A persona is a profile of one specific person who represents the people you are building for, with a name, built from evidence you collected rather than from imagination. It is not your target market. It is who you are building this app for.

Once you have one, you stop talking about users. You talk about Lisa.

It took me years to understand what a persona was actually supposed to be. It represents your user. It is not your idealistic user, and it is not who you hope will turn up.

What it is for

A persona earns its keep at the moment you reach a decision and do not know which way to go. You look at the persona and ask what she would do.

The test of whether yours is any good is whether your whole team can say her name out loud in a meeting without it feeling silly. Al wants to be able to service his wheelchair himself. That is a sentence a team can design from. That is why it is called a persona: you are personifying this person.

That is not a trick to feel productive. It is a real decision tool. I come to a feature and I think, would Lisa like this? Her primary concerns are knowing about the kids' days and communicating, so maybe a feature tracking the kids' location, which was another theme I saw come up, is not as important to her. That question has an answer, and "would users like this" does not.

Build it from evidence, not from your head

The persona you make is based on the evidence you collect from verbatim statements, meaning the things people actually said in your interviews, in their words.

You do not want to just randomly make stuff up. If you randomly make stuff up, you are making a product for an imaginary user. That is almost as bad as making a product just for yourself, as far as the growth and potential success of your product go.

If you are building it for yourself, then profile yourself as a user, honestly, and know that you are the target market.

Who to pick

You want the customer profile for the person most likely to love this app. Not just someone who will use it. Someone who will love it.

And you should only have one primary user, and then subsets. Deciding that is the hardest part of this whole thing.

For my nanny app I specify parents working outside the home, because I had to decide who the app is for, and people working inside the home and outside the home have completely different use cases.

Go past demographics

Age brackets tell you almost nothing on their own. What you want is what a group wants and why.

Take a travel app, and three sets of people a demographic filter would hand you as one audience. Couples whose kids have recently left home want to be told what to do, because they were tied down for so long and now they want a guide. Couples who never had children spent their adulthood discovering things on their own, and being handed an itinerary is the opposite of what they came for. A couple still travelling with two or three kids can find things themselves, and are relieved not to have to.

Same ages, same holiday budget, three different products.

The same thing is true of my own course. A group that wants to be millionaires and a group that wants to provide for their family and be happy are not one audience, whatever they have in common on paper.

More than one persona, when there really are two sides

The nanny app has two. Sarah, the nanny, is inputting information, so I have to build all of the forms, and the experience of putting things in has to work best for her. Lisa, the mum, technically could fill things in, but she is not doing it six hours a day. I do not mind a little friction for Lisa. I care about Sarah, because Sarah is the user for inputting.

That is what a second persona is for, a different job inside the product, not a second market you would like to have.

Also worth separating: who uses it and who you market to. You might build the app for the students, and market to their parents.

Skip the star ratings

Some templates have ratings on them, like stars out of five. I do not think ratings are good, because they can accidentally bias people.

If you put frustration at two out of five meaning it matters a bit but not that much, someone else reads two out of five as not important at all. Use short words instead of numbers.

A note on listening

When I did the user interviews for the Rivian mobile app, we had a profile that Rivian's user research team had set, and we found multiple people who matched it. You would think you would hear the same things from each of them. You did, if you listened carefully. If you listened only to what they said, they said different things.

The persona is built from the careful listening, not from the transcript surface.

Keep reading

How to Define Your Target Market When You Have an App Idea

Affinity Mapping: How to Make Sense of Your Interview Notes

How to Talk to Users When You're a Developer and Find It Deeply Uncomfortable

Learn this with me

You build your persona in class 3 of From Passion to Product, my free six-week live course, out of the verbatim statements you collected in your online research, and before you interview anyone. At that point a persona is roughly 75% guessing and 25% evidence. The guess gives you a direction, and the interviews in the next class correct it.

The course is free, runs six weeks and is taught live, and the persona you build is for your own app.

Join the waitlist for the next free cohort, or get the App Strategy Workbook and build yours at your own speed.