Market Segmentation: How Startups Pick a First Segment

Market segmentation is the practice of dividing a market into groups of people or companies that share characteristics, needs or behaviors, so you can serve one group well instead of everyone badly. The four classic types are geographic, demographic, psychographic and behavioral. B2B teams add firmographic.

Most guides assume you already have customers, a survey budget and a database to slice. If you are pre-launch, you have none of that. This article shows how to segment with no customer data, pick a single beachhead segment, and use that choice to set your MVP scope and find your first users.

Short on time? Jump to the worked example or the checklist at the end.

What market segmentation is (and what it is not)

Market segmentation means splitting a broad market into smaller groups whose members have something meaningful in common, such as the problem they have, how they buy, or the size of their business. The point is to make decisions easier: who to build for, what to say, where to show up. For a deeper companion read, see a founder’s guide to market segmentation with an example.

Segmentation is often confused with the two steps that follow it. They are a sequence:

Market segmentation vs customer segmentation

The two terms overlap, but they start from different places.

• Market segmentation looks at the whole potential market, including people who have never heard of you. It is what you do before launch.
• Customer segmentation groups the people who already buy from you, using real purchase and usage data. It is what you do after launch, to retain and upsell.

If you have no customers, you are doing market segmentation, and you will be working from research and conversations rather than analytics.

The 4 types of market segmentation (plus the 5th you will actually use)

Here are the four standard types, each with a one-line example.

• Geographic: grouping by location. A pet-walking marketplace launches in one city, not a whole country.
• Demographic: grouping by age, income, job, family status. A budgeting app for recent graduates.
• Psychographic: grouping by values, attitudes and lifestyle. A meal-kit service for people who care about cooking from scratch.
• Behavioral: grouping by how people act: how often they buy, what triggers them, how loyal they are. A fitness app aimed at people who quit gyms within three months.

The question “what are the 5 methods of market segmentation?” usually gets one of these additions:

• Firmographic: the B2B version of demographics. Company size, industry, revenue, number of locations.
• Needs-based (problem-based): grouping by the job people are trying to get done. This is the one most useful for startups, because it points straight at what to build.
• Technographic: grouping by the tools people already use, such as companies on a particular accounting platform.

Different sources count five differently. The usual lists are the four above plus firmographic, or plus needs-based.

A quick example: Coca-Cola

Coca-Cola is the textbook case. It segments by demographics and lifestyle (Diet Coke and Coke Zero Sugar for people cutting sugar), by geography (local flavors and pack sizes in different countries), and by behavior and occasion (small cans for individuals, large bottles for families and gatherings). It works because the company already has years of sales data. You do not, so your version needs to be smaller and cheaper to test.

Why segmentation matters more before you launch

A large company can afford to be wrong about a segment for a year. A small team cannot. With limited time and money, you can serve one audience well or several audiences poorly.

Your segment choice quietly decides most of what comes next:

• Messaging: the words you use on the landing page depend on who is reading.
• Features: a solo practitioner and a hospital network need very different products.
• Channels: one segment hangs out in a Facebook group, another on LinkedIn, another nowhere online.

Segmentation is also the step that turns a vague audience into something usable. If you are still at “anyone who needs this,” start with defining your target market, then come back here to compare the candidates.

What makes a good segment

Whatever type you use, test each segment against five criteria:

• Measurable: you can roughly size it and tell who is in it.
• Reachable: you know where to find these people and can contact them.
• Substantial: big enough to build a business on, even if small at first.
• Distinct: they behave or need something differently from other segments.
• Actionable: you can act on the difference with a specific feature, message or channel.

How to segment a market with no customer data

No customers does not mean no evidence. It means you gather evidence from outside your own product. Here is a five-step process.

Step 1: Write the problem and the job

Start with one sentence: “People who ___ struggle to ___, so they currently ___.” Then write the job they hire a solution for. For a clinic booking app: “Keep my calendar full without spending evenings on messages.” Segments are easier to find when you know the job.

Step 2: List 4 to 6 candidate segments

Use problem-based cuts first, then firmographic or behavioral ones. Avoid demographics-only lists like “women 25 to 45.” Good candidates are specific: “solo physiotherapists,” “yoga studios with two or more instructors,” “HR teams at 200-person companies.”

Step 3: Gather evidence

You can do this in a week or two without a budget:

• Talk to 5 to 10 people from each of your top candidates.
• Read forums, subreddits and Facebook groups where they complain.
• Read one- and two-star reviews of competing tools.
• Check public data such as industry association counts or business registries for rough size.

