DEV Community

kanta13jp1
kanta13jp1

Posted on

User Interviews for Indie Developers: 5 People Reveal the Real Problem

#ai #automation #indiedev #buildinpublic

User Interviews for Indie Developers: 5 People Reveal the Real Problem

The best way to avoid "built it, no one uses it" is user interviews. 5 people is enough to see patterns. Here's the method I use.

Why Interviews, Not Surveys 

Surveys (wrong): "Is this useful? → Yes/No" teaches you nothing
Interviews (right): observing behavior reveals the real problem
Enter fullscreen mode Exit fullscreen mode

Users can't accurately say what they want. But they can describe when they struggle. Interviews are behavioral observation, not opinion collection.

The Rule of 5

From Jakob Nielsen's research: 5 user interviews reveal 85% of problems.

1 person:  unique problems (noise)
3 people:  patterns start emerging
5 people:  major problems are covered
10+ people: diminishing returns → low ROI
Enter fullscreen mode Exit fullscreen mode

No large-scale research needed. Start by talking to 5 people.

Question Design: Ask About Past Behavior 

❌ Wrong: "Would you use this feature?"
   → Answer: "I think I would" (hypothetical, not real)

✅ Right: "When was the last time you struggled with [problem]? What did you do?"
   → Answer: actual behavior pattern
Enter fullscreen mode Exit fullscreen mode

5 core questions:

  1. "Tell me about [problem area]." (opening)
  2. "When was the last time this was a problem for you?" (make it concrete)
  3. "What did you do then?" (observe behavior)
  4. "What was the hardest part?" (find the pain)
  5. "How do you handle it now?" (find the workaround)

How to Run the Interview 

Duration: 30-45 minutes
Format: Zoom / in person (recording or notes required)
Setup: 1-on-1 (group interviews skew responses)
Enter fullscreen mode Exit fullscreen mode

Don't do this:

❌ Pitch your product (it becomes sales, not research)
❌ Over-agree "right, exactly!" (introduces bias)
❌ "But wouldn't feature X solve this?" (pushing your answer)
Enter fullscreen mode Exit fullscreen mode

Do this:

✅ "Can you tell me more about that?" for depth
✅ Don't fear silence (let them think)
✅ Ask "why?" three times to reach root cause
Enter fullscreen mode Exit fullscreen mode

Note-Taking 

Good notes: exact words used ("I always forget to...", "when X happens I tend to...")
Bad notes: your interpretation ("bad UX", "missing features")
Enter fullscreen mode Exit fullscreen mode

Organize notes within 24 hours while memory is fresh.

Finding Patterns (Analysis Across 5 Interviews) 

Step 1: Write each "struggle" on a sticky note (one per note)
Step 2: Group similar notes (affinity diagram)
Step 3: Treat anything 3+ people mentioned as a "problem"
Step 4: Hold 1-2 person mentions as tentative (not enough evidence)
Enter fullscreen mode Exit fullscreen mode

5 people × 3-5 struggles = up to 25 notes. Group them and you'll get 3-7 themes.

Example from My Project

After launching the AI University (280-company AI tool database), I interviewed 5 users:

What they said:
  "I don't know which one to choose"
  "I want a comparison table"
  "I want to know what it's actually like to use"

What I heard:
  Feature list → comparison axes → real experience
  Needed to shift from "database" to "tool for deciding"
Enter fullscreen mode Exit fullscreen mode

This insight drove the expansion of the /competitors comparison page.

How to Find Interviewees 

For indie developers:
  1. Comment section of your dev.to / Qiita articles
  2. X (Twitter): "Anyone struggling with [problem]?"
  3. Reddit / Discord communities around the topic
  4. Email existing users (high open rate)

Incentive: $10-20 Amazon gift card or free plan upgrade
Enter fullscreen mode Exit fullscreen mode

Don't fear rejection. If you reach out to 10 people and 5 talk to you, that's enough.

Summary

User interviews aren't market research — they're empathy building. The problem lives in the user's life. Listen before you talk about your product. 5 conversations will show you what to build.

Top comments (0)