A Room Built for Decisions

A room built specifically for decisions still needs someone in it willing to disagree.

We have started to run a weekly forum for anything that needs a senior decision — the whole point being that it doesn't get lost in the usual meeting shuffle. It's a genuine structural fix, and it works. But this week the room itself was thin — enough people away that a few proposals didn't get the scrutiny they deserved. That's a different problem to solve than "where do we make decisions," and it sent me back to a book I'd been meaning to finish, on how people actually make good decisions under uncertainty. A few ideas from it are worth carrying into the room.

Resulting. We judge decisions by how they turn out, not by how they were actually made. A good call can still go badly; a bad one can get lucky. Confuse the two often enough and you stop learning anything — you just get better at telling stories about outcomes after the fact.

Self-serving bias. We credit our wins to skill and blame our losses on luck — and, just as unhelpfully, do the reverse when judging other people's decisions. Left unchecked, it makes it almost impossible to learn honestly from either.

Habits, not willpower. Fixing bias isn't about trying harder in the moment — it's about changing the routine around a familiar cue and reward, so a better habit forms without having to relitigate it every single time.

CUDOS. A useful charter for any group trying to think clearly together:
Communism — share information openly rather than protecting what makes you look bad
Universalism — judge evidence by the same standard no matter where it came from
Disinterestedness — stay alert to your own conflicts of interest
Organized Skepticism — actively invite dissent rather than politely avoiding it

Devil's advocate. The phrase comes from the Catholic Church, which used to appoint someone specifically to argue against a candidate's sainthood. Deliberately assigning the dissenting view, rather than hoping someone volunteers for it, turns out to matter.

10-10-10. Before a decision, ask how you'll feel about it in ten minutes, ten months, and ten years. It's a quick way of pulling your future self into a conversation your present self would otherwise have alone.

Backcasting and premortems. These two are opposites, and I’d only come across premortems previously. Backcasting starts from a future where everything's gone brilliantly and works backward to see what had to be true to get there. Premortems start from the opposite place — imagining everything's already gone wrong — and work forward from that to mitigate it before it does. Personally, I've always found premortems more fun: people are far more willing to talk enthusiastically about what's gone wrong than about what's gone right.

None of this is exotic. It's mostly a discipline of building dissent and honest hindsight into a room on purpose, rather than hoping enough smart people sitting in it produces those things by accident. Meanwhile, the less abstract stuff kept moving too — a cyber assessment return closed out on short notice, a genuine team effort, and a goodbye to a colleague who's been one of the steadiest, most reliable voices on the programme.

Lesson learned

A dedicated forum for decisions isn't the same thing as a good decision. Structure gets you the room; deliberate dissent and honest hindsight are what actually have to happen inside it.