UX/UI Design, A to Z ┃ 2.2 Empathy Maps: Understanding the Problem From the User's Side

Welcome back! Missed you :)

An empathy map isn't a table where you imagine a user. It's a tool for connecting evidence from research into a single point of view.

Last lesson, your affinity diagram found patterns across many users. This one moves the opposite way — down into one user's experience, connecting what they say, do, think, and feel.

And it draws a line most beginners skip:

The line between what you heard and what you're guessing.

What Is an Empathy Map?

An empathy map is a collaborative tool that organizes what your team knows about one type of user. It pulls qualitative data from interviews and observation into one place, so everyone can see what the user says, does, and how they make decisions in context.

The core discipline is honesty about your sources. What you directly heard and observed goes down as evidence. Thoughts and feelings — which you can't observe directly — get marked as assumptions to verify later. Blur that line and the map becomes fiction with a professional layout.

Why Use One?

• Compresses research. Turns interviews and observations into something short and readable.
• Aligns the team. Replaces several private mental models of "the user" with one shared view.
• Surfaces patterns and gaps. Reveals repeating needs, mismatches between words and actions, and questions you haven't answered.
• Leads to problem definition. Uses the user's goals and obstacles to sharpen the problem worth solving.

When Do You Make One?

Right after qualitative data comes in — interviews, observation — and before you define the problem or start generating ideas.

Running an affinity diagram first is efficient: you cluster the raw research, then map it. But that order isn't mandatory, and the map keeps changing as new findings arrive.

Before You Start

Three things to get right first:

1. One user, one map

• Pick a single person or one clearly defined user type. Never mix different user groups onto the same map — their goals and behaviors will contradict each other.

2. Gather your evidence

• Interview quotes, observation notes, open survey responses.

3. One note, one thing

• And keep sourced facts visually separate from team assumptions.

How to Fill It In

This follows Dave Gray's 2017 Empathy Map Canvas. One user type sits in the center; around them you connect the observable outside and the internal motivations.

A useful way to read the canvas: the outer sections (See, Say, Do, Hear) are things you can observe from the outside. The center (Think & Feel) is what you have to infer — and the whole map builds toward it. Here's each section with a travel-app example.

1. Who are we empathizing with?

• Who is this person, and what situation or role are they in?

🟨  A busy professional trying to plan a quiet trip in very little time.

2. What do they need to do?

• What goal, decision, or task are they trying to accomplish?

🟧  Finalize a satisfying itinerary within limited time and budget.

3. What do they see?

• What's in their environment — competitors, content, what others around them are doing?

🟩  Travel apps, price-comparison sites, colleagues' social posts, travel ads.

4. What do they say?

• What words did they actually use in the interview?

🟦  "Planning needs to be as simple as possible."

5. What do they do?

• What behavior was actually observed or reported?

🟪  Searches during their commute, screenshots places, compares prices.

6. What do they hear?

• Which voices influence them — friends, family, colleagues, media?

🟥  A colleague's recommendation, family advice on budget, a travel YouTuber's review.

7. What do they think and feel?

• The concerns, hopes, and motivations that aren't visible. The Pains and Gains.

⬛  Pains: anxiety about wasting scarce time, fear of a trip going wrong.

⬜  Gains: a genuine rest, the satisfaction of an efficient plan.

Keep two sections honest.

In Say, use real quotes. In Think & Feel, treat everything as a hypothesis to confirm through research — because you're inferring an inner state you cannot see.

From Filled-In Map to Insight

Filling the boxes isn't the finish line. The value is in the relationships between sections. Three places are especially worth looking:

• Goals and struggles that repeat across multiple interviews
• The gap between what a user says and what they actually do — often the richest signal on the whole map
• The Pain blocking a Gain — the obstacle standing between the user and the outcome they want

Example insight:

"Busy professionals have to plan trips in very little time, so they need guidance that narrows choices quickly — not more information to sift through."

That say-do gap deserves special attention. Someone insists budget is their top priority, then books the convenient option at twice the price. Neither statement is a lie — but the gap between them is where the real need usually hides.

Mistakes to Avoid

1. Filling blanks without evidence.

• If you don't know, don't invent — write "needs more research." An empty box is more honest than a fabricated one.

2. Mixing user groups on one map.

• Different goals and behaviors belong on separate maps.

3. Turning quotes into your interpretation.

• Record the direct quote and the observed fact first; mark interpretation separately.

4. Treating the map as final truth.

• It's a hypothetical model summarizing current research. New data revises it.

5. Thinking it replaces personas or research.

• An empathy map synthesizes findings. It doesn't replace the research itself or a finished persona.

The Goal Is a Sharper Problem

An empathy map is not decoration for feeling like you understand the user. It exists to separate fact from assumption, connect the user's goals, behavior, and emotion, and define the problem you actually need to solve more precisely.

A good empathy map isn't the one with every box full. It's the one that shows you what to learn next and which problem to tackle.

Key Takeaways

• An empathy map connects research evidence into one user's point of view — it is not imagination in a template.
• One user or user type per map. Mixing groups produces contradictions.
• Separate what you heard and saw from what you're inferring. Quotes are evidence; thoughts and feelings are hypotheses to verify.
• The insight lives in the relationships between sections — especially the gap between what users say and what they do.
• An empty box beats a fabricated one. "Needs more research" is a valid entry.
• It synthesizes research; it doesn't replace research or a persona.
• The point is a sharper problem definition, not a feeling of understanding.

Before You Go

A 10-minute exercise

Below are real-style fragments from one interview with a food-delivery app user. Sort each into the right empathy map section — and flag which ones are assumptions you'd need to verify.

a. "I just want to order in two taps and be done."

b. Reordered the same restaurant four times last month (from their history)

c. Scrolls past new restaurants without opening them

d. A friend keeps recommending places they never try

e. Probably worried about wasting money on food they won't like

f. Wants to feel taken care of after a long day

1. Place each fragment: Say, Do, See, Hear, or Think & Feel.
2. Which two are inferences, not things the user told or showed you? (Look hard at e and f.)
3. Now find the say-do gap. Compare a with b and c. What does the contradiction suggest the user actually needs?

The gap between "two taps and done" and reordering the same place four times isn't a contradiction to resolve — it's the insight. Speed matters more to this user than variety.

What's Next

User Personas: Built From Research, Not Assumption

You started this series with a proto-persona — a guess. Now you have research. This is where the two meet.

We'll cover what separates a research-based persona from that early guess, and how to build one from the interviews, affinity diagram, and empathy map you've already made.

Remember the assumptions on your proto-persona? Time to see which ones survived!👍

If this helped, stick around — follow the blog for the next one, and send it to a friend who's just starting out.

Thank you! 🙌