Empathy Interviews: Questions, Structure, and How to Run Them

An empathy interview is a structured conversation designed to understand someone's lived experience — their feelings, frustrations, values, and mental models — not just their observable behavior. Rooted in human-centered design and design thinking, empathy interviews reveal the emotional layer beneath what users do, helping teams build products and services that feel intuitively right.

How Empathy Interviews Differ from Standard User Interviews

Traditional user interviews focus on tasks: what do you do, how often, what happens when something breaks? Empathy interviews go deeper: how do you feel about it? What matters to you? What was the hardest part emotionally?

The distinction sounds subtle but produces radically different insights. A task-focused interview might reveal that users struggle with your onboarding flow. An empathy interview reveals why: they feel embarrassed when they do not understand something, and your current design puts that failure on public display. That emotional insight changes everything about how you redesign.

Empathy interviews became a cornerstone of the Stanford d.school design thinking methodology and have since spread through product design, education, healthcare, and social innovation. They are especially powerful at the start of a project — before any solution exists — when the goal is to fully understand the human problem you are trying to solve.

According to IDEO, teams that run empathy research before beginning design work are significantly more likely to identify the right problem to solve. The investment in understanding before building pays off through reduced iteration cycles and products that resonate immediately.

When to Use Empathy Interviews

Empathy interviews are the right tool when:

• You are starting a new project and need to understand the human context before defining solutions
• Your product has strong usability metrics but low emotional engagement
• Users are churning and you cannot identify a functional reason why
• You are entering a new market or designing for a new user segment
• Your team is building on assumptions that have never been tested with real humans

They are less useful when you need to measure how widespread a problem is (use a survey) or evaluate whether a specific design works (use usability testing). Empathy interviews are for building understanding, not measuring or evaluating.

Empathy Interview Questions by Phase

Opening: Build Rapport and Get Context

The first 5–10 minutes are critical. The goal is to help the participant feel safe sharing honestly. Start broad and personal.

• "Tell me about yourself and what you work on."
• "Walk me through your typical [day/week] around [topic]."
• "What is the most important thing to you about [domain]?"
• "How long have you been dealing with [situation]? What has changed over that time?"

Experience Questions: Understand What Happens

Move into the participant's actual experience. Stay in the past — hypotheticals produce unreliable data. Real stories are the raw material of empathy.

• "Tell me about the last time you [relevant activity]."
• "Walk me through that situation from beginning to end."
• "Who else was involved when that happened?"
• "What did you try first? What happened when that did not work?"
• "Tell me about a time when [activity] went really well."
• "Tell me about a time when it was especially difficult or frustrating."

Feeling Questions: Access the Emotional Layer

This is where empathy interviews diverge from standard user research. These questions access the data that behavioral interviews miss.

• "How did that make you feel?"
• "What were you thinking at that point?"
• "Was there a moment when you felt stuck or unsure?"
• "What was the hardest part emotionally?"
• "How did you feel when it was finally resolved?"

Value Questions: Surface What Matters

The deepest level of empathy research reveals the underlying values and priorities that drive behavior. The "5 Whys" technique works well here.

• "Why is that important to you?"
• "What would success look like in that situation?"
• "What would be the worst possible outcome?"
• "If you could change one thing about [experience], what would it be and why?"
• "What does [topic] represent for you beyond the practical side?"

Closing: Catch What You Missed

• "Is there anything about your experience with [topic] that we have not talked about that you think I should understand?"
• "If you were advising someone designing [product/service] for people like you, what would you tell them?"

Step-by-Step: Running an Empathy Interview

1. Set an empathy-first intention before you begin Consciously shift your goal: you are not validating a hypothesis. You are trying to understand this person's world. Put your assumptions in a drawer. Your job in this conversation is to listen, not to confirm.

2. Open with their story, not your product Do not mention your product, concept, or design for the first 15–20 minutes. Let the participant narrate their experience in their own words. Context you gather before introducing your lens is more valuable than anything collected after.

3. Follow feelings, not features When someone describes something that happened, ask "How did that feel?" and "What were you thinking at that point?" These questions consistently surface emotional data that behavioral questions miss.

4. Ask about the extremes "Tell me about a time when [activity] went really well" and "Tell me about a time when it was especially hard." Extreme experiences reveal values and pain points more clearly than descriptions of typical experiences.

5. Use the 5 Whys When someone describes a frustration or preference, ask "Why does that matter to you?" Then ask why that matters. Five levels of why typically surface the root value or fear driving the behavior. Most behavioral researchers stop at the first why — empathy interviewers keep going.

