Insights / Continuous transformation
Before you break up a good team, spark it again
I lead a small development team that has delivered four software projects in a row. Each one built on the last. None of them was a moonshot.
The roster has changed along the way. Some people moved on to other work, and new people joined as interns. The core held, though, and so did the momentum. Each project started from where the last one finished.
Each one also started the same way, with a Spark session. That's the point of this article. (Well, that and giving a meeting a better name than "brainstorm.")
The expensive part is the team
In 1965, researcher Bruce Tuckman described the stages small groups go through: forming, storming, norming, and performing. In 1977, Tuckman and Mary Ann Jensen added a fifth stage, adjourning, when the work ends and the group breaks up. Tuckman later admitted they couldn't find a perfect rhyme for it.
Every leader has lived through this. A new team spends a long stretch figuring out who does what, arguing about it, and settling on how to work. Then, if you're lucky, it performs.
And then, very often, we take it apart. The project closes. People get reassigned. The next initiative forms a brand new team and pays the whole cost again.
I'm not saying teams should live forever. That would be nonsense. But a team that has just hit its stride is the most valuable tool an organization has for making change. Treat it as an asset worth protecting.
Spend the performing window on purpose
The answer is a shorter loop.
Light a small spark. Ship one meaningful change. Measure it in terms the business cares about: revenue gained, cost avoided, faster service, or faster time to market. Then do it again, with the same team, while it still has its momentum.
That's continuous transformation at team scale. Iterate. (Yes, I see the pun in my company's name. It was free.)
What a Spark session looks like
A Spark session is a short working session that narrows a team's attention to the one change that matters most right now. My team ran one at the start of our first project, and again at the start of each one after it.
It borrows from formats that already exist, especially the Lightning Decision Jam from the design agency AJ&Smart and 1-2-4-All from Liberating Structures. What sets it apart is how it starts and how it ends.
It starts from prompts. People think best when they have something concrete to react to. Bring five to seven starting points, each with a question that points people at their own work. When I put one together for a group of IT service delivery leads, the prompts included:
- Repeat problems. What does your team fix every week that never goes away?
- Manual toil. Which task would your team happily never do again?
- Waiting time. Which request takes longest, and where does it sit waiting?
- Handoffs. Where does work get misrouted or bounced between groups?
- Rubber stamps. Which approval always happens and never changes the outcome?
- Repeat fixes. Which problem always gets the same fix from the same person?
- Metrics. Which number do you report that nobody acts on?
Swap in prompts that fit your world. A sales team, a finance office, and a product group will each have their own.
It ends with a pilot. The run of show takes about 30 minutes:
- Introduce the prompts (3 minutes). Keep it brief.
- Think alone (7 minutes). Each person picks the two prompts that hit closest to home, then names a specific process, a first step, and the biggest blocker. Quiet thinking first keeps the loudest voice from setting the agenda.
- Share (15 minutes). One idea each, around the room. Group similar ideas together.
- Vote and commit (5 minutes). Pick one. Give it an owner, a measure, and a date to report back.
Then pilot it small. Report back. And Spark again.
Where this goes wrong
A Spark session finds the next useful step. Someone still has to make sure the steps add up to a strategy.
It also fails in two predictable ways. The first is the session that produces a beautiful wall of sticky notes and nothing else. (Possibly the most expensive art in the building.) Give every pilot an owner before you schedule the next session. The second is skipping the measure. When budget season arrives, a number is what keeps a team together.
What a Spark session can give you is the evidence for keeping a team together.
Before you break up a good team, ask
- Is this team performing right now?
- What did its last change deliver in revenue, cost, speed, or time to market?
- What would one more loop cost, compared with forming a new team from scratch?
- When is the next Spark session on the calendar?
Spark it again
Four projects in, my team still starts each new one faster than it started the first, because everyone already knows how to work together. That's the return on keeping a good team together.
Spark it. Ship something. Measure it. Then spark it again.
Ready to run your first Spark session?
Tell us about your team and what you want to change. You will hear back from Lawrence.
