Describe what someone is doing when the problem happens, not who they are on paper. "Parents trying to get a daycare spot" is a target market. "Working women with young children" is a census category, and you cannot go and find those people to talk to.

You need this before anything else in the research, because you cannot run user interviews if you do not know who your user is.

Start with behaviour and circumstance

Write down the moment the problem occurs and what the person is doing in it. Who else is involved. What they are using right now to cope. How often it happens.

You will be wrong about some of it, and that is the point. What you are writing is a guess that your interviews will confirm, disprove or reframe. Afterwards that person is no longer made up in your head, they are based on evidence.

I got mine wrong in a useful way. I assumed parents with nannies were commuting and working outside the house, with the nanny starting early, because that was my own situation as a nanny. That was 10 years ago and the landscape has changed.

Separate the user from the customer

This is the distinction that changes products. For my nanny app, nannies are the users, and the parents are the customer, because the parents are the ones who will pay.

Once you see that split, you can also see that they may have different problems. The parents might not care about a shared book at all, and might care enormously about working out what the nanny is owed this week. You could sell a product to a group who does not want the main feature.

Write down both: who uses it, and who pays.

Narrow until you can find two of them

A practical way to cut: keep narrowing the group until it feels almost too specific, then go and find two people from each segment you have written down and talk to them about their experience. Two per category is enough to tell whether your segments are real.

You will know you have found a market when they start repeating each other. After four or five people of the same kind, the same problems come up again and again, in different words. That repetition is the signal that these people belong to one group rather than being individuals you happen to know.

And when they do not repeat each other, that is information too. It usually means the group you drew is actually two groups.

The question people forget to ask

Can you stand a thousand of this person in your life?

If the product works, these people are in your life for years: your support inbox, your reviews, your community. You will know quite quickly in the interviews. It is either "I want a thousand users like you" or "I do not want a thousand of you in my life", and that is a legitimate input into who you build for.

When you find out someone is not your person

That happens during competitive research too. Looking at who a competing product attracts tells you what a segment expects, and sometimes it tells you that the segment is not yours. Headspace and Calm are both meditation apps and they are not serving the same person, whether or not their makers say so.

Do you need market size numbers?

A basic sense of it, yes. Your numbers do not have to be exact, they just have to show what is there, with a note about whether this is a niche where growth will be slow because you have to go and find each person. Skip financial projections and user-growth predictions entirely. They set expectations that nothing supports.

If you decide to go broad anyway

Sometimes the honest answer is that your thing serves people at a stage rather than people of a type, and a narrow persona would be a fiction. That is a legitimate choice and it is not free: broad means harder marketing, vaguer copy, and a longer search for the people who are actually yours. Make it deliberately, write down why, and revisit it once you have talked to twenty people.

Keep reading

How Do I Validate an App Idea Without Building It?

How to Find Your First Users for a New App, Without a Following

Is My App Idea Good Enough? Here's How to Actually Find Out

Learn this with me

Defining your target user is class 2 of From Passion to Product, my free six-week live course, including the difference between a primary and a secondary user. You turn that into a persona in class 3 and then test it against real people in class 4, which is the order that matters: a target market you decided on Monday is a guess until someone outside your head confirms it.

Six weeks, live, and you do it on your own idea.

Join the waitlist to hear when the next free cohort opens, or start with the App Strategy Workbook.