Six Weeks, Then Rest

Basecamp, the software company behind the project-management tool of the same name, builds its products in six-week cycles. Then it stops. The next two weeks are a deliberate cool-down in which teams review, adjust, and rest before the next cycle begins. Ryan Singer, who helped shape the approach over 15 years at the company, put it into a book called Shape Up, and the six weeks is not arbitrary: long enough to finish something meaningful, short enough that everyone can feel the deadline coming.

I keep thinking about that cool-down because almost every organization I coach discovers the fast loop and then tries to live inside it forever. It's an understandable mistake. A small team running short, honest experiments learns faster than any plan can, and once a team of around ten is stable enough that a couple of absences don't stall it, cycles that once took three or four months can shrink to six to eight weeks. In my book I call that a Hyper-Loop. It really is a different rhythm, and the energy it produces is real.

It is also a rhythm people can hold for a while, and "a while" is doing real work in that sentence. What I've seen work is roughly three cycles at the fast pace, followed by a deliberate lengthening: eight to twelve weeks, then three to four months, then back toward the steady seasonal rhythm a community can sustain without anyone having to think about it. Not stopping. Slowing on purpose, before exhaustion makes the decision for you.

Basecamp put the pause inside the cycle. I would argue for putting it inside the plan. Either way the principle is the same, and AI makes it more urgent: when the tools make the next experiment cheaper, the pressure to run the next one immediately gets stronger. The teams that keep learning are the ones that decide, in advance, when to slow down.

From Dancing in the Wind: An Adaptive Leadership Framework in the Age of AI.

Where in your organization has a sprint quietly become the permanent pace?