UX Research — How I Actually Do It (Without Corporate Fluff)
UX research is frequently overcomplicated with academic jargon, endless personas, and bloated slide decks that nobody reads. True user research is simple: observe real humans attempting to accomplish a real task, identify where they hesitate or get frustrated, and eliminate the friction.
1. The "Shadow & Observe" Method
Instead of asking users “What features do you wish you had?”, I ask them to share their screen and complete their routine workflow while thinking aloud. People are notoriously bad at predicting what they want, but their cursor never lies when they get lost.
2. Spotting the "Say-Do Gap"
In surveys, users frequently claim: “I want 20 custom filtering options and deep configuration settings.” But when you track actual production telemetry, 92% of users stick to the smart default view. Research must separate aspirational feedback from actual behavioral patterns.
3. Rapid 48-Hour Clickable Prototypes
Testing static Figma mockups only tests visual layout. Testing a quick clickable React/Next.js prototype with real keyboard input, realistic latency, and live validation tests the actual muscle memory of the product. Testing early with 5 target users reveals 80% of usability bottlenecks.
4. Non-Leading Interview Techniques
- Avoid leading questions: Instead of asking "Did you find the checkout easy?", ask "Walk me through how you finalized the purchase."
- Embrace awkward silences: When a user hesitates, resist the urge to help immediately. The 5 seconds of struggle will tell you exactly what visual cue is missing.
- Dig into the "Why": When a user expresses frustration with a button, ask "What were you expecting to happen when you clicked that?"
- Group observations by friction severity: Critical blockers (cannot finish task), Cognitive friction (slowed down), and Aesthetic feedback (cosmetic opinions).
Good UX research does not validate your preconceived ideas; it humbles your assumptions and shows you the product through fresh, unjaded eyes.
5. From Raw Notes to Code Action Items
The outcome of a research sprint should never be a static PDF report that gathers dust in Google Drive. It should be a crisp list of actionable UI tweaks: simplifying form fields, adding autofocus states, rewriting confusing microcopy, or rethinking information architecture.
