Why “who we are” is structure, not soft culture.
We’ve spent this series arguing for speed and autonomy: shorten the lag, push decisions to the edge, let units act on what they see. Which should make you nervous. A fast, decentralized organization is also one where lots of people are making consequential decisions without checking with the centre. What stops that from becoming expensive chaos — everyone moving quickly in slightly different directions?
The answer is identity — and the claim of this essay is that identity is not soft culture. It’s load-bearing structure. It’s the thing that makes speed safe.
Start with a quieter but equally important shift, in how we measure whether an organization is doing well. The default scoreboard is financial: revenue, margin, quarterly earnings. The trouble is that these are lagging indicators — they describe fitness that already happened. They tell you the organization was healthy last quarter, not whether it can survive the next surprise. The cybernetic standard is different: viability — the capacity to maintain your identity, keep creating value for the people you serve, and keep thriving in an environment that won’t hold still. That’s almost exactly Beer’s definition, in plainer words, and it’s a direct rebuke to running an organization by its rear-view mirror.
Now connect viability to identity, because they turn out to be the same point seen from two angles. As structure loosens and speed rises, what lets a decentralized unit act without central clearance is a clear, shared, genuinely understood answer to two questions: who are we, and what will we not do? A team that deeply understands the organization’s purpose and boundaries can make a fast local call and be confident it’s the call the whole organization would endorse. They don’t need to ask permission, because identity has already answered the question that permission exists to answer.
That’s why identity is structure, not decoration. A strong, hard-to-copy sense of purpose lowers the coordination cost of every single decentralized decision. It’s the cheapest possible alignment mechanism: instead of a rule for every situation or an approval for every move, you have a shared understanding that lets thousands of independent judgments cohere. It even doubles as a competitive advantage — a distinctive identity is hard for a rival to replicate, and it guides adaptation without anyone at the centre directing traffic.
The compact version: identity is what makes speed safe, and viability is what identity is for. Strip out identity and decentralization drifts — autonomy with no shared centre becomes fragmentation. Strip out viability as the goal and you optimize the rear-view metrics while the organization quietly loses the capacity to survive. The two together are the missing half of every “move fast” prescription.
The honest caution is real and uncomfortable: identity can ossify. The very thing that holds a fast organization together can harden into the rigidity that adaptivity was supposed to escape. Held too tightly, identity becomes “this is how we’ve always done it” and refuses the change the environment demands. Held too loosely, it stabilizes nothing and the drift returns. And here’s the part neither the practitioner accounts nor the classic model fully resolve: how does an identity itself adapt — change what the organization is — without dissolving into something with no centre at all? That’s one of the genuinely open problems in this whole area, which is precisely why it’s worth arguing about in a room full of people who’ve tried it.
The takeaway for anyone running a fast, distributed team: don’t treat “who we are” as a values poster. Treat it as the most important piece of infrastructure you have. It is what lets you grant autonomy without losing coherence — and the work of keeping it both strong and adaptable is real work, not an offsite slogan.
Next week: the newest pressure on all of this — what AI actually changes about where humans are essential.
Part of the countdown to Metaphorum 2026 — “100 Years of Stafford Beer” (17–19 September 2026, Alliance Manchester Business School, Manchester), where I’m convening a half-day “Cybernetic Futures” workshop [workshop slot — TBC]. Programme, themes, and registration: conference2026.metaphorum.org. Abstract deadline: 17 July 2026.