90 minutes. After discussions with the facilitators, we changed that to 2 phases. Besides this control of the plan, I also had their help in assigning the groups. Being Employees, they had a much better feel for the interpersonal dynamics of the team. Since I wanted to avoid personality conflicts within the groups, it was a natural win-win.

The affect of engaging the facilitators in making the plan was enormous! The best example I can point to was how one of my trainees insisted on being a facilitator. He is easily one of the best Devs in the department. But he doesn’t often reach out beyond development. I had thought he’d be less interested in the people-side of things, like this retro. Still he was among the group I thought I could trust facilitate, so I asked him.

After the brief training, and the applied changes to the plan, I had the group draw straws. I had more trainees than I needed. In the end he objected to be ‘excluded’, and one of the other trainees volunteered for back-up duty. As we continued to implement the plan he became one of the most staunch supporters I had. And he leaned into the effort heavily. Now admittedly, not all that was just from ownership in the plan. But it helped.

Then came the retro. The facilitators and I setup a good sized conference room. We divided the department into their assigned teams. And In the end, we grouped then like this: the Leadership team, the BAs, the Team Leads, and 3 groups of devs. The Dev groups had members from each of the delivery teams.

Then we were off. I opened the meeting explaining the format, and the ground rules. These rules gave the facilitators leverage to guide conversations away from gripe sessions and towards valuable knowledge sharing and brain-storming. Once the groups started, I kinda became a glorified time-keeper.

I rarely had to jump in to help settle a group to keep them on track, because the facilitators had it on lock. Each of them knew the characters in their assign group and how best to interact with them. As a result, I wandered about waiting for the designated time to pass.

There was a moment where I had to cut one of the group members off during the ‘share’ phase. Aside from one moment, the retro flew by without a hitch. When it was over, I held the Facilitators back. Not being in all the conversations, I couldn’t tell what the group feeling was at the end. So I had to ask.

Unanimously, the facilitators believed the retro was a success. Even if we didn’t get a ton of actionable items out of it. They felt it was evident that the department’s mood had lifted as a result. As we discussed, several areas for improving the format surfaced. Additionally, we found several conversations that still needed to happen. As our reserved time on the room drew to a close, I thanked each of them for offering their services, and their time to me. I congratulated them on executing the retro fantastically. And I offered specific compliments to how each had worked with their groups. Then we dismissed.

For my part, I still feel I could have done better. But seeing as this was the first time I’ve had to run a retro from more than 10 people, I guess it was a good first run. I’ve still got a lot more to write about on this. So be sure to come back next week, when I’ll discuss how I trained the facilitators!