User Research

Also known as: Customer DiscoveryUX ResearchQualitative Feedback

💡The Concept

User Research is the systematic investigation of your target audience's behaviors, needs, and motivations. It exists to invalidate your assumptions before you spend expensive engineering hours building a product nobody actually wants. True research focuses on what users *do*, not what they *say* they will do.

⚠️The Trap

The most dangerous trap is asking leading, hypothetical questions like 'Would you pay $10/month for this feature?' Humans are terrible at predicting their future behavior and want to please the interviewer. They will say 'yes' to your face and then never open their wallets when the product launches.

🎯The Action

Conduct 'Jobs-to-be-Done' interviews focused entirely on the past. Instead of asking what they want you to build, ask: 'Walk me step-by-step through the last time you tried to solve this problem. What exactly did you do? What tool did you use? How much time did it take?' Pain lives in the past, not the future.

🌍Real-World Example

When building the original iPhone, Apple didn't ask Nokia flip-phone users 'What features do you want on a phone?' Users would have just asked for a better physical keyboard (which BlackBerry delivered). Apple observed the extreme friction of carrying an iPod, a phone, and a separate internet communicator, solving an underlying behavioral pain point the market didn't know how to articulate.

⚡Pro Tips

Record interviews (with permission) and transcribe them. The specific words users use to describe their pain should become the headline on your landing page.

Talk to extreme users. Average users accept the status quo. Power users who have hacked together 5 spreadsheets to solve a problem will explicitly show you exactly what to build.

Stop talking. The longer the uncomfortable silence after a question, the deeper the insight the user is about to give you.

🚫Common Myths

✗Myth: “Henry Ford said 'If I had asked people what they wanted, they would have said faster horses.'”

✓Reality: Even if true, a 'faster horse' clearly indicates the user's core Job-to-be-Done is 'getting to point B quickly.' The car solves the exact same problem, just far better.

✗Myth: “You need a massive sample size to get valid qualitative data.”

✓Reality: According to the Nielsen Norman Group, testing 5 users generally uncovers 85% of core usability problems. If 5 consecutive people stumble on the same poorly labeled button, testing 50 more people is a waste of time.

📈Industry Benchmarks

Usability Problem Discovery Rate

Nielsen Norman Group rule for qualitative usability testing

Optimal ROI

5 Users (Discovers ~85% of issues)

Diminishing Returns

15 Users (Discovers ~95% of issues)

Wasted Resources

> 25 Users (Redundant findings)

Source: NN/g: 'Why You Only Need to Test with 5 Users'

🎮Decision Scenario: The Misleading Survey

Your team launched a quick survey asking 10,000 active users of your fitness app: 'Which feature do you want next: Calorie Tracking or Wearable Integration?' 80% voted for Calorie Tracking.

Engineering Time Required

3 Months

Survey Result

80% choose Calorie Tracking

Current Retention

Declining

Decision 1

Your Lead Engineer warns that building a massive food database for calorie tracking will take the entire quarter, delaying several necessary bug fixes. The CEO wants to follow the survey data.

Commit all resources to building the Calorie Tracker. The data is overwhelmingly clear, and ignoring 8,000 users is dangerous.Click to reveal →

You launch the tracker 3 months later. Only 5% of users actually log their food because it's tedious. You missed the critical bugs, and overall retention crashes. You fell for the 'Aspirational Goal' trap—people voted for who they want to be, not what they will actually do.

Delay the build. Conduct 10 in-depth qualitative interviews with users who voted for the Calorie Tracker to understand their underlying motivation.Click to reveal →

The interviews reveal users actually just want a simple way to know if they 'burned a lot' during their workout to feel good about eating a large dinner. You build a simple 'Intensity Score' in two weeks, which achieves the same psychological goal with massive adoption and saves your engineering resources.

🧪