TL;DR: Empathy Map
- An empathy map is a 4-quadrant tool capturing what users Say, Think, Do, and Feel.
- It turns raw research data into a shared, visual understanding of your user.
- Empathy maps ≠ user personas – they’re complementary tools, not substitutes.
- Most teams build empathy maps on assumptions; the ones built on real research actually move the needle.
- AI-powered research tools can now populate an empathy map in 30 minutes without recruiting a single participant.
Here’s a situation you’ve probably been in: your team is two weeks from launch and someone asks, ‘Do we actually know what our users are frustrated by?’ A quick round of sticky notes follows. Everyone shares their gut feelings. The session ends with a Miro board that’s visually impressive but built almost entirely on internal assumptions.
That’s the empathy map problem no one talks about. The tool itself is solid – it’s the data going into it that’s usually thin.
This guide covers everything you need to build a proper empathy map – the quadrants, the process, real industry examples, and the mistakes that quietly wreck them. And at the end, there’s a practical look at how to get the user data needed to fill one honestly, including how tools like Articos are making that part dramatically faster.
What Is an Empathy Map and How to Use It in UX Design
An empathy map is a collaborative visualization tool that captures what a specific user – or user group – says, thinks, does, and feels. Originally developed by Dave Gray at XPLANE, it’s been widely adopted across design thinking, agile, and product development workflows.
The core purpose is deceptively simple: get everyone on your team looking at the same user, from the same angle, at the same time. Without that shared frame, designers optimize for usability while engineers optimize for performance and marketing writes copy for people who’ve never actually used the product.
| Definition | An empathy map is a structured framework used to articulate what a team knows (or needs to learn) about a particular type of user. It externalizes user knowledge into a shared visual, helping teams align on user needs and make better product decisions. |
Empathy maps work best at the start of a design or product cycle – before wireframes, before feature specs, before anyone has decided what ‘the solution’ looks like. They’re also useful mid-project when user research reveals new behavior patterns that don’t match initial assumptions.
The Four Quadrants of an Empathy Map (PAA)
Traditional empathy maps are divided into four quadrants, with the user or persona placed in the center:
| Quadrant | Captures | Data Source | Watch For | Example |
| Says | Verbatim quotes from interviews/sessions | User interviews, usability tests | What they say vs what they do | “I want something I can trust immediately.” |
| Thinks | Internal thoughts, worries, beliefs – not always voiced | Interviews, diary studies, contextual inquiry | Gaps between Thinks and Says | “Am I making the wrong choice here?” |
| Does | Physical actions, behaviors, workarounds | Observation, session recordings, analytics | Workarounds = unmet needs | Refreshes page 3×, shops 4 tabs at once |
| Feels | Emotional state during the experience | Tone, body language, survey adjectives | Contradictions with Says/Does | Anxious: worried about hidden fees |
Some teams add two additional zones below the quadrants – Pains and Gains – to capture user frustrations and motivations. This is optional but useful when you need to feed the empathy map directly into a value proposition or Jobs-to-be-Done framework.
See and Hear Quadrants
Some empathy map frameworks extend beyond the core four quadrants (Think, Feel, Say, Do) to include:
- See – what does the user observe in their environment? What do they see their peers doing? What marketplace signals reach them?
- Hear – what are they hearing from colleagues, media, influencers, or family that shapes their worldview and decisions?
These extended quadrants are particularly useful in B2B contexts where peer influence and industry signals heavily shape behavior.
Empathy Maps vs. Journey Maps
| Empathy Map | Journey Map | |
| Focus | Who is the user? (mindset, emotions, worldview) | What does the user experience over time? (touchpoints, actions, feelings at each stage) |
| Time dimension | Snapshot | Sequence |
| Best used for | Understanding user psychology before design | Designing the experience across a complete interaction |
| When to create | Early discovery phase | After defining the user and the service |
Empathy maps and journey maps are complementary. Create empathy maps to understand who you’re designing for; create journey maps to design the experience they’ll have.
How to Create an Empathy Map Step by Step
Building an empathy map takes about 60–90 minutes when done properly – shorter if you’re working from pre-existing research, longer if you’re doing it live with a team. Here’s the process:
Step 1: Define the User You’re Mapping
Pick one user or persona per empathy map. If you have three user segments, you need three maps. Mixing them produces a fictional composite that describes no one accurately.