6. Listen more than you talk A healthy ratio: 20% talking, 80% listening. Resist the urge to fill silence. Comfortable pauses often precede the most honest answers. When a participant goes quiet, wait 5 full seconds before speaking.

7. Do not problem-solve during the interview If a participant describes a frustration, your job is not to reassure them or suggest solutions. Acknowledge what you heard ("That sounds really frustrating") and ask a follow-up question. Problem-solving mode shuts down vulnerability.

8. Close with reflection Summarize what you heard in 2–3 sentences and ask: "Does that capture your experience accurately? Is there anything important I missed?" Giving participants the chance to correct your summary often surfaces the most honest data of the session.

Key Things to Know

• Not a usability test: Empathy interviews do not evaluate your product. They explore the human context your product lives in. Introduce your product only at the end, if at all, and frame it as "I want to show you something and get your reaction" — not "we built this to solve your problem."

• No artifacts during the interview: Do not show wireframes, prototypes, or screenshots during the main interview. The presence of a solution biases how people describe problems. They start evaluating what you built instead of describing what they experience.

• Emotional signals are data: Note not just what was said, but tone, hesitation, laughter, and discomfort. These signals tell you where the emotional weight is. A participant who laughs nervously when describing a process is telling you something important.

• Pair up when possible: Run empathy interviews with one person asking questions and one person taking notes. The questioner can focus entirely on the emotional thread of the conversation without split attention.

Synthesizing Empathy Interview Data

Raw empathy interview data is rich but unwieldy. The standard synthesis tool is the empathy map — a four-quadrant framework capturing what users Say, Think, Feel, and Do.

How to build an empathy map:

1. After each interview, extract quotes and observations into sticky notes
2. Organize them into the four quadrants: Says (direct quotes), Thinks (what they seem to believe), Feels (emotional signals), Does (observed behaviors)
3. After 5+ interviews, look for patterns across the maps
4. Identify tensions: where do what people Say and what they Feel conflict? These tensions reveal unmet needs.

From your empathy maps, distill a "Point of View" statement: "[User type] needs [need] because [insight]." This reframes your design challenge around a human truth instead of a feature request.

How AI Scales Empathy Research

One limitation of traditional empathy interviews is that they require significant time to moderate well — and a skilled empathy interviewer needs experience to follow emotional threads naturally. This makes depth-first empathy research expensive to scale.

AI-powered platforms like Koji address this constraint. The AI interviewer is trained to recognize emotionally significant statements and probe deeper when participants express frustration, uncertainty, or strong feeling. It asks follow-up questions like "Can you tell me more about how that felt?" and "What was going through your mind at that point?" consistently — across every session, regardless of volume.

This means a team can run 20 empathy-style interviews in a week rather than 3–4. The emotional depth that previously required a skilled human moderator is built into the AI's probing strategy. Researchers can focus entirely on synthesis and insight generation rather than facilitation logistics.

Tips & Best Practices

• Record all sessions with permission and read transcripts carefully. Emotional nuance is lost in summary notes — the specific words people use often carry more meaning than any paraphrase.

• Share raw stories from empathy interviews with your team. Data summaries strip out the emotional texture that drives real empathy. When your engineer hears a customer describe feeling embarrassed, that lands differently than reading "users find the error states confusing."

• Look for what people care about versus what they say they want. These are often different. A participant might say they want a faster checkout, but an empathy interview reveals they actually want to feel confident they made the right decision.

• Revisit empathy research every 6–12 months. User feelings and values shift as your product matures and as the market changes. What was true about your users two years ago may not be true today.

Frequently Asked Questions

Q: How is an empathy interview different from a regular user interview? A: A regular user interview focuses on behavior and tasks — what you do, how you do it, what goes wrong. An empathy interview focuses on feelings, values, and lived experience — the emotional context behind behavior. Both are valuable; empathy interviews go deeper.

Q: How long should an empathy interview last? A: 45–60 minutes is typical. Emotional depth takes time to develop. The first 20 minutes establish rapport and context; the richest insights usually surface in the second half of the session.

Q: Can empathy interviews be done remotely? A: Yes — video calls work well. Pay close attention to tone of voice and facial expressions. AI-powered interview platforms can also surface emotional depth through follow-up probing on emotionally charged language, making remote empathy research scalable.

Q: How many empathy interviews do I need? A: Five to ten interviews typically reveal primary themes. With AI-powered platforms, running more interviews requires no additional moderator time — there is no reason to stop at five if you want higher confidence.

Q: What do I do with empathy interview data after synthesis? A: Build an empathy map, then distill a Point of View statement: "[User type] needs [need] because [insight]." Use this to reframe your design challenge around a human truth before any ideation begins.