Market research for a startup is not a survey of a thousand people or a report bought from an analyst firm. It is a short, focused effort to answer the handful of questions that decide whether an idea can work, using sources a founder can reach in days and check for themselves. This guide sets out those questions, where to find each answer, and the order to do it in.
What market research means for a startup
Market research is gathering evidence about customers, competitors and demand so you can make a decision with less guesswork. A large company does it to size a known market or defend its share. A startup does it for a different reason: to find out whether a specific customer has a problem worth paying to solve, who already sells to them, and where there is room for a new product.
That difference changes the method. You need fewer, more specific answers, and you need them quickly, because the cheapest time to change an idea is before anything is built. Most of the work is desk research (reading what already exists) followed by a small amount of primary research (talking to the people you want to sell to).
The five questions startup market research has to answer
- Who is the customer, and what problem do they have? Specific enough to find them: “owners of independent restaurants with one location”, not “the hospitality industry”.
- How many of them are there, and what would they pay? A rough, bottom-up count multiplied by a realistic price is enough at this stage.
- Who already serves them? Direct competitors, indirect ones and the spreadsheet, agency or habit customers rely on today.
- What do those options leave out? The features, customers, price points or promises nobody covers well.
- How would you reach them? Whether buyers search for the problem, where they gather, and what reaching each one is likely to cost.
If your research answers these five, it has done its job. If it produces a long document that answers none of them clearly, it has not, however thorough it looks.
Desk research: what you can learn without talking to anyone
Start here, because it is fast, free or cheap, and it makes your customer conversations far better. Each source answers a different question:
| Source | What it tells you |
|---|---|
| Competitors' websites | Who they target, what they offer and how they position themselves |
| Pricing pages | What customers already pay, and in what unit |
| Reviews (G2, Capterra, app stores) | What customers like and, more usefully, what they complain about |
| Keyword data | Whether people search for the problem, and how competitive those searches are |
| Forums and communities | How customers describe the problem in their own words |
| Public statistics (Census, BLS) | How many potential customers exist, for a bottom-up market size |
| Funding and job posts | Which competitors are growing, and where they are investing |
The competitor half of this is covered in detail in competitor analysis for startups, including a worked example. For pricing specifically, see how to find competitor pricing; for demand, see how to estimate market demand with search volume and how to estimate market size for a new product.
Primary research: talking to customers
Desk research tells you what exists. Only customers can tell you whether the problem hurts enough to pay for a fix. Two kinds of primary research matter most at this stage:
- Problem interviews. Short conversations with people in your target customer group about how they handle the problem today, what it costs them and what they have already tried. Ask about past behavior, not hypothetical futures: “what did you do last time?” is more reliable than “would you use this?”.
- A commitment test. A pre-order, a deposit, a letter of intent or a paid pilot. Compliments are cheap; a commitment is evidence.
Surveys are useful later, to measure how common something is once you know what to ask. Early on, they mostly confirm what the person writing the questions already believed.
The order to do it in
A sequence that works for most founders, and fits in one to two weeks:
- Write the idea as one specific sentence: the customer, the problem and the outcome.
- Map the competitors and collect the same facts for each, with sources and dates.
- Collect their pricing and convert it to one comparable unit.
- Read their reviews and group the complaints into themes.
- Check search demand for the problem, not only for your product category.
- Estimate the market bottom-up: customers you can reach multiplied by what they would pay.
- Talk to ten or more potential customers, using what you learned to ask better questions.
- Decide: go, change the customer or the offer, or stop.
The final step is the point of all the others. The guide to validating a startup idea covers how to make that decision, and the business idea validation checklist gives each part a pass condition.
Common startup market research mistakes
- Top-down market sizing. “1% of a $50 billion market” tells an investor you have not counted your customers. Count them bottom-up.
- Only researching the famous competitors. The small product aimed at exactly your customer is often the one that matters.
- Leading questions. Asking people whether they like your idea measures politeness, not demand.
- Undated facts. Prices and features change; a figure without a source and date cannot be trusted or updated.
- Researching instead of deciding. Research is finished when it changes or confirms a decision, not when there is nothing left to read.
Automate the desk research with SignalDart
SignalDart is an AI-powered startup market research and competitor intelligence platform. From a one-sentence description of your idea, it does the desk-research half of this guide: it finds and ranks your competitors, reads their pricing and feature pages, measures search demand, scores the market gaps and returns an Opportunity Score with a plain verdict, citing the page behind every figure. You can export the result as a market research report. The customer conversations are still yours to have, and they go better when you already know the market.