A SWOT analysis is one of those tools everyone has heard of, and most people who run one end up with a page of four boxes full of general statements. Strengths: a motivated team. Opportunities: the market is growing. That page helps nobody, because no decision changed as a result of it.
The problem is not the tool. The problem is that the four boxes get filled with opinion, and when it is your own project your opinion about yourself is not neutral. So here is a stricter version: every box has an entry test, and anything that fails the test does not get written down.
One rule makes it useful: every line must produce a decision
The test that matters: after you write a line, ask what it makes you do next week. If there is no answer, delete the line.
A motivated team produces no decision. We are the only supplier in the area that delivers within 24 hours produces one, because that is something you can put in a message and something you can go and check against a competitor.
The difference between those two is not eloquence. It is that the second one is specific and can be proved wrong. Any statement nobody could contradict is not information.
Strengths and weaknesses: what you have, not what you feel
These two boxes are about things inside the project, which means things you control. The right source for them is not your own sense of it. The right source is customers who paid.
Try this. Take the last ten customers who paid, and look at what each one said when you asked why they chose you. If you never asked, that is a five minute call each, and it is the cheapest market research available. Whatever repeats in those answers is a real strength. Whatever you wrote down and cannot find in their words is your opinion, not your strength.
Weaknesses work the same way in reverse. Ask the customer who left, or the one who enquired and never came back. One practical point here: write down the things you actually lose business over, not the things you dislike about your own setup. The site looks dated is not a weakness if nobody left because of it.
Knowing who the customer is comes before all of this, and I covered it in how to identify your target market.
Opportunities and threats are outside you, so do not guess them
These two are about things outside the project, and this is where the biggest mistake happens. People write an opportunity based on something they heard or felt. There is demand for this service, based on what exactly?
There is a free tool that helps with this box: Google Trends. You can see interest in a search term over years, compare two terms, and see the difference between one region and another.
But you have to understand the number before you build on it. Google states in the Trends help documentation that the data is normalized, meaning these are not absolute search counts. Each point is divided by total searches for that place and time, and the result is scaled from 0 to 100.
Google also says that low volume terms show as zero, and that for those terms the data "reflects statistical noise rather than actual search interest". So if you search your idea and see a zero, that does not mean there is no demand. It means the term is smaller than Trends will display.
And the most important sentence on the whole page: Google says Trends "is not a scientific poll and shouldn't be confused with polling data", and that the data "should always be considered as one data point among others before drawing conclusions". The tool confirms or denies a direction. It does not measure the size of a market.
The other cheap ways to test demand before building anything are in how to know there is demand for your product before starting.
Threats: write only what could happen within a year
The threats box goes wrong when it becomes a list of worries. Competition. The economy. Those are not threats, they are the weather.
A useful threat is specific and near. Our two largest customers are 60% of revenue, and if one leaves we are in trouble. That produces a decision immediately: start looking for new customers tomorrow.
A practical rule: if a threat could not happen within twelve months, remove it. Not because it is unreal, but because you cannot act on it now.
If one of your threats is a competitor undercutting you, that is its own situation and I wrote about it in how to handle a competitor who is cheaper than you.
Turning it into a decision: the boxes are read in pairs
This is the step that makes the difference, and most people skip it. You do not read the analysis one box at a time. You read it in pairs.
Strength with opportunity: this tells you where to spend. You have a real strength and there is demand for it, so that is the focus for the next quarter.
Weakness with opportunity: this tells you what to fix. There is demand and you are not ready, so either fix the weakness quickly or let the opportunity go deliberately rather than by forgetting about it.
Strength with threat: this tells you how to defend. Your strength is what you lean on when the threat arrives.
Weakness with threat: this is the red list. Anything here needs a decision with a date on it, not monitoring.
If you cannot find a single pair, what you produced was descriptive rather than analytical, and that is the moment to go back and delete the general statements.
One page, with a date on it
A useful analysis looks like this: one page, three to five specific lines per box, and under each pair one decision written as a verb with a date.
Put a date on the page. That is not cosmetic. The analysis stays valid until something significant changes, and the date is what lets you reread it in three months and see whether you were right. That is what turns the page from an exercise into something you learn from.
The disciplined version of this, meaning how to filter accounts and decide who to approach and who to leave alone, is written up more strictly by BDG Labs in an ideal customer profile for outbound: four hard tests. It is written for B2B companies, but the tests themselves work as a ruler for your strengths box.
The honest limits
A SWOT analysis does not tell you what to do. It organises what you already know, and the decision comes out of the pairs above rather than out of the boxes. So anyone expecting the page to tell them where to start will be disappointed.
Second, the tool is far more useful for a running business than for one still in your head. If you have not started, you have no real strengths or weaknesses to write, and everything you put down is a forecast. In that case testing demand matters much more than analysing.
Last and most important: this gets done once and then forgotten. It takes time, nobody returns to it, and it becomes a page describing a market that has moved. If you are not going to open it again in three months, the time you spent was wasted. Set the review date while you are writing it, not afterwards.
Every marketing goal needs a business goal underneath it; if there is none, nobody knows what the activity is for. That idea anchors Part 2 of my free book Growth Engine, along with a way of writing a goal so that anyone reading it knows when you have succeeded: Part 2, Goals: where are you going (in Arabic).