If you haven’t done user interviews yet, now is the time – or consider running a quick round of AI-powered user research to gather initial behavioral and attitudinal signals before the mapping session.
Step 2: Gather Your Research Materials
Pull together any qualitative data you have: interview transcripts, usability session notes, support tickets, NPS survey comments, session recordings. These are your raw inputs. No data = a guessing session, not a research session.
| Contrarian Take Worth Noting | Most empathy maps you’ll see online are built almost entirely from team assumptions. They feel like research because they’re structured and visual – but without real user data as input, they’re just organized opinions. That’s a fundamental gap the top guides quietly ignore. |
Step 3: Set Up Your Canvas
Draw or open a digital template divided into four quadrants. Label them Says, Thinks, Does, Feels. Put the user name/persona in the center. Tools like FigJam, Miro, or even a physical whiteboard all work – the format matters less than the conversation that happens around it.
Step 4: Fill in the Quadrants (Start with Says and Does)
Begin with what you can directly observe: verbatim quotes for Says, and recorded behaviors for Does. These are harder to fabricate and anchor the session in actual evidence.
Step 5: Work Through Thinks and Feels – Separately
This is where most teams rush and conflate the two. Here’s the distinction that matters:
- Thinks = cognitive layer. What belief or assumption is shaping their behavior? Often unstated.
- Feels = emotional layer. What’s the underlying emotional response – anxiety, excitement, frustration?
| How do I fill out the Thinks and Feels section in an empathy map? | For Thinks, look for what the user believes but doesn’t say out loud – self-doubt, assumptions about your product, background worries. For Feels, use emotional adjectives paired with context: not just ‘frustrated’ but ‘frustrated because the pricing page didn’t answer her main question.’ If you’re stuck, ask: ‘What would this person be telling a friend about this experience tonight?’ |
Step 6: Look for Contradictions
The most useful insights come from mismatches – where Says conflicts with Does, or where Thinks contradicts Feels. A user who says ‘I research everything carefully’ but impulse-purchases within 10 minutes is telling you something important. These contradictions are where product opportunities hide.
Step 7: Identify Pains and Gains
Add these optional rows at the bottom. Pains capture what frustrates, blocks, or worries the user. Gains capture what they’re hoping to achieve or feel. This directly feeds into product positioning and feature prioritization.
Empathy Map vs User Persona: What’s the Difference
This is probably the most Googled question in this space, and the short answer is: they’re not the same thing, and you shouldn’t have to choose between them.
| Dimension | Empathy Map | User Persona |
| Focus | Mindset & emotional state during an experience | Full profile: demographics, goals, behavior patterns |
| Format | 4 quadrants (Says, Thinks, Does, Feels) | 1-page character sheet with photo, bio, goals, frustrations |
| Granularity | Specific to a context or task | Broader; covers multiple touchpoints |
| Best used for | Understanding a specific user moment | Aligning team on target user type overall |
| Created from | Research sessions, interviews, observation | Aggregated research across many users |
| Relationship | Can feed INTO persona creation | Can be visualized THROUGH empathy maps |
The practical sequence: run research → build individual empathy maps per research participant → identify patterns → use those patterns to construct or refine a persona. Empathy maps are often a step in persona creation, not a replacement for it.
| Should I create one empathy map for each persona? | Yes – each empathy map should represent one user or one user segment. If you have three personas, build three maps. Blending them into one produces averaged, inaccurate insights. That said, you can create an ‘aggregated’ empathy map that represents a user segment by combining themes from multiple individual maps – but only after doing the individual ones first. |
How Empathy Maps Help You Find User Pain Points and Goals
According to research from Forrester, companies that lead in customer experience outperform laggards by nearly 80% in revenue growth. Empathy maps are one of the more practical tools for closing that gap – specifically because they force teams to articulate where users are struggling, not just where they’re succeeding.
The connection between empathy maps and pain points is structural. Here’s how each quadrant reveals friction:
- Says: Direct complaints and workarounds signal surface-level frustrations (‘I always have to Google this’)
- Thinks: Unvoiced doubts and mental blocks reveal deeper anxieties (‘Am I using this right?’)
