A SWOT analysis looks simple: four boxes, four headings. The hard part is filling them well. The version most people build is decoration. It is a slide full of generic phrases that nobody disputes and nobody acts on. The version below is the method we use with clients, and it produces a matrix that feeds directly into strategy instead of sitting in a deck untouched.
There is a useful way to think about the difference. A decorative SWOT lists things like "strong brand" and "growing market" because they sound safe. A working SWOT lists facts that are specific enough to disagree with, and each fact ends up paired with a decision. The whole craft of this exercise is in the pairing step, not in the box drawing. If you only take one thing from this guide, take that: the boxes are a means to an end, and the end is a set of concrete strategic moves.
Step 1: Agree on the scope before anything else
The biggest mistake is starting without a shared scope. SWOT for the whole company, one business unit, one product line, or one market entry are different exercises, and a matrix built for one of them will mislead you if you use it for another. Decide up front:
- What entity are we analyzing? Name the exact business unit, product, or geography. "The company" is usually too vague.
- What time horizon: current state, or the next 12 months? A snapshot of today and a view of next year produce different items.
- What decision is this SWOT feeding? A go-to-market plan, an investment case, a turnaround, or a hiring plan each needs different facts.
Write the scope at the top of the page in one sentence. It stops every subsequent argument. When two people disagree about whether something is a strength, the scope usually settles it: the fact is a strength relative to this entity, this horizon, and this decision, or it is not in scope at all.
A useful benchmark: an experienced consultant spends the first 10 to 15 minutes of a workshop on this step alone. If your team is arguing about scope after 30 minutes, your analyst has not finished step two yet.
Step 2: List the facts, not the opinions
Before sorting anything into boxes, collect the facts about the business. This is where most teams shortcut, and it is the reason their SWOT reads like a PR page. Gather, in one place:
- Revenue and cost structure, by product line and by customer segment
- Customer concentration and churn rates, including your three largest customers as a share of revenue
- Competitive moves and market conditions, meaning what changed in the last 12 months
- Regulatory changes and constraints that affect pricing or distribution
- Internal capabilities and bottlenecks, from headcount to production lead time
- Any data you already hold: sales pipeline, support tickets, renewal rates, unit economics
The rule: every item is something you could point at with a number or a source. "Strong culture" without "retention is 30% above industry average" is an opinion, not a fact. If you cannot attach a number or a named source to an item, either find one or drop the item.
This step also has a discipline about volume. A good working list has 30 to 60 raw facts. A list shorter than that means you have not looked hard enough. A list longer than that means you are including noise. Cut anything that does not change a decision, and you will naturally land in the working range.
Step 3: Sort into exactly one box
Use the internal/external test:
- Internal (true today, about the company) goes to Strength or Weakness
- External (about the market around it) goes to Opportunity or Threat
One fact, one home. If the same thing appears in two boxes, you are listing opinions, not building analysis. Internal versus external is the line people cross most: a weakness that is actually an industry condition ("rising input costs" for everyone) belongs in Threats, not Weaknesses. A strength that depends on a competitor disappearing is not a strength yet.
Another common sorting error is time. An opportunity is an external condition you can act on within your planning horizon. A condition that only matters in year five is a trend worth noting, but it does not belong in the matrix if it does not change this plan.
A simple test for each item: would a reasonable person with the same data place it differently? If the answer is yes, you have an opinion disguised as a fact, and it goes back to step two until it is anchored to a number.
Step 4: Ask the two questions that make it useful
The matrix earns its keep when it answers:
- Can we win? Strengths versus Opportunities: where do our assets meet an opening? This is the upside question, and it should produce at least one specific move.
- Can we lose? Weaknesses versus Threats: what combination breaks us, and what is the cheapest hedge? This is the downside question, and it should produce at least one specific protection.
Anything that answers neither question gets cut. This is the filter that separates a working SWOT from decoration. Every surviving item should be able to complete one of two sentences: "This lets us do X" or "This stops X, so we need Y".
Keep the counts honest. A matrix where every cell has seven items is a sign of weak prioritization. A strong matrix often has two or three items per cell, each one load bearing. If your strongest cell has more than five items, rank them and cut the tail; the client cannot act on twenty initiatives.
Step 5: Write the linkage sentences
For each cell that survives, write what it means for strategy. The format is simple: "S2 + O1: our brand lets us enter the mid-market before competitors", or "W3 + T2: our reliance on two suppliers and their capacity squeeze means we fund a second source this quarter". Now the matrix hands you the strategy section of your deck instead of sitting apart from it.
The linkage sentence is the deliverable. Everything before it is plumbing. If you ship a SWOT without linkage sentences, you have shipped a list. If you ship the linkage sentences, you have shipped a plan with the evidence attached.
Common mistakes to avoid
- Box filling for its own sake: a full matrix with no decisions. Every item must connect to a move.
- Generic language: "strong team", "good culture", "competitive pricing". Each is a placeholder for a fact you have not written down yet.
- Mixing internal and external: putting industry conditions in Weaknesses or competitor behavior in Opportunities. Use the internal/external test on every item.
- No prioritization: twenty strengths read as zero strengths. Rank the cells and cut the tail.
- Static one-off use: a SWOT is only useful at the moment you make a decision from it. Rebuild it when scope, time horizon, or the decision changes, which usually means twice a year or before each major move.
- Skipping the downside: teams love the Opportunities box and avoid the Threats box. The Threats box is where you find the cheapest insurance, so do not under-fill it.
How to present the results
Two formats work in practice. For a written report, use a two by two matrix followed by a page of linkage sentences grouped by decision. For a workshop, put the four boxes on a wall, agree the facts, then physically move the strongest items into a decisions column as you write the linkage.
Whichever format you use, end with a numbered list of actions, each tied to a specific item. If an action cannot be tied to an item, it does not belong on the same page as the SWOT.
Speed it up
The box filling is mechanical. You can generate a structured SWOT in a minute and spend your time on the linkage, which is where the value lives: free SWOT generator. Paste in your business facts, review the matrix, and write the linkage sentences from there.