One team wobbles for a fortnight and gets on with it. The other doesn't. Here's what separates them.

Two teams in the same organisation get hit with the same thing. A restructure, a market shift, some change to how the work gets done.

One of them wobbles for a fortnight and then gets on with it. The other doesn't, and it isn't because the people are any less capable. On paper you'd struggle to tell them apart.

It's common to try to explain this with the things we all assume. Stronger personalities over there. Better luck. A more experienced leader. None of that holds up once you've watched them 'adapt' a few times, because it doesn't explain why a group who handled last year's upheaval perfectly well came apart under this one.

Most of us have built reliable teams. They do good work under the conditions the work was designed for, and that takes years and it's nothing to sniff at. Resilient is a different property. It's what those same people can do once the conditions stop holding. When the brief is vague or half-finished, the priorities move on Friday, the process doesn't cover this, and nobody has told them what to do yet.

It's easy to assume resilience is something a team either has or doesn't, a quality of the people in it. It isn't. It's a practice, and practices have structures underneath them, and structures can, and should, be built on purpose.

Software teams are the right place to look for those structures. Not because of the technology, but because these teams have had to work this out deliberately, under pressure, for the better part of twenty years, and they've left the workings on the board. They operate in permanent conditions of incomplete information, so they built a set of habits designed to shorten the distance between deciding what to do, noticing where they're going well and wrong, and adjusting. From the outside those habits look like ceremony. But, really, they're scaffolding, and the difference matters more than you might think.

You can install every good team or leadership structure in the catalogue and get nothing back.

Amy Edmondson's work is what everything in adaptive teams rests on. Studying hospital teams in the late nineties, she found something she initially thought was an error in her data: the better-performing nursing units were reporting more mistakes, not fewer. They weren't making more. They were in climates where reporting one was survivable, so the mistakes surfaced, and surfacing them was exactly what made those units better.

She called the condition psychological safety, and it's been misread ever since as a synonym for being nice to each other. Which it's not. It's the shared belief that you won't be punished or humiliated for speaking up with a question, a concern, a mistake, or a half-formed idea.

The bit that matters most for anyone worried this means going soft: Edmondson pairs safety with high standards, not instead of them. Safety with low standards produces a comfortable team that doesn't do much. Standards with no safety produces anxiety, and people who hide problems until they're too big to hide. You want both at once, which is harder and considerably more interesting.

You can hear it, if you listen for it:

None of these structures make a team adaptive on their own. They hold a shape open, and what fills it is people deciding it's safe to think out loud in front of each other.

You can start with one of them this week. That's how it gets built, in the ordinary run of a month when nothing much is happening and none of it feels urgent.