- Does: Workarounds in behavior show where your product is failing them without them complaining (‘opens a spreadsheet instead of using the built-in feature’)
- Feels: Emotional responses are early churn signals. Anxious users don’t stay. Embarrassed users don’t refer.
The Pains/Gains section at the bottom translates these signals into actionable product language. Pains become feature problems to solve. Gains become outcomes to promise. That’s the direct line between an empathy map and your product roadmap.
For B2B SaaS teams in particular, the Thinks and Feels quadrants often surface the real switching costs – not price, but fear of disruption or sunk-cost attachment to existing workflows.
Real Empathy Map Examples for SaaS, E-commerce, and Mobile Apps
Abstract frameworks make more sense with concrete examples. Here are three – one per industry – based on composite user research patterns.
Empathy Map Example: B2B SaaS (Project Management Tool)
Persona: Maya, 34, Product Manager at a 60-person startup
- Says: “My team is in three different tools right now and nothing talks to each other.”
- Thinks: “If I pick the wrong tool, I’ll have wasted three months of adoption effort.” (Migration anxiety – rarely voiced)
- Does: Keeps a personal ‘master spreadsheet’ on the side to track what the tool doesn’t capture. Opens competitor websites before renewals.
- Feels: Overwhelmed by feature bloat, skeptical of vendor promises, cautiously optimistic when she sees a clean onboarding flow.
- Pain: Can’t get buy-in from engineers who think the tool is ‘too PM-centric’.
- Gain: Wants to look competent in front of her VP with clear sprint visibility.
Empathy Map Example: E-commerce (Checkout Abandonment)
Persona: Tariq, 29, casual online shopper buying a gift
- Says: “I don’t want to create an account just to buy one thing.”
- Thinks: “Is this site legit? I’ve never heard of it.” (Trust gap – expressed through hesitation, not words)
- Does: Opens the return policy in a new tab before adding to cart. Searches ‘[brand] reviews Reddit’ mid-checkout.
- Feels: Impatient with a 4-step checkout, anxious about entering card details on an unfamiliar site.
- Pain: Surprise shipping costs at step 3 of 4.
- Gain: Wants the gift to arrive before Friday with zero friction.
Empathy Map Example: Mobile App (Fitness Tracker)
Persona: Priya, 26, casual runner trying to build a habit
- Says: “I open it every day for about the first week, then I forget about it.”
- Thinks: “Other people seem to use this consistently – what am I doing wrong?” (Comparison anxiety)
- Does: Logs only when she remembers. Skips the goal-setting onboarding. Turns off notification permissions on day 2.
- Feels: Motivated on Day 1, guilty by Day 10, avoidant by Day 14.
- Pain: Notifications feel generic and slightly shaming.
- Gain: Wants to feel like the app is on her side, not tracking her failures.
| Note on These Examples | Notice what makes them useful: specificity. ‘Opens competitor websites before renewals’ and ‘Searches [brand] reviews Reddit mid-checkout’ are the kinds of behavioral details that only show up in real research – not team brainstorms. That’s the difference between a map that informs decisions and one that decorates a Confluence page. |
Limitations of Empathy Mapping
Empathy maps are useful but have real limitations:
- Relies on researcher assumptions – without actual user research, an empathy map reflects the team’s beliefs about the user, not the user’s reality
- Static snapshot – empathy maps don’t capture how user thinking and behavior changes over time
- Oversimplification risk – collapsing complex human behavior into four quadrants loses nuance; treat empathy maps as starting points, not endpoints
- Can entrench wrong assumptions – if created without validation, they can give false confidence in user understanding
Common Mistakes to Avoid When Making an Empathy Map
What mistakes should I avoid when making an empathy map?
Most empathy map guides stop at ‘here’s the template.’ Here are the mistakes that actually kill the value of one:
1. Filling It with Assumptions Instead of Data
This is the big one. ‘I think our user feels frustrated’ is not empathy map content. If you don’t have research, run some first. Even 3–5 short user interviews produce far more accurate inputs than an entire team’s gut instincts combined.
2. Treating It as a One-Time Exercise
Empathy maps go stale. User behavior shifts, markets change, onboarding flows evolve. Schedule a quarterly review – or at minimum, revisit whenever you’re about to make a significant product decision.
