Slime Mold Org Design (Sports Car vs Big Rig)
Coordination cost grows with the square of headcount — drive the vehicle you actually have.
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 93%
Even with a team of good, collaborative, hardworking people, coordination cost scales roughly with the square of the number of people involved. Most companies either fight this or pretend it does not exist. Komoroske's model: treat the org as a vehicle. Small = sports car, steerable by a founder. Large = big rig, which does not mean 'go slow', it means 'pivot less'. Or you deliberately split into a swarm of sports cars and accept visible incoherence as the price of autonomy.
Origin
Komoroske's 'slime mold' deck, written at Google and widely shared across the industry — perplexity's co-founders cited it when describing how they organize. Slime molds are the biological analogy: decentralized organisms that find solutions to problems no one was searching for.
Core principles
- 01Coordination cost grows with the square of the number of people who must agree, commit and actually work together.
- 02You cannot beat the coordination force by willpower — you can only choose an architecture that respects it.
- 03Big rig ≠ slow. Big rig = pivot less, be intentional about steering, invest in program management and planning slack.
- 04The swarm option trades visible coherence for autonomy and antifragility. Both are legitimate — Apple picked coherence, AWS picked the swarm.
- 05Autonomy-focused tech companies are already more slime-mold than they realize; fighting that fact is what hurts.
How to run it
- 1
Measure the vehicle you are actually driving
Count the people who must genuinely coordinate on a decision, not the headcount on the org chart. That number, squared, is your coordination drag. Be honest about whether you are still a sports car.
Pro tip The founder-steering privilege — everyone accepting that you may steer — quietly evaporates as you scale. Notice when it has gone.
Watch out Driving a big rig like a sports car makes you a danger to yourself and others, and grinds the engine.
- 2
Choose the architecture explicitly: coherent big rig or swarm of sports cars
Either accept the big rig — steer less often, more deliberately, with heavier program management and planning slack — or split into autonomous units and accept that outsiders will see that team A and team B clearly did not talk. Decide which cost you are willing to pay.
Pro tip Apple runs the coherence play (the illusion of perfect product coherence) and executes it marvellously. AWS runs the swarm play — fifteen ways to do everything, but a powerful, antifragile suite.
Watch out The one thing you cannot do is want both. Trying to have swarm autonomy and big-rig coherence produces the worst of each.
- 3
If you chose the big rig, pivot less and buy slack
Reduce the frequency of direction changes, invest more in program management and process, and build slack into the planning process so surprises can be absorbed without derailing a shared launch date.
Pro tip Slack is not waste — it is the shock absorber that lets many teams still hit the same launch.
Watch out Frequent 'small' pivots are what actually grind the engine at scale, not any single big one.
- 4
If you chose the swarm, defend the incoherence
Accept and defend that surface-level inconsistency between teams is the deliberate price of autonomy and speed. The payoff is an overall suite that is powerful and antifragile and finds solutions to problems you did not know you were searching for.
Pro tip Decide in advance how bad the incoherence is allowed to be for your specific customer — that is the real dial.
Watch out In consumer products with a single integrated surface, visible incoherence can be genuinely fatal.
In the wild
Apple treats the illusion of perfect coherence across its products as strategically vital and organizes to preserve it — accepting the coordination cost and the reduced pivot frequency that come with it.
→ Marvellous execution on integrated product coherence, at the price of speed and bottom-up autonomy.
AWS made the opposite call — there are roughly fifteen different ways to do everything and the teams evidently do not coordinate. The organization does not treat that as a defect.
→ A very powerful, antifragile product suite where the swarm collectively covers ground no coherent plan would have found.
Common mistakes
Steering the big rig like a sports car
Founders who keep making sharp, frequent direction changes as headcount grows produce whiplash, ground gears and burned-out teams — the vehicle physically cannot respond that way.
Reading 'big rig' as 'go slow'
The constraint is on steering frequency, not on velocity. Big rigs move fast; they just cannot swerve. Confusing the two makes leaders slow down execution instead of stabilizing direction.
Pretending coordination cost does not exist
It is emergent and quadratic even with excellent, collaborative people. Ignoring it means the org silently spends most of its energy on coordination while leadership blames individuals for slowness.
Is it for you?
Best for
Founders and execs scaling past the point where they can personally steer, deciding between a coherence architecture and an autonomy architecture.
Not ideal for
Teams under roughly a dozen people — the coordination cost is not yet the binding constraint, and the vehicle metaphor adds nothing.
From the transcript
“the core Dynamic that makes organizations hard to navigate as they get larger even if you assume everybody's actively good at what they do actively…”
“think of a company like a vehicle when the company is very small you can drive it you can steer it like a like a…”
“you grow into the size of a big rig if you drive a big rig like a sports car you're going to be a danger…”
“people go oh it means go slow no no no it means pivot less”
“Apple has chosen the former is it very important to them to have the illusion of perfect coherence their products it worked very well for…”
From the episode
Thinking like a gardener not a builder, organizing teams like slime mold, the adjacent possible, and other unconventional product advice
Alex Komoroske (Stripe, Google)