User Research — Design for Real People

What you'll learn

• Why "you are not the user"
• Research methods: interviews, surveys, usability testing
• Personas and user journeys
• How research prevents building the wrong thing

You are not the user 🧑‍🤝‍🧑

The most expensive mistake in product design is building something beautiful that nobody needs — or that real people can't figure out. The cause is almost always the same trap: the designer assumes everyone thinks like them. But you are not the user. You know where every button is because you made it; a first-timer is lost. User research is how you escape your own head and design for reality.

Key research methods:

• Interviews — talk to real (potential) users one-on-one. Ask about their problems and habits, not "would you like this feature?" (people are bad at predicting their own behavior). Listen far more than you talk.
• Surveys — questions sent to many people. Great for quantitative signal ("70% abandon at the payment step") but shallow — pair with interviews for the why.
• Usability testing — the gold standard. Put your design (even a rough prototype) in front of a real person, give them a task ("try to buy this item"), and watch silently as they struggle or succeed. You'll spot problems you were completely blind to. Amazingly, testing with just ~5 people uncovers most major issues.

Tools to organize what you learn:

• Personas — a short profile of a typical user ("Sara, 28, busy nurse, shops on her phone between shifts, low patience for slow apps"). It keeps the team designing for a specific human, not a vague "everyone."
• User journey map — the step-by-step path a user takes to reach a goal, marking where they feel frustrated or delighted. It reveals the painful moments to fix.

The mindset shift: research replaces "I think..." with "I observed...". Opinions are cheap and often wrong; evidence from real users is what builds products people actually adopt. Even a little research — testing a prototype with five friends — dramatically improves your design and saves you from confidently building the wrong thing. This is also what separates a designer from a decorator: designing the right solution, verified with real people.

Quick, cheap tests you can run this week 🧪

Not all research requires recruiting strangers and scheduling calls. A few lightweight tests fit into an afternoon and still catch major problems:

• The 5-second test. Show someone your design for exactly 5 seconds, then hide it and ask: "What was this? What was it for? What would you click first?" If they can't answer, your hierarchy and messaging aren't communicating fast enough — real visitors also glance and judge in seconds, so this simulates that ruthless first impression cheaply.
• Card sorting. Write every piece of content or feature on its own index card (or sticky note), then ask a few people to group the cards into categories that make sense to them, and name each group. Do this with 5 people and patterns emerge — this is how you build navigation/IA (from the wireframing lesson) that matches how real users think, instead of how your team happens to be organized internally.
• The "first click" test. Show a screen and a task ("find where to update your password"), and just watch where they click first — nothing else. Research shows that if the first click is right, people usually complete the task; if it's wrong, they often struggle. One click tells you a lot.
• Guerrilla testing. No lab required — sit in a coffee shop or ask 3 coworkers "can I get 5 minutes of your time?" and run a mini usability test on your phone or laptop. Imperfect, but some real-user feedback beats zero, and it costs nothing but asking.

None of these require a research budget or a dedicated researcher — they require only the willingness to show your work to a real person before you're finished, which is the actual hard part.

Try it yourself: run a 5-second test today 🔍

Pick any screen or page you've designed (or even a screenshot of an app you admire) and do this right now:

1. Find a friend, family member, or coworker willing to give you 60 seconds.
2. Show them the design for exactly 5 seconds (count it, or use a phone timer), then hide the screen.
3. Ask: "What is this? What's it for? What would you click or tap first?"
4. Write down their answer exactly, without correcting or explaining anything, even if they got it "wrong."

If their answer matches what you intended, your hierarchy and messaging are working. If it doesn't, you've just learned — for the cost of one minute — something no amount of staring at your own design would have revealed, because you already know what it's supposed to say.

Real-world case study: the checkout field that cost millions 🔍

One of the most famous stories in UX research concerns an e-commerce checkout form that required every visitor to create an account before buying — an extra "Register" step with a password field, standing between a shopper and their purchase. Analytics and user research consistently pointed to the same problem: a huge share of shoppers abandoned their cart at exactly that step, frustrated at being forced to create yet another password just to buy something.

The team running the research proposed a small, almost boring change: replace the "Register" button with a "Continue" button, and simply ask for an email address at checkout — turning account creation into an optional step offered only after the purchase was complete, framed around removing friction rather than demanding commitment upfront. The change was tested and rolled out, and the resulting jump in completed purchases was large enough that the story is still cited in UX and conversion-research circles as a canonical example of how research-driven insight into a single, unglamorous field can move revenue enormously.

The lesson isn't "always remove registration" — it's that research turned a vague, arguable business opinion ("users should create accounts") into an observed fact ("this exact step is where real people quit"), and that a famously small, unglamorous change, backed by evidence instead of assumption, outperformed a much flashier redesign would have.

Summary

• You are not the user — research replaces assumptions with evidence
• Methods: interviews (why), surveys (how many), usability testing (watch them struggle)
• Personas and journey maps keep design focused on real people and pain points
• Cheap tests you can run this week: 5-second test, card sorting, first-click test, guerrilla testing
• The famous checkout case study: one research-driven field change moved revenue enormously
• Even testing with ~5 people uncovers most major problems

Next: Wireframing & IA — turning research insights into the skeleton of a design.