---

It Catches the Failures Your Persona Cannot

TL;DR

• An empathy map is a one-page artifact that captures what a real user is going through in a specific moment — what they Say, Think, Do, and Feel — and it exists because most product failures are small misses that look like nothing in isolation and like churn in aggregate.
• The teams that get value from empathy mapping aren’t filling boxes; they’re hunting for the gaps between what users say and what users actually do, which is where the truth sits.
• Run it in 60–90 minutes against one user type and one specific moment, with real interview quotes and support tickets in the room, and walk out with two or three decisions — not a wall of sticky notes everyone admires and nobody acts on.

Anyone Who’s Relied on Public Transport Knows What a Bad Product Feels Like

Anyone who’s relied on public transport knows the feeling. You step off one train, and you see the next one pulling away. You didn’t do anything wildly wrong. You were just off by a minute. Maybe you took the wrong exit. Maybe you followed a sign that wasn’t clear. The result is the same. You’re stuck waiting, recalculating, getting quietly annoyed at a system that quietly assumed you had perfect timing and zero pressure.

That’s what a lot of product experiences feel like. Teams spend time on what’s visible — screens, flows, features — while the real problem sits under the surface. They don’t actually understand the user’s situation. Not the “ideal user” with time to explore. The real one. Distracted, under pressure, switching between tools, trying to get something done without thinking too hard about your product at all.

Most product failures don’t look dramatic. They look like small misses. Tiny moments where someone loses confidence, hesitates, or has to improvise. A label that isn’t clear. A status that doesn’t update. A “what now?” moment, right when they can least afford it. Empathy mapping is one of the simplest ways to catch those misses before they become churn, workarounds, and support tickets.

An Empathy Map Captures What It’s Like to Be a Specific User in a Specific Moment

An empathy map is a one-page way of capturing what a user is experiencing in a specific moment through four lenses — Says, Thinks, Does, Feels. If a persona is your “who,” an empathy map is your “what it’s like to be them, right now.”

It’s popular in product discovery and design thinking workshops for a reason. It makes a team stop arguing from opinion and start aligning around reality. Instead of everyone holding a slightly different version of the user in their heads, you build a shared view. One that’s grounded in what people actually say and do, and honest about what they might be thinking and feeling. A good empathy map also shifts the team’s default question from “what should we build?” to something far more useful: “what is the user trying to get through, and what keeps tripping them up?”

The Four Quadrants Aren’t a Template — They’re Four Different Angles on the Same User

A lot of teams treat empathy maps like a template to fill. The value isn’t in filling boxes. It’s in noticing patterns, and especially contradictions, that show up when you look at a user from more than one angle. Here’s what each quadrant is actually doing, and how to tell when one of them is lying to you.

Says captures what the user explicitly says — in interviews, support calls, usability tests, casual feedback. “Why is this taking so long?” “I just need it approved.” “I don’t want another tool to learn.” This quadrant is useful because it captures expectations and language. It tells you how users frame the problem in their own words, which is often different from how your team frames it.

Thinks is where insight often hides. People don’t always say what they think, especially when they’re being polite, unsure, or worried about sounding incompetent. “If I mess this up, it’ll come back to me.” “I don’t trust this system. I should keep a backup.” “I’m already late. Please don’t make this harder.” The Thinks quadrant is not a license to invent a story. It’s an informed inference based on evidence — tone, hesitation, repeated questions, what they avoid, what they double-check, what they quietly work around.

Does is what the user actually does. Observable behavior, steps, workarounds. It’s often the most honest quadrant because it doesn’t rely on memory or good intentions. Refreshes the page repeatedly. Screenshots an error and messages someone. Uses Excel “just in case.” Abandons the flow and tries again later. If your empathy map is thin here, it usually means you haven’t watched enough real usage. You’ve listened. You haven’t observed.

Feels captures the emotional weather of the experience. It doesn’t need to be dramatic to matter. Quiet emotions drive behavior all the time, especially around trust. Anxious when the status is unclear. Irritated when asked to repeat steps. Relieved when there’s a clear next action. Confident when actions lead to immediate confirmation.

One quick rule that helps: the most useful insights live in the gaps. If a user says “it’s fine,” does three workarounds, and feels anxious, you’ve learned something bigger than a polite comment.

A Useful Empathy Mapping Session Looks Nothing Like a Sticky-Note Ritual

Empathy mapping works best as a short, structured workshop — usually 60 to 90 minutes — after you’ve collected user inputs. You don’t need an elaborate setup. You need evidence, and you need a focused scenario.