The 7-step market research process covers these methods in more detail.

Step 4: Score each segment

Rate each candidate from 1 to 5 on five factors:

• Pain: how badly does this problem hurt them today?
• Ability to pay: do they already spend money on this or similar problems?
• Reachability: can you find 20 of them in two weeks?
• Size: is it large enough to matter?
• Competitor weakness: how poorly do existing tools serve them?

Do not weight everything equally if you have a reason not to. Early on, pain and reachability usually matter most, because they determine whether you get a first user at all.

Step 5: Pick one beachhead segment and write down why

A beachhead segment is the single group you aim to win first. Write one paragraph that says who they are, why you chose them, and what evidence would make you change your mind. That last part keeps the decision honest.

Worked example: a scheduling app for physiotherapists

Say you want to build a scheduling app for physical therapy practices. Your research gives you four candidates: independent physiotherapists, small clinics (3 to 10 staff), gyms with in-house physios, and corporate wellness programs. The scores below are illustrative, based on what a founder might learn from a handful of interviews and review reading.

Independent physios win. They run their own calendars, so the pain of no-shows and double bookings is personal. They are easy to find through professional groups and local associations. Big enterprise tools are overkill for them, and generic booking apps do not handle treatment-plan sessions well. Corporate wellness pays well but scores 1 on reachability, which makes it a poor first bet.

What the choice changes

The scores are a thinking tool, not a verdict. If interviews contradict them, change the numbers.

Turn your segment into MVP scope and first-user outreach

A segment only matters if it changes what you build. Take the segment’s top problem and turn it into one core journey. For the physiotherapists, that is “patient books and shows up.” Then list three to five features that make that journey work end to end. If you need help drawing the line, read how to decide what goes into version one.

Everything else goes on a “later” list, labeled with the segment it belongs to. Multi-staff rosters are not a bad idea. They are a small-clinic feature, and small clinics are your second segment. Keeping that list stops scope creep from the people you are not serving yet.

Finding your first 20 people

A good segment is one where you can name the places its members gather. For physiotherapists, that could be:

• Professional association directories and member forums
• Facebook and LinkedIn groups for private practitioners
• Local continuing-education events
• Small practices you can message directly

Ask questions about their current behavior, not your idea: How do you handle bookings today? What happened the last time a patient did not show? What have you tried and dropped? Offer a short call, then ask whether they would try an early version. The first 1,000 users playbook covers what comes after those first conversations.

Inside TwoSprint, the segment you choose carries through research, features, user journeys and tasks, so your plan stays tied to one person instead of drifting back to “everyone.”

Common segmentation mistakes

• Too broad. “Everyone with a phone” is not a segment. If you cannot picture where to find 20 of them, narrow it down.
• Too many at once. Three segments means three sets of messages, features and channels, with a team of two.
• Demographics only. Two 35-year-old founders can need completely different tools. Segment by problem or job-to-be-done first.
• Choosing by size alone. The biggest segment is often the hardest to reach and the best served.
• Never revisiting. Your first choice is a hypothesis. Review it after your first 20 to 50 users.

Signs it is time for the next segment

• Users in your first segment are retained and refer others.
• You can describe their journey without guessing.
• Inbound requests start coming from an adjacent group.
• Growth in the first segment is slowing because you are reaching most of it.

Startup segmentation checklist

Copy this into your notes and fill it in.

• Problem and job written in one sentence each
• 4 to 6 candidate segments listed, using problem-based, firmographic or behavioral cuts
• 5 to 10 conversations held per top candidate
• Forums, reviews and public data checked for each
• Each segment scored 1 to 5 on pain, ability to pay, reachability, size and competitor weakness
• Each segment checked against measurable, reachable, substantial, distinct and actionable
• One beachhead segment chosen, with a written reason and a note on what would change your mind
• Persona and core journey written for that segment
• 3 to 5 MVP features listed, with everything else parked under later segments
• List of 20 named people or places to reach them
• Review date set for after your first 20 to 50 users

Doing this with TwoSprint

If you want to run these steps in one place, TwoSprint walks you from a rough idea to a plan: candidate segments, a scored pick, then a persona, a core journey and a short feature list. It follows the same flow as how to turn a raw idea into a product plan. A spreadsheet and a few interviews will also get you most of the way. What matters is that you choose one segment and write down why.