PM Empathy Interviews Guide
(2026 Edition)
6 good questions, 5 bad questions to avoid, the 5-user rule explained, and a 6-step synthesis process to turn interviews into product decisions.
Build User Empathy Daily — Free →6 Questions That Work
✅ “Walk me through the last time you [did X].”
Gets a specific story, not generalisations
✅ “What was the hardest part of that?”
Surfaces the real pain, not perceived pain
✅ “What did you try to do about it?”
Reveals current workarounds — the real competitor
✅ “What happened next?”
Keeps the story going; reveals downstream consequences
✅ “Tell me more about that.”
Best follow-up question in interviewing — opens people up
✅ “What would a perfect solution look like?”
Reveals aspirational needs, not feature requests
5 Questions to Avoid
❌ “Would you use a feature that...?”
People overstate future intent by 20–40%
❌ “Don't you find it annoying when...?”
Leading question — tells the user what to say
❌ “How much would you pay for...?”
Stated willingness-to-pay correlates poorly with actual behaviour
❌ “Do you like our product?”
People are polite. They'll say yes to strangers asking.
❌ “What features should we build?”
Users are bad at designing products. Ask about problems, not solutions.
The 5-User Rule Explained
5 users catches ~85% of usability issues (Nielsen)
After 5, returns diminish quickly for a single problem area
If you need to segment (power users vs new users), 5 per segment
For broad discovery (new market), 10–15 is worth it
For diary studies or long-form, smaller samples over time often beat more users
6-Step Synthesis Process
Transcribe within 24 hours — memory fades fast, nuance gets lost
Highlight user quotes that made you stop and think
Tag each insight: pain point, workaround, delight, unmet need
Look for themes across 3+ interviews — one user's frustration is anecdote; three is signal
Translate themes into product hypotheses: 'If we addressed X, then Y would change'
Share with your team — insights that live in your head don't change the product
FAQ
How often should PMs run empathy interviews?
At least 1 per week, as a sustainable baseline. Teresa Torres recommends this as the minimum for continuous discovery. PMs who do this consistently build instinct that analytics-only PMs can't replicate. 20 minutes with one user each week = 17 hours/year of direct user exposure, which compounds into pattern recognition over multiple years.
How do you find users to interview?
For existing products: in-product opt-ins (best signal), email outreach to specific user segments, post-support-ticket follow-ups. For new/no users: your personal network, LinkedIn outreach, subreddits/communities where target users hang out, paid panels (UserInterviews.com). Offer ₹500–₹2000 incentive depending on length. Aim for 70%+ show-up rate; below that, improve your pre-interview reminders.
What's the biggest mistake PMs make in empathy interviews?
Talking too much. Good interviews are ~80% user talking, 20% PM. PMs who over-explain, lead with hypotheses, or interrupt miss insight. The discipline is hard — the instinct is to fill silence. Best rule: after the user pauses, count to 3 before speaking. They'll often continue with the most important thing.
Build User Empathy Daily
Daily PM scenarios that train the discipline of empathy before solution.
Start Free Trial →