User Research for PMs
(2026 Edition)
The 6 research methods every PM should know, when to use each, how many users you actually need, and how to turn research into product decisions — not reports.
Practice User Research Scenarios — Free →User Interviews (1:1)
Understanding why users do what they do. Uncovering pain points. Learning about context of use.
Sample size
5–8 per persona. Patterns emerge after 5. Returns diminish after 10.
⚠️ Avoid: Asking 'would you use X?' — users are bad at predicting their future behaviour. Ask about past behaviour instead.
✅ Good for
Early-stage discovery, post-launch learning, persona development
❌ Bad for
Feature prioritisation (use quant), UX validation (use usability tests)
Surveys
Quantifying something you already qualitatively understand. Measuring perceptions at scale.
Sample size
Minimum 100 responses for meaningful inference. 300+ for segment cuts.
⚠️ Avoid: Asking leading questions. 10+ questions (completion rate drops). Free-text when multi-choice would work.
✅ Good for
NPS, CSAT, feature importance ranking, persona attribute validation
❌ Bad for
Understanding WHY users do things (surveys reveal what, not why)
Usability Testing
Validating that users can actually use a design. Finding UX friction.
Sample size
5 users catches ~85% of usability issues (Nielsen's rule).
⚠️ Avoid: Leading users through tasks. Asking them what they're thinking instead of letting them think aloud naturally.
✅ Good for
New flows before launch, redesign validation, complex feature testing
❌ Bad for
Understanding demand for a feature (they'll comply even if they wouldn't use it)
Diary Studies
Understanding behaviour over time, in natural context, without researcher present.
Sample size
10–15 participants for 1–2 weeks.
⚠️ Avoid: Asking for too much effort per entry. Forgetting to send reminders.
✅ Good for
Habits, long-feedback-loop products, mobile behaviour, multi-touchpoint journeys
❌ Bad for
Quick insights, large-scale trends, features you need to ship next sprint
Session Recordings & Heatmaps
Passive observation at scale. Spotting unexpected user behaviour.
Sample size
Review 20–50 sessions per hypothesis. Use heatmaps at aggregate level.
⚠️ Avoid: Relying on it alone — session recordings show behaviour without context. Pair with interviews.
✅ Good for
Funnel diagnosis, unexpected error behaviour, spotting design dead zones
❌ Bad for
Understanding user goals or intent (you see what, not why)
Card Sorting / Tree Testing
Information architecture decisions. What should be grouped with what?
Sample size
15–30 participants for meaningful clustering.
⚠️ Avoid: Using closed sorts when you don't know user mental models yet. Use open sorts first.
✅ Good for
Menu redesigns, navigation structure, category taxonomy
❌ Bad for
Everything else — it's a specialised method for IA only
6 User Interview Questions That Work
Ask these in order. Let silence breathe. Never lead.
“Walk me through the last time you [did X]. What happened?”
Gets a specific story, not a generalisation
“What was the hardest part of that?”
Surfaces the real pain point
“What did you try to do about it?”
Reveals current workarounds (often the real competitor)
“How did that work out?”
Gets context on why existing solutions aren't enough
“What happened next?”
Keeps the story going; reveals downstream consequences
“Is there anything else you think I should know?”
Catches unexpected insights and closes respectfully
FAQ
How much user research should a PM do?
At least 1 user interview per week per product area is the minimum sustainable cadence (Teresa Torres' continuous discovery recommendation). This is ~2 hours per week and dramatically improves product quality. PMs who do zero research outside of what UX researchers provide ship products that miss user nuance. PMs who do research continuously build intuition that no amount of data can replicate.
Do PMs need to run user research themselves or rely on UX researchers?
Both. Work with UX researchers for complex studies (diary studies, segmentation, ethnography). Run lightweight research yourself for ongoing learning (weekly user calls, informal usability tests, PM-run interviews for feature validation). PMs who outsource ALL research to UXR end up with slower feedback loops and weaker user intuition.
How do you recruit users for interviews in India?
For existing users: in-product invites (best conversion, most representative), email outreach to segments, or post-support-ticket follow-ups. For non-users or new segments: UserTesting.com, Respondent.io (for paid panels), LinkedIn outreach, subreddits/communities where your target users hang out, or your personal network with a clear ask. Offer ₹500–₹2000 incentive depending on interview length.
Build User Empathy in 2 Minutes a Day
Daily PM scenarios that sharpen user intuition and research-based decision making.
Start Free Trial →