Empathy Map: Exercising empathy with your user

Most teams skip user research when there's no budget or time for it. Instead of guessing, they default to assumptions — and those assumptions go untested. The Empathy Map exists to make that gap visible. It forces you to write down what you think you know about your user, and — more importantly — to see where your knowledge has holes.

Here's how I use it, when it works, and where it breaks down.

What an Empathy Map actually does

The Empathy Map is a visual canvas that organizes what you know (or assume) about a user into seven areas: who they are, what they need to do, what they see, say, do, hear, think and feel. It was originally created by the consultancy Xplane for business model design, but it's become a standard tool in UX for building proto-personas — personas based on hypotheses rather than research.

The important distinction: an empathy map is not research. It's a structured way to surface assumptions so you can decide which ones to test first.

The seven areas, and what to actually put in them

The map has seven sections. For each one, I'll include the standard questions and a practical note on what makes the answers useful.

1. WHO are we empathizing with?

Be specific. "Small business owner" is too broad. "A bakery owner in a small town who processes 20 sales a day on a phone with limited data" gives the team something real to work with.

2. What do we want this person to DO?

This section anchors the map to an outcome. Without it, the rest of the exercise turns into abstract character-building.

3. What do they SEE?

4. What do they SAY?

If you're filling this out without direct user quotes, label every answer as an assumption. That honesty is the whole point.

5. What do they DO?

6. What do they HEAR?

7. What do they THINK and FEEL?

This last section is where teams tend to project the most. My rule: if you can't tie a "think and feel" answer to something observable — a behavior, a quote, a pattern — flag it as speculation.

Where empathy maps work — and where they don't

Empathy maps work best in two situations:

Where they break down: when teams treat the map as a finished artifact instead of a hypothesis sheet. I've seen teams fill out a map once, pin it to the wall, and never update it — even after talking to real users. At that point, the map becomes decoration, not a tool.

The other risk is groupthink. If one person — usually the most senior — fills in the map and everyone else nods, you've just documented one person's assumptions with a collaborative veneer. The fix: have team members fill sections independently before comparing answers. The disagreements are where the real learning happens.

How to run the exercise

The format is simple. You can use paper and post-its, a whiteboard, or any digital canvas.

The output isn't the map itself — it's the list of things you now know you need to validate.

The practical takeaway

An empathy map is useful exactly to the degree that you treat it as disposable. Fill it out, use it to plan your next research round, then update or discard it when you have real data. The moment you treat it as truth, it stops working.

If you're using it on a project right now, try this: after filling in the map, count how many answers are based on direct observation vs. assumption. If more than half are assumptions, you've just built your research backlog.