The framework I use to pick a framework

The most common message I get from people using FrameworkList is some variation of: "I don't know which one to use." The library has 100+ frameworks. Browsing the catalog top-to-bottom is the wrong way to find the right one — you'll spend longer choosing than thinking.

After several years of doing this, the procedure I actually use is three questions. Two minutes. Yes, this is a framework for picking a framework, and yes, I have noticed the recursion.

Question 1: What kind of work is this?

Frameworks live in five buckets:

• Triage — sorting a pile of stuff you already have. Eisenhower, MoSCoW, Pareto.
• Prioritization — ranking options by some criteria. RICE, ICE, Weighted decision matrix, Kano.
• Decision — choosing between specific alternatives where the answer matters. Pros/Cons, Decision tree, Regret minimization, Cost-benefit, 10/10/10.
• Strategy — figuring out what to be doing over a longer horizon. SWOT, Five Forces, BCG Matrix, Blue Ocean, Wardley, JTBD.
• Diagnosis — understanding what's actually going on. 5 Whys, Fishbone, OODA, Cynefin, First Principles.

If you can name the bucket in under 30 seconds, you've eliminated 80 frameworks. If you can't name the bucket, you don't yet know what kind of problem you have — that's a useful thing to notice. Spend the next 10 minutes writing down what you're trying to do until you can.

Question 2: How reversible is the decision?

This is the question that surprises people.

• Low-stakes, easily reversible — use the simplest framework that gives you a defensible answer. Pros/Cons, ICE. Don't overthink. The cost of being wrong is small, the cost of deliberation is real.
• High-stakes, hard to reverse — use something heavier and pair it with a premortem. SWOT or Five Forces and a premortem. The cost of being wrong is huge; the cost of deliberation is rounding error against that.
• Strategically defining — use the multi-framework approach. SWOT to map, Premortem to stress-test, Decision matrix to choose, OKRs to commit. Reversibility close to zero; deliberation should match.

The mistake people make: they apply the same framework weight regardless of stakes. Either they over-deliberate routine decisions (six hours of SWOT for which CRM to buy) or under-deliberate strategic ones (a fast Pros/Cons for which market to enter).

Match the framework's weight to the decision's weight.

Question 3: Will other people need to see the answer?

A solo thinker can use any framework. The artifact is for them.

But the moment you need to share the result with stakeholders, partners, or your team, half the frameworks fall out. Some produce great solo output but bad meeting artifacts (5 Whys reads as a stream of consciousness). Some produce great meeting artifacts but require facilitation (Six Thinking Hats works in person, falls flat in Slack).

If you'll need to share, prefer frameworks that produce structured, one-page visual outputs. SWOT, RICE, Eisenhower, BCG, Lean Canvas, Decision matrix. These all collapse to a single image you can drop in a doc.

If you won't need to share, you have full freedom. Use the one that fits your own thinking style.

The procedure end to end

1. Name the bucket (triage / prioritization / decision / strategy / diagnosis).
2. Score reversibility (easily reversible / hard to reverse / strategically defining).
3. Decide on shareability (just for me / for the team).

That's three answers. The intersection of those three almost always points to one or two frameworks. Open the library, look at those, pick the one whose example matches your situation closest, run it.

Examples I run a lot

I'll save you the trouble of reading the matrix. These are the frameworks I personally reach for most:

• "Should I do X this week?" — Eisenhower Matrix. 60 seconds.
• "Which of these 15 features ship next quarter?" — RICE. 90 minutes with the team.
• "We're about to make a big bet. What could go wrong?" — Premortem. 45 minutes.
• "Should we enter this market?" — SWOT + Five Forces in sequence. Two sessions, a week apart.
• "I cannot tell what is wrong with this team." — 5 Whys, then Fishbone if 5 Whys runs out of road.

If you're new to the library, those five cover ~70% of the decisions you'll face. The other 95 frameworks become useful as you get more specific.

When none of this helps

Sometimes the answer to "which framework should I use" is none of them. The problem isn't underspecified or stuck — it's just hard, and a framework won't make it easier. In those cases, the right move is to talk to someone who has solved the same problem before. Frameworks are a way to think clearly when you're alone with the problem. A 30-minute call with someone who's been there is faster and better when it's available.

If you'd like to compare notes, the welcome email is from a real address. Reply to it.

— King Mark