Start by picking one user type and one moment. Not “our target audience.” Something specific, like “first-time admin setting permissions,” or “ops manager approving invoices before end of day.” Empathy maps get powerful when they’re precise.

Then bring real inputs into the room. Interview quotes, usability notes, support tickets, onboarding call recordings, analytics patterns, sales call themes. Anything that smells like reality. If you don’t have evidence yet, you can create a hypothesis map, but label it honestly as a starting point and treat it as something you will validate.

From there, the flow is simple. Begin with silent capture — everyone adds notes to the four quadrants without debating them. This reduces groupthink. People notice different things, and you want the room to surface those differences before you converge. Next, cluster what’s on the map into themes. You’ll start seeing repeated patterns — unclear status, fear of blame, doesn’t trust automation, needs faster confirmation, switches tools mid-task. This is where the map stops being a collage and starts becoming a story.

Then do the part that separates a useful workshop from a sticky-note ritual. Call out contradictions and gaps. Where do “Says” and “Does” conflict? What does the user seem to need but never states directly? Which quadrant is oddly empty, and therefore signals you need more research?

Finally, end with decisions. Not “great insights.” Decisions. A couple of design principles, the top risks to validate, and one or two scenarios to prototype and test next. If nothing changes after an empathy mapping session, you didn’t really do empathy mapping. You did a well-intentioned exercise and then went back to building from assumptions.

Dennis Has Twelve Minutes Between Meetings and the Empathy Map Tells You Why That Matters

Let’s say you’re building an internal tool where an operations manager approves vendor invoices. On paper, it’s a straightforward workflow. In real life, it’s a transfer point with tight timing, pressure, and consequences if something goes wrong.

Scenario: Dennis has 12 minutes between meetings. He needs to approve three invoices before the finance cut-off. He’s switching between email, your platform, and a shared drive. He isn’t here to explore. He’s here to get through it.

Dennis Says: “Where do I see what’s pending?” “Why do I need to upload the same thing again?” “Just tell me what’s missing.”

Dennis Thinks: “If this goes wrong, I’ll be the one answering for it.” “I don’t trust that ‘submitted’ means ‘received’.” “I can’t afford to get stuck here.”

Dennis Does: Opens the invoice in email, checks numbers in a spreadsheet, returns to the platform. Downloads a copy “for records.” Messages finance to confirm receipt. Approves one invoice, then pauses because the status didn’t change.

Dennis Feels: Anxious when status is unclear. Irritated when the system asks for the same input twice. Relieved when there’s a clear checklist. Confident when actions lead to immediate confirmation.

Notice what this gives you. You can stop debating features in the abstract and start asking better questions. What would make the status unmistakable? Where can we reduce tool switching? What’s the smallest change that increases trust? Which step causes the “missed connection” moment under time pressure? That’s the difference between building something that looks good in a demo and building something that holds up on a Tuesday afternoon.

Empathy Mapping Fits Right Before You Lock In Direction — and Right After You Realize You Already Did

Empathy mapping is most useful right before you lock in direction. When you’re defining the problem, shaping requirements, deciding what to prototype, or trying to align stakeholders who all have different opinions about what “users want,” it’s a fast way to get the room back to evidence.

It’s also a great reset button when a product has already started moving and the team feels stuck. If the work has become a debate about preferences, empathy mapping brings everyone back to context. What the user is dealing with. What they are trying to accomplish. What they will tolerate, and what they will not.

How an Empathy Map Differs From the Other Discovery Artifacts You’ll Be Asked to Produce

Artifact	What it tells you	When it earns its place
Persona	Who the user is across many moments	You’re defining the segment you’re designing for
Empathy Map	What it’s like to be that user in one specific high-pressure moment	You’re designing a flow that has to work under pressure
Journey Map	What path the user takes end-to-end across time	You’re surfacing friction across the full experience
Jobs to Be Done	What outcome the user is hiring the product to achieve	You’re deciding whether to build a feature at all

The four are complementary, not interchangeable. The empathy map’s specific job is to ground one moment in evidence rather than imagination. Trying to make a persona do the empathy map’s work produces a generic profile that designs nothing well; trying to make an empathy map cover the whole journey produces vague truths nobody can act on.

The Four Failure Modes That Quietly Turn the Map Into Theater

The first mistake is making it up. If the map is mostly your team’s imagination, it becomes a mirror, not a user tool. It might still feel convincing, but it won’t protect you from building the wrong thing.

