Framework

Circle of Competence

Buffett and Munger's model for knowing exactly where your understanding ends — size doesn't matter, honesty about the edge does.

Circle of Competence Find this on the mind-map

Introduction

Charlie Munger and Warren Buffett made this famous as an investing discipline, and it's since spread widely into leadership. Your circle of competence is the set of things you actually understand — where your knowledge and experience run deep enough that your judgment holds up. It has nothing to do with how many years you've put in or how confidently you talk.

How It Works

The model works like a target with three zones:

Zone What it contains How you lead there
Center Deep expertise. Places where you've made real decisions, seen the outcomes, and sharpened your judgment over years. Trust your instincts. Move with clarity and speed.
Edge Topics you're exposed to and interested in, but haven't gone deep on and don't have a track record in. Slow down. Ask questions, explore, bring others in.
Outside Territory you don't know at all. Decide here and you're mostly guessing. Say so out loud. Defer, delegate, or deliberately learn your way in.

Circle of Competence diagram

Your circle isn't fixed. It grows through real effort — learning, hands-on experience, feedback, doing it again and again — and it can't be faked or made bigger just by saying so. Most leaders overestimate the size of their own circle. The point of mapping it isn't to make it look big. It's to make it real.

Why It Matters for Leaders

You're expected to have answers on strategy, people, metrics, markets — the list never ends. The honest response isn't to be good at everything. It's knowing exactly where you actually stand. When you bluff, the bad decision doesn't stay contained — it grows as it moves through your team and out into the business.

Overconfidence outside your circle has a name: the Dunning-Kruger effect, where people with limited knowledge in an area overestimate how competent they are. It applies to you as readily as to anyone else. Assume you know enough about an unfamiliar technology, wave off your own experts' input, and make the call anyway, and you usually end up with an expensive, inefficient mess. The fix is explicit: know what you don't know, check your knowledge gaps on a regular basis, and pair this with the Johari Window to surface your blind spots.

There's a newer version of the same trap. AI has pulled confidence and competence apart. When you produce great output with AI's help, the pride you feel is hard to tell apart from the pride of having actually done the thinking yourself — so the gap between how competent you feel and how competent you are widens without you noticing. Self-assessment has gotten less reliable at exactly the moment it matters most, both for you and for anyone reading your output.

The upside is trust. Be straight about your limits and you earn credibility with your team and your stakeholders, get better information feeding into your decisions, and open up room for other people to lead where their circle runs deeper than yours.

A concrete example helps. Say you're a first-time engineering manager. Inside your circle: writing and reviewing production code, debugging, system design, effort estimation, tech debt trade-offs. At the edge: running one-on-ones, setting team OKRs, running cross-team planning, speaking up in leadership reviews. Outside: handling team dysfunction, hiring across unfamiliar stacks, org-level politics, coaching underperformers, comp planning, engineering strategy. You can have opinions in all three zones. You only get real outcomes in the first one.

Your chat with Kyoshi about the Circle of Competence framework

What to do next

Test your knowledge of Circle of Competence

Take the quiz →

Practice Circle of Competence

Apply Circle of Competence to a real situation

Skills related to Circle of Competence

Frameworks related to Circle of Competence

Books related to Circle of Competence