Empathy Maps in UX Research

An empathy map is a collaborative visualization that captures what a team knows about a user's behaviors, thoughts, and emotions. It's not a timeline or a journey. It's a snapshot that gives teams a shared foundation for design decisions.

Traditional empathy maps divide into 4 quadrants with the user or persona at the center:

• Says captures direct quotes from research.
• Thinks surfaces what users might be processing but not saying out loud.
• Does records observable behaviors.
• Feels notes their emotional state.

Empathy maps come in two forms. A one-user map reflects a specific person, usually built from a single interview or a diary study entry. An aggregated map synthesizes patterns across multiple users who share similar behaviors, and it's often a useful starting point before creating personas.

Empathy maps are most useful at the start of a design project, before requirements are defined and concepting begins. That's when teams most need a shared, grounded picture of the user to avoid defaulting to assumptions.

Both the process of building a map and the finished artifact offer value:

• Analyze qualitative research. Empathy maps help teams make sense of research notes, interview transcripts, and survey answers, and surface gaps in what they still need to find out.
• Communicate users to the team. The finished map gives everyone, including stakeholders, a visual reference for user attitudes and behaviors that can protect design decisions from bias.
• Collect data directly from users. When filled in by users themselves after a session, an empathy map can act as a secondary data source and a compact summary of that research.^[1]

A typical empathy map includes four quadrants:

• Says: What the user says about the product. Ideally, this section contains real quotes from users recorded during interviews or user testing sessions.
• Thinks: What is the user thinking about when interacting with a product? What occupies the user's thoughts? What matters to the user?
• Feels: This section contains information about the user's emotional state. What worries the user? What does the user get excited about? How does the user feel about the experience?
• Does: What actions does the user take? What actions and behaviors did you notice?

Some of these quadrants may seem ambiguous or overlapping. For example, it may be difficult to distinguish between the Thinks and Feels sections. Do not focus too much on being precise — if an item appears to fit into multiple quadrants, just pick one.^[2]

The Says quadrant captures what users say out loud during interviews or usability studies. The goal is to fill it with verbatim quotes pulled directly from your research, not paraphrased summaries.

Direct quotes are valuable because they reflect exactly what users expressed, without interpretation. A researcher listening to a participant say "I don't understand what to do from here" gets something more actionable than a note that reads "user was confused."

Some examples of what goes in Says:

• "I am loyal to Delta because I never have a bad experience."
• "I want something reliable."
• "I don't understand what to do from here."

The Thinks quadrant captures what users are thinking throughout the experience, based on your qualitative research. The key question to ask: what occupies users' minds, and what matters most to them in this context?

Thinks can overlap with Says. A user might say "this is taking too much time" and think it simultaneously. But the more revealing entries are the ones users would never say out loud. Users might hold back because they feel self-conscious, want to seem polite, or are unsure whether their frustration is valid.

You won't get this data directly. Instead, infer it from behavioral cues: hesitations, facial expressions, body language, and moments where what users do contradicts what they say. Interview techniques like think-aloud protocols, where users narrate their thought process in real time, can also surface thoughts that would otherwise stay hidden.

Some examples of what goes in Thinks:

• "This is really annoying."
• "Am I dumb for not understanding this?"
• "This is taking too much time."

The Does quadrant captures the actions users take, based on what you observe or uncover in research. The goal is concrete, physical behavior: what users actually do, not what they say they do.

Some examples of what goes in Does:

• Refresh a page several times
• Shop around to compare prices
• Check the size chart

Pay attention to moments where actions contradict words. A user might express satisfaction but repeatedly backtrack through a flow, or complain about a feature they keep using anyway. These contradictions are some of the most valuable things an empathy map can surface. They point to unmet needs and design tensions that users themselves may not be able to articulate.

The Feels quadrant captures the user's emotional state during the experience. Each entry typically pairs an emotion with a short sentence for context, making it clear what triggered the feeling.

To fill this quadrant, ask: what worries users, what frustrates them, and what makes them feel confident or excited? Emotions are inferred from research, not stated outright, so look for cues in tone, word choice, and behavior.

Some examples of what goes in Feels:

• Impatient because the pages load too slowly
• Confused by too many contradictory prices
• Worried about making a mistake

While you can create an empathy map solo, doing it with a team brings in multiple perspectives and builds shared ownership of the findings.

Follow these 5 steps:

1. Define your scope and goals. Decide who the user is and what task or experience you are mapping.
2. Gather your materials. Work on a whiteboard with sticky notes and markers, or use a remote collaboration tool like Miro or FigJam.
3. Collect your data. Run interviews, direct observations, contextual inquiries, or diary studies. Bring all findings to the team before you start mapping.
4. Generate ideas as a group. Have everyone read through the data and add sticky notes to each quadrant independently.
5. Cluster and name themes. Bring similar notes together, label the clusters, and discuss what patterns emerge.

Once the map is complete, digitize it and add context: include the user or persona, any open questions, and the date and version number.

The 4-quadrant format works well for initial analysis, but it's not the only option. If you need more detail, add quadrants like Goals, Pains, and Gains to capture what the standard map leaves out.

For UX-focused workshops, the format adapted by Paul Boag, a UX consultant and author with over 30 years in the field, is worth considering. Boag found the traditional map too generic for design work and built a version around UX-specific questions:

• Feelings: How do users feel about the experience? What matters to them?
• Tasks: What are users trying to complete?
• Influences: What people, places, or things shape how users act?
• Pain points: What obstacles, fears, or frustrations are users hoping to overcome?
• Goals: What is users' ultimate goal, and what are they trying to achieve?

The broader principle holds: empathy maps are flexible tools. Adapt the format to fit your research goals rather than using it as a fixed template.^[3]

Empathy maps are most useful at the beginning of the design process, right after user research. Creating empathy maps helps synthesize research observations, reveal deeper insights into users' needs, and see things from their points of view.

Empathy maps provide you with the information needed to develop requirements and concepts. You must understand users' attitudes and behaviors before creating solutions — be it content ideas, webpage design, app prototypes, or a new service offering.

They can also be used throughout the design process and revised as new data becomes available. A sparsely populated map indicates where more user research needs to be done.