3. Confusing ‘Says’ with ‘Thinks’
Says = what they actually said in your session, verbatim or close to it. Thinks = what they held back, believed internally, or couldn’t articulate. If you’re writing ‘Thinks: users want a faster checkout’ – that’s either a Says quote or a Gains statement. It’s not a thought.
4. One Map for All Users
A SaaS product with three ICPs needs three empathy maps. Combining a power user and a casual user into one map creates a fictional average person who doesn’t exist and can’t be designed for.
5. Skipping the Contradictions Review
Most teams fill the quadrants and call it done. The real analysis happens when you compare across quadrants – specifically, where Says contradicts Does, or where Feels contradicts the actions being taken. Those gaps are your highest-signal design opportunities.
6. No Clear Owner or Next Step
An empathy map without a downstream action is wall art. Assign each session an owner responsible for turning the map’s insights into specific research questions, feature requirements, or copy changes.
7. Using It to Validate Decisions Already Made
This happens more than anyone admits. The map gets created post-decision to justify a roadmap item. At that point it’s not user research – it’s rationalization. The map should inform decisions, not confirm them.
Where to Find Free Empathy Map Templates
There are several solid options depending on your workflow:
- Miro – Best for remote teams doing live collaborative sessions. Their template includes Pains/Gains zones and works well on a shared board.
- FigJam (by Figma) – Good if your team is already in Figma. Clean and editable.
- Mural – Similar to Miro. Useful for structured facilitation with larger teams.
- Canva’s empathy map tool – Simple and visual. Good for presentation-ready versions after a session.
- Printed whiteboard format – For in-person workshops, a hand-drawn version on a whiteboard or A3 paper often generates more discussion than a polished digital version.
| When NOT to use a template | Templates are starting points, not straitjackets. If your product context involves highly specialized workflows (healthcare, legal, industrial B2B), consider customizing the quadrant labels to match the specific context. ‘Thinks’ might become ‘Compliance Concerns’ and ‘Feels’ might become ‘Risk Tolerance’ – the underlying logic stays the same, but the prompts become more relevant. |
How Articos Helps You Build Empathy Maps Backed by Real Data
The biggest obstacle to a useful empathy map isn’t the template – it’s getting research-grade user data to fill it with. Traditional methods involve recruiting participants, scheduling sessions, running interviews, transcribing them, and then synthesizing. That process typically takes 4–6 weeks minimum.
Articos cuts that down to under 30 minutes. The platform uses AI-powered synthetic personas that are built to reflect real human behavioral and attitudinal patterns – without you having to find, schedule, or incentivize a single participant.
Here’s what that looks like in practice for empathy mapping:
- Define your target user: Enter the demographic, role, and context for the person you’re mapping.
- Run your questions: Ask what you’d ask in a real interview – about their process, frustrations, goals, and emotional state.
- Get structured output: Articos returns insights organized by attitudinal and behavioral themes – exactly what goes into Says, Thinks, Does, and Feels.
- Fill your empathy map: Transfer the insights into your map with actual data behind each quadrant.
This is particularly useful for startups and agencies who need research velocity that traditional methods can’t deliver – and for teams exploring new personas before committing to a full recruitment-based study.
The platform also delivers 90% organic-synthetic parity, which means the insights consistently align with what you’d hear from real participants – without the politeness bias or social pressure that sometimes skews live sessions.
📌 Image Placement 5: Articos dashboard showing synthetic persona insights being mapped to empathy map quadrants – clean product screenshot style (place in this Articos section)
Final Thoughts
Empathy maps are one of those tools that look easy but are only as good as what you put into them. The structure is the simple part. The discipline of filling them with real user data – and actually acting on what they reveal – is where most teams fall short.
If you want research that reflects what users actually think and feel – not what your team imagines they think and feel – the data collection step is where it starts. For teams who can’t spend six weeks recruiting and interviewing, Articos makes that step fast. The map itself is still yours to build.
| Try Articos Free – Build Your Empathy Map in 30 Minutes No recruitment. No scheduling. Real insights from AI personas that reflect your target user. Start your free trial at platform.articos.com |
