A manager at a past client of mine once had a new request come in. The new request would impact multiple different teams, that would all have to make changes to their individual pieces and then integrated. Because it was potentially a large change, he asked all the teams involved to come up with an estimate and they came back with a total estimate of ten to twelve weeks.
He decided to run this as an experiment to see what he could learn about this multi-team collaboration. He hand picked a group of individuals - one from this team, perhaps two from this other team, and so on. He put all these people into a room together and set the boundaries of the experiment. “I want to see how far you can get in two days. What will you learn? How much can you complete? What lessons will we be able to take from this?”
Three hours later, they came back to him, having completed the work. Not merely having talked through a design or an approach; having completed the work, end to end. Something that they had estimated would take 10-12 weeks.
How is this possible?
First, there were no hand-offs. Every skill that they needed, was already in the room. There were no delays while we wait for another team do do their part. They didn’t have to wait for any approvals as the people who could do that were already in the room as well.
Second, they had only one thing in progress.
The key here is that they had no dependencies and no multi-tasking on other work. Everything they needed was right there and this was their top priority for two days. They were 100% focused and they delivered.
Is this a fluke or is this a teachable moment?
I’ve introduced hundreds of teams to ensemble programming. When they do that for the first time, the vast majority are able to complete a story end to end in about a day and a half. These are teams that can rarely finish a story that fast when working in their usual ways.
In the last couple of weeks, one of my clients held a two day big-room planning session with all of their stakeholders present. They completed more in two days than they would have been able to complete in months, had they been working in their normal way. In my experience, this is typical of big-room planning.
I think the clear lesson here is that this is reproducible when we satisfy the two criteria of focusing on one thing and breaking the dependencies. Put all the right people together and let them work.
Note that putting all the right people together does not require that they be physically in the same room. I’ve seen amazing results when everyone is fully remote. The key is active collaboration, not where one is sitting.
The inverse is also true. Putting people physically together doesn’t provide any of this benefit if they aren’t actually collaborating. I see too many companies forcing people into close proximity and then letting them work individually. All the benefit is lost when we do that.