Map internal factors (Strengths, Weaknesses) and external factors (Opportunities, Threats), then convert the grid into a short list of strategic moves.
Albert Humphrey (popular attribution from 1960s–70s planning practice); later formalised with TOWS by Heinz Weihrich
SWOT is a simple lens to structure situational analysis. It separates what’s inside your control (capabilities, assets, constraints) from what’s outside (market shifts, regulation, tech, competition). The real value comes when you turn the four lists into actionable strategies (the TOWS step), not from generating a long catalogue of observations.
Two axes
SWOT synthesis – combine quadrants to create options:
Company and BU strategy;
Market entry;
Product/portfolio review;
Competitor and partner assessment;
Personal/career planning.
Define the objective & scope – what outcome and horizon is this SWOT for? (e.g., “UK SMB launch in 12 months”).
Populate each quadrant (evidence-first)
Strengths – proof in metrics/assets: unique data, low CAC, patents, distribution, culture.
Weaknesses – bottlenecks/constraints: capital, brand, churn, unit economics, missing capabilities.
Opportunities – external tailwinds: regulatory change, underserved segments, adjacent jobs, tech costs falling.
Threats – external headwinds: new entrants, platform risk, substitutes, compliance, macro shifts.
Prioritise – score each item by impact × certainty; keep the top 3–5 in each box.
Do TOWS pairing – write 1–2 concrete options per pairing (S–O, W–O, S–T, W–T).
Choose & test – pick 2–3 options; add owners, metrics, and a 90-day test plan.
Review cadence – refresh quarterly or when a major assumption changes.
Laundry lists – too many vague items; force evidence and cut to the vital few.
Mixing internal/external – keep “we can change” separate from “we must respond to”.
No synthesis – a SWOT without TOWS leaves you with description, not decisions.
Wishful thinking – treat items as hypotheses with sources; avoid platitudes (“great team”) without proof.
Static snapshot – trends move; set explicit review triggers.
One size fits all – run SWOT for a specific objective/market, not the whole world.
Click below to learn other mental models
Order matters → permutations. Order doesn’t → combinations. Adjust for with/without replacement.
In blockchains, value tends to concentrate at the shared protocol layer rather than the application layer, though modular stacks and wallets can shift where value accrues.
A visual of sequential decisions with probabilities and payoffs; fold back to compute expected value.