── ── Mental model
Dunbar's Number
Dunbar's number (~150) is the cognitive limit on stable social groups the human neocortex can sustain, with nested layers at ~5, ~15, ~50, then outer bands at ~500 and ~1,500. Derived by Robin Dunbar (1992) from primate neocortex ratios. Key implication: below ~150, informal trust coordinates behavior; above ~150, formal hierarchy becomes architecturally mandatory — not a cultural choice.
Run Dunbar's Number on a real problem
Bring something you're actually deciding — free, in the browser.
How it works
Step 1 — Identify layer: headcount → Dunbar layer (5/15/50/150/500/1,500) → distance to next breakpoint.
Step 2 — Audit coordination: how does the group coordinate? (informal trust / formal process / hierarchy) — where are breakdowns?
Step 3 — Diagnose mismatch: does coordination mechanism match group size? (below 50: direct knowledge ok; 50–150: structured trust ok; above 150: formal hierarchy required)
When to use it
- team headcount is approaching or crossing 150 (or 50, or 500)
- someone says "coordination is getting harder," "we lost the startup feel," or "we don't know who does what anymore"
- deciding whether to split a team, open a new office, or add a management layer
When not to use it
group is well below 50 and problems are clearly interpersonal; core issue is incentive misalignment rather than group size (use principal-agent instead).
Worked example
Scaling an AI-Native Startup Team (2023–2026)
Between 2023 and 2026, a distinctive class of company emerged: the AI-native startup. Powered by generative-AI product demand and an unusually deep venture-funding environment for AI, these companies raised large rounds early and hired far faster than the classic SaaS growth curve. The pattern that recurred — reported across founder essays, investor commentary, and press coverage of companies such as OpenAI, Anthropic, and a wave of well-funded application-layer startups — is that capital compresses the timeline to each Dunbar breakpoint. A team that historically took…
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skillsUseful? Star the repo — stars help other builders find it.
Related mental models
A meaningful empirical claim must specify what observations would refute it.
The Feynman Technique tests whether understanding is genuine (can reproduce, predict, extend) or surface (can recognize, recall jargon).
In a small company the founder is the critical infrastructure: when their energy, health, or judgment fails, the company fails.
Goodhart's Law: when a metric controls behavior, people optimize the metric rather than the underlying goal.