Market Study Analysis: A Lean Guide for Startup Founders

A market study analysis is a structured check of whether enough people want what you plan to build, who else serves them, what is changing around them, and how big the opportunity is. For a startup, it is the cheapest way to find out that an idea is weak before you spend money and months on it.

You do not need a consulting budget. A founder can run a useful market study analysis in about five days, with an AI-drafted starting point, a handful of user conversations and a spreadsheet. This guide walks through the steps with one worked example, then shows how to turn the findings into a product plan.

Most guides stop at a report. This one ends with a decision (go, pivot or stop) and a plan you can build from.

What a market study analysis is (and what it is not)

A market study analysis answers five questions about a specific market: how big it is, how much demand exists, who the competitors are, what trends and rules affect it, and which conditions would help or hurt a new entrant.

The terms get mixed up, so here is how they differ in practice.

Market research feeds the market study, and the study feeds the plan. If you want more depth on the research methods themselves, see our 7-step market research guide.

A corporate report might run 80 pages and take a quarter. A startup needs something smaller: enough evidence to decide whether to keep going. Aim for a short document and a clear verdict.

Before you start: define the market and the question

Studies drift when they have no purpose. Settle three things first.

• The decision. Write the one decision the study must support. Example: “Should I build an MVP of a booking app for independent physiotherapy clinics?”
• The market, in one sentence. Who, which problem, which context. Example: “Solo and two-person physiotherapy clinics in the US that take bookings by phone and lose appointments to no-shows.”
• The time box and the bar. Five working days, and a pass/fail line written down now, before you see the data.

The bar is the part people skip. Without it, every result looks like a reason to continue. A sample bar: at least 6 of 10 interviewees describe the problem unprompted, at least 3 say they would try a pilot, and the reachable segment supports your revenue goal.

Step 1: Start from an AI research draft

On day one, ask an AI tool for a first draft. It is fast and it gives you a structure to test. A useful prompt names the market sentence, then asks for: likely segments, the size drivers, known competitors and their pricing, trends and regulation, and the questions a skeptic would ask.

You can also start from an AI business plan generator, which gives you the same kind of first pass inside a plan structure.

Verify the draft with this checklist:

• Does every number have a source you can open? If not, delete it or mark it unverified.
• What is the date of each source? Drop anything old enough to be out of step with the market.
• Do all the named competitors exist? Visit each website. Invented or defunct companies are common.
• Are the prices and features current? Check the live pricing page.
• Does the market definition match yours, or is it a much bigger category?
• What did the draft leave out? Ask what a local or niche player would add.

Step 2: Demand analysis

Demand analysis asks who needs this, how many of them, and how often the problem comes up. Start with signals you can collect without talking to anyone.

• Search behavior. Free keyword tools show whether people search for the problem or for existing solutions.
• Forums and communities. Look for threads where people complain about the problem or ask for tools.
• Reviews of existing products. Three-star reviews of competitors list unmet needs in the customers’ own words.
• Job posts. If companies hire people to do the task manually, the problem has a budget.

Then talk to people. Interview 5 to 10 target users, 20 to 30 minutes each. Ask about the last time the problem happened, what they did about it, what it cost them and what they tried. Do not pitch. People are polite about ideas and honest about past behavior.

Add a short survey if you can reach more people, mostly with open questions. We list 25 in open survey questions for idea validation.

For an early idea, interviews beat surveys. A survey needs a good question list and enough respondents to mean anything. Interviews tell you what to ask.

Step 3: Competitor and offer analysis

List three kinds of alternatives: direct competitors, indirect ones (a different product solving the same problem) and the do-nothing option, which is often a spreadsheet, a phone call or a sticky note. The do-nothing option is usually your biggest rival.

For each, record pricing, who they target, how they position themselves and where their users complain. Our guide on how to map your rivals and find your gap covers the comparison in detail.

You are looking for a gap you can own: a segment they ignore, a job they do badly, a price point they miss. A gap that only you can see is a warning sign. A gap that users mention in reviews is an opening.

Step 4: Environmental analysis

PESTEL covers political, economic, social, technological, environmental and legal factors. A full PESTEL is overkill for a first study. Keep only the factors that could change your plan.

• Regulation. A clinic app may touch health data rules. A pet-walking marketplace may face licensing or insurance rules in some cities.
• Technology shifts. A new platform feature or API could make your idea easier, or obsolete.
• Buying cycles. If clinics set budgets once a year, a launch in the wrong month costs you a season.
• Economic pressure. Are your target customers cutting or adding spending on this category?

Write one line per factor: what it is, and whether it helps, hurts or does nothing. Skip any factor with no effect on the plan.

Step 5: Segment and size the market

Pick a first segment before you size anything. “Everyone with a pet” is not a segment. “Full-time workers in mid-size US cities who own a dog and are away 9 hours a day” is. For the method, see how to pick a first segment.

Then size it bottom-up: count the customers you can reach and multiply by what they would pay. Top-down numbers (“the pet industry is worth billions”) tell you nothing about your own revenue.

Interpret the numbers with the decision in mind. A big TAM with a tiny SOM is still a tiny business for now. What matters for a startup is whether the SOM supports your goals and whether you can name the channel that gets you the first customers.

Worked example: a lean study from start to finish

Fictional idea: PawRoute, a marketplace that connects dog owners with vetted walkers in mid-size US cities. The numbers below are illustrative, not real market data.

Decision: build an MVP in one city, or drop the idea. Time box: five days. Bar: at least 6 of 10 owners report a recurring pain, a gap exists against current apps, and a single-city SOM supports at least $60,000 in yearly revenue.

Verdict: go, with conditions. Demand passed the bar (8 of 10). The gap is reliability with a backup walker. The SOM clears $60,000 but not by much, so the first test is whether we can get 50 households and 15 walkers in 60 days. If that fails, we pivot to a narrower niche (such as senior dog care) or stop.

Turn the study into a product plan

A study that ends in a PDF gets forgotten. Every finding should land in a section of your plan, which is the thread of turn a raw idea into a product plan.

This also keeps scope honest. If a feature does not tie back to a finding, it waits. See how to define MVP scope and map user journeys before you code.

TwoSprint is built for this step. You bring the study and the idea, it turns them into a structured plan, and the plan carries through to a design system, tasks and a build, so the findings do not get lost between research and code.

Common mistakes

• Confusing market size with reachable demand. A large TAM says little about the customers you can actually get.
• Only desk research. Reports cannot tell you why someone cancelled a walker at 7 a.m. Talk to people.
• Trusting AI numbers without sources. If you cannot open the source, do not use the number.
• Studying forever. Set the time box and the bar, then decide. More data rarely changes a clear result.

Lean market study checklist

• Decision written down in one sentence
• Market defined in one sentence: who, problem, context
• Time box set (5 days) and pass/fail bar written before research
• AI draft created and every claim verified or removed
• Signals gathered: search, forums, reviews, job posts
• 5 to 10 interviews done, notes saved with quotes
• Direct, indirect and do-nothing alternatives mapped
• Gap identified and backed by user comments
• Only plan-changing PESTEL factors listed
• First segment chosen
• Bottom-up TAM, SAM and SOM calculated with assumptions shown
• Verdict recorded: go, pivot or stop, against the bar
• Findings mapped to target user, features, MVP scope, journeys and tests

Doing this with TwoSprint

Once the study gives you a verdict, the slow part is turning it into something buildable. TwoSprint takes your idea and findings and helps you shape the target user, features, MVP scope and journeys into one plan, then carries it into design and tasks. If you want to try it, start with the AI business plan generator.