The second is trying to cover everyone. One map for all users quickly turns into vague truths like “users want convenience.” Keep it narrow: one persona, one moment.

The third is treating the map as final. People change. Tools change. Workflows change. A strong empathy map is something you revisit after new research, not something you file away after the workshop.

And the fourth — the most common — is you don’t translate it into action. The map is not the deliverable. The clarity is. If the empathy map doesn’t lead to decisions about what to validate, what to simplify, what to prototype, it won’t earn its place in the process.

Two Numbers That Should Change How You Think About Skipping This

The Standish Group’s CHAOS Report (September 2020) found that only 31% of software projects meet success criteria on time, on budget, and with required features — and inadequate requirements specification ranks among the top three causes of failure. That failure starts precisely where teams design for the ideal user instead of the real one running your flow on a Tuesday afternoon with twelve minutes to spare.

Aggarwal et al. in the Princeton GEO paper (arXiv:2311.09735, KDD 2024, November 2023) showed that documents grounded in named statistics, citations, and structured evidence outperform narrative-only documents by over 40% at downstream decisions. The same principle applies to your empathy map: one filled with real quotes, observed behavior, and named scenarios will outperform one filled with what your team imagines users probably feel — every time the team has to defend a design decision later.

Frequently Asked Questions

What is empathy mapping in product discovery? Empathy mapping is a product-discovery technique that captures what a specific user experiences in a specific moment across four quadrants — Says, Thinks, Does, Feels. It complements a persona by adding moment-level context rather than stable demographic description, and it surfaces the contradictions between what users claim and what they actually do.

How is an empathy map different from a user persona? A persona describes who the user is across many moments — role, demographics, goals, technical comfort — and is stable. An empathy map describes what it is like to be that user in one specific high-pressure moment, and is tied to that scenario. The persona answers who you’re designing for; the empathy map answers how to design the flow that has to work when that person is under time pressure.

How long does an empathy mapping session take? A well-run empathy mapping session takes 60 to 90 minutes after the team has already collected evidence. The session itself does not include the research phase. Sessions that try to gather evidence inside the workshop produce maps that reflect the team’s imagination rather than the user’s reality.

What evidence does an empathy mapping session need to be useful? Real inputs in the room: interview quotes, support tickets showing where users get stuck, usability test recordings, onboarding call notes, analytics traces showing abandonment patterns, and sales call themes about prospect objections. Without these, the team fills quadrants with assumption rather than evidence, which produces a mirror rather than a user tool.

When does empathy mapping not earn its place? It doesn’t earn its place when the user moment is already well-understood across the team and observable behavior is uncontested — typically for minor iterative improvements on flows that have been live for a year and accumulated rich behavioral data. For new flows, new user segments, regulated workflows, or moments under high time pressure, skipping the empathy map almost always produces a beautifully designed flow that breaks the first time a real user runs it.

Can you build an empathy map without user interviews? You can build a hypothesis empathy map without interviews — but label it honestly as a starting point, a set of assumptions to validate, not a finished artifact. Treating the hypothesis map as final is the failure mode that turns empathy mapping into team imagination. The map only earns its weight once real quotes, observation notes, and support evidence have filled the four quadrants.

A Closing Thought

Public transport isn’t frustrating because people are bad at commuting. It’s frustrating when the system quietly assumes ideal conditions — perfect timing, perfect signage, zero pressure.

Products become frustrating for the same reason.

Empathy mapping helps you design for real conditions. People are busy, accountable, distracted, and often trying to complete your task in between ten other tasks. If you can build for that user, you don’t just improve UX. You build trust. And trust is what keeps people coming back.

Sources

• Standish Group International. CHAOS Report 2020: Beyond Infinity. September 2020.
• Aggarwal, Pranjal et al. “GEO: Generative Engine Optimization.” Princeton University / Georgia Tech / IIT Delhi, arXiv:2311.09735, KDD 2024, November 2023.
• Gray, Dave; Brown, Sunni; Macanufo, James. Gamestorming: A Playbook for Innovators, Rulebreakers, and Changemakers. O’Reilly Media, July 2010.
• Nielsen Norman Group. “Empathy Mapping: The First Step in Design Thinking.” https://www.nngroup.com/articles/empathy-mapping/

---

Run empathy mapping inside a structured Product Discovery Workshop. Inheaden’s discovery process produces the empathy map, persona, journey map, and Jobs to Be Done as four connected artifacts — not isolated exercises

Book a 30-minute call to figure out whether your team needs a fresh empathy map for a new flow or a reset session for a stuck one: → https://inheaden.io/contact