Every time I tell someone to do market research before they start, the first answer is the same: that needs a research firm and money I do not have. That is wrong, and not only because a free alternative exists. For a small business the free alternative is genuinely better.
A research firm hands you a report, and a report is something you read. When you do the research yourself you hear the customer with your own ears, and that changes your decisions in a way no document does.
So here is the method I actually use. Four sources, three of them completely free, and one page at the end.
First: market research answers three questions, not twenty
Most people who try this drown because they do not know what they are looking for. Start with the questions, and there are only three.
One: do people actually have this problem, and roughly how many? That is a size question. Two: how are they solving it today? That is a competition question, and the competitor is often not a company. It is a spreadsheet, an employee, or nothing at all. Three: are they already paying somebody for this? That is a money question.
Any fact that does not serve one of those three, however interesting, gets left out. That is the difference between research that takes a week and research that never ends.
Second: the official numbers are free and you are not using them
This is the cheapest step and the most skipped. National statistics agencies publish detailed data for nothing, and most people would rather guess.
I opened the Saudi General Authority for Statistics site today while writing this. The headline indicators published on its front page were a population of 35,300,280 from the 2024 estimates, Saudi unemployment at 6.5 per cent for the second quarter of 2026, and inflation at 1.7 per cent for October 2026.
That was the front page, with no sign-up and no request. More useful still, the site carries a statistical database and an open data portal, so you are not limited to headline indicators. You can go down to region and sector.
There is a second thing worth noticing on that page: a schedule of upcoming releases with dates. When I read it there were specific releases scheduled for the following week, including agricultural statistics, short-term business statistics, a construction cost index and international trade. If your sector has a release coming, you can wait for it instead of buying a number.
Egypt's CAPMAS works the same way, and its income, expenditure and consumption survey is directly useful to anyone trying to understand purchasing power. Start with the official numbers before any paid tool.
Third: Google Trends, and what it does not tell you
Trends is free and useful, and most people read it wrong. So here is what Google itself documents about it.
On its FAQ about Google Trends data, Google states that the tool uses a sample of searches rather than all of them, that each data point is divided by the total searches of the geography and time range it represents in order to compare relative popularity, and that results are scaled on a range of 0 to 100 based on a topic's proportion to all searches on all topics.
Notice the word relative. A score of 100 does not mean large volume. It means the highest point in the range you selected. The same page says Trends only shows data for popular terms, and that search terms with low volume appear as zero.
And the most important sentence on the page: Google says Trends is not a scientific poll, that it should always be considered as one data point among others, and that it is not a perfect mirror of search activity. It adds that small random fluctuations, statistical noise, are most noticeable on low-interest queries.
So if your subject is specialised, what you see in Trends is probably noise rather than signal. Use it to compare two terms, or to find seasonality. Do not use it to estimate the size of a market, which is exactly what it gets used for.
Fourth: ten conversations beat a hundred survey responses
This is the step nobody wants to do and the one that matters most. Ten conversations with people from your audience, twenty minutes each, will teach you more than any report.
It has one rule: never ask about the future. A question like would you buy something like this gets a yes almost every time and is worth nothing. Ask about what already happened. The last time you had this problem, what did you do? Did you pay anybody? How much? What annoyed you about the solution you used?
The past is reported more honestly than the future, and that is the basis of any useful customer conversation. I wrote about the questions that surface the real problem in how to find the client's real problem.
And if you are still deciding who you are talking to in the first place, that is the question before all of this, and it is in know your customers: when you try to please everyone you usually reach nobody.
Competitors: read what they do, not what they say
Competitor analysis is not visiting their websites and reading the about page. That is the least useful thing they publish.
What helps is elsewhere: the comments on their posts, especially repeated complaints; their Google Maps reviews; what they put in their ads, because they are paying for it so it is probably working; and what they quietly removed from their services after a year.
That last one is the best free signal available. A service that got removed was not making money, and somebody else paid for that lesson.
Once you know where you sit among all that, what actually makes your offer different rather than a slogan is in what makes a brand different.
The result fits on one page
Research that ends in a folder is not research. Research that ends in one page with five things on it becomes a decision:
One: who the customer is, in a sentence. Two: the problem in their words, not yours. Three: how they solve it now, and for how much. Four: the single official number supporting that size exists. Five: the question you still cannot answer.
The fifth item is the most honest line on the page, and the first one removed when somebody is selling the plan to themselves. Research whose output is entirely confirmations was not research, it was justification.
The full sequence for reviewing where you stand before writing a marketing plan is in how to review your position before a marketing plan.
Honest limits
This method does not produce a market size figure. If you need a number to put in a document for an investor, the three sources above will not get you there. They get you something else, which is understanding, and for a small business owner that changes decisions more than a number does.
The figures I quoted were read on 16 November 2026. Statistical indicators get updated, and inflation and unemployment change every quarter. Open the site yourself and take the current number rather than copying mine.
Ten conversations is a small and biased sample. The people who answer you are not a random sample. They are the ones with time, or the ones who know you. That is a real limit, and the way to reduce it is to go looking specifically for someone who refused to buy or bought from a competitor, because those are the most useful people to talk to and the hardest to reach.
And there is one case where this method does not work at all. If your product is genuinely new and nobody is solving the problem in any form, then questions two and three have no answers, and you are not doing market research, you are running an experiment. The difference is that its result comes from actually selling something to somebody, not from asking.
If you would rather build this as a structured skill with real measurement tools than assemble it from scattered sources, the team at UCourses have written on what an analytics course must actually cover in their analytics course for marketers. That is the paid, structured route; the above is the free one, and both work.
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).