Design and UX
UX research
Research is how a design team stops arguing from taste. It covers interviews, field visits, usability sessions, surveys, card sorts and analytics, and the real skill is choosing between them: knowing when watching five people beats counting five thousand, and when a number is the only honest answer. This page sets out the methods, the vocabulary and the traps that make findings look sturdier than they are.
Why this topic exists: Interviews, usability tests, surveys, analytics and when to use which: NN/g's Research Methods and User Testing topics; SFIA User research; the fourth course of Google's UX programme.
Every product decision rests on a belief about the people who will use it. Research is the discipline of testing that belief before it hardens into a feature. SFIA treats user research as a distinct skill, apart from the analysis and evaluation that follow, and the separation is worth keeping: gathering evidence, making sense of it and judging a design against it are different acts, and teams that blur them tend to find exactly what they hoped to find.
Two questions that sort the methods
Nielsen Norman Group arranges methods along several axes, and two of them do most of the work. The first asks whether you are recording what people report or observing what they do, and the two disagree more often than newcomers expect. The second asks whether you want to understand why something happens or to measure how often it does. Where a method sits on those axes decides what it can prove.
| Method | What it captures | Where it fits |
|---|---|---|
| Interviews | Goals, habits, vocabulary and frustrations as people describe them | Early, before anything is designed |
| Field or contextual study | Behaviour in the real setting, interruptions included | Early, when the context itself is unknown |
| Usability test | Where people stall, misread or abandon a task | Throughout design, on sketches as much as on builds |
| Card sort and tree test | How people group content, and whether they can find it in a proposed structure | While shaping the information architecture |
| Survey | Attitudes at scale, and nothing reliable about behaviour | Checking a pattern already seen in sessions |
| Analytics and A/B tests | What large numbers of people did, never why | After launch, against a baseline |
Read top to bottom, the table also traces a project's arc. Early work is generative, discovering which problem deserves solving; mid-project work is formative, shaping a design while change is still cheap; late work is summative, asking whether the shipped thing performs. Running a summative method at the generative stage produces confident numbers about the wrong question.
From sessions to findings
Raw notes are not insight. The working vocabulary here starts with the affinity diagram, where observations are clustered until themes surface; then the persona, a composite of real participants that keeps a team honest about whom it serves; and the journey map, which lays out someone's steps, feelings and sticking points over time. Each is only as good as the sessions behind it, and a persona assembled from marketing segments instead of interviews is fiction with a stock photograph attached. Findings that reach a product decision are usually short, specific and tied to a clip of a real person struggling, because nobody argues with the clip. Google's UX programme gives research a course of its own, a reasonable place to practise planning a study and reporting on it. The natural next steps are UX and interaction design, where findings turn into structure, and accessibility, which widens whom you recruit.
The question that spoils most first studies
Asking people whether they would use a feature. Predictions about future behaviour are unreliable, and polite participants overstate their enthusiasm. Ask about the last time they met the problem, or hand them a task and watch.
Next to this topic
- UX and interaction designInformation architecture, flows, patterns and the behaviour of interfaces: NN/g's Interaction Design and Information Architecture topics, IxDF's UX and HCI literature, SFIA User experience design.
- UI and visual designLayout, typography, colour, hierarchy and the craft of the screen: NN/g's Visual Design topic, IxDF's UI design, colour theory and Gestalt literature.
- Design systemsTokens, components, documentation and governance so that many hands ship one product: the site's design-system concept, IxDF's brand-guidelines literature, and the meeting point with front-end code.
- Accessibility in designWCAG's four principles and conformance levels, assistive technology and inclusive design: NN/g's Accessibility topic, SFIA Accessibility and inclusion, W3C WAI.
- Prototyping and design toolsFigma, Sketch, Adobe XD and the prototyping loop from paper to clickable: NN/g's Prototyping topic; Udemy's Design Tools shelf; the third and fifth courses of Google's UX programme.
- Graphic design and illustrationComposition, type, image and print, and the Adobe and Affinity toolchain: SFIA Graphic design; IxDF's graphic design literature; Udemy's largest design shelf after 3D.
- 3D, animation and motion designModelling, rigging, rendering and motion in Blender, Maya, 3ds Max, ZBrush, After Effects and Cinema 4D: SFIA Animation development; Udemy's largest design shelf.
- Web designLayout for the browser, responsive design, WordPress themes and the hand-off to code: Udemy's Web Design shelf; the programming field owns the build.
- Design assets and licensingTemplates, stock media and fonts, and the licences that decide what you may do with them: the site's design-assets concepts and the /digital-assets/ partner picks.
- Design thinking and service designProblem framing, ideation and the end-to-end service: IxDF's design-thinking and service-design literature; NN/g's Design Process and Customer Journeys topics; SFIA Customer experience.
Concepts to know
Glossary entries with the reason each one matters here.
- Design system
Research findings feed the system's patterns.
Certifications that test it
Vendor exams and free certificates; facts, cost and the preparation path are on each page, and the certifications hub has them all.
Frequently asked
- How many participants does a usability test need?
- For finding problems in a qualitative test, a small round is usually enough, and several small rounds between design changes teach more than one large study. Quantitative benchmarks are another matter: measuring a success rate or time on task with any confidence takes far more people. Decide which you are running before you recruit.
- How is UX research different from market research?
- Market research asks who might buy and why; UX research asks how people use something and where it lets them down. They share methods such as interviews and surveys, but market research segments audiences and sizes demand, while UX research watches behaviour against a specific design or task. Good teams read each other's reports.
- Can designers run their own research?
- Yes, and in small teams they usually do. The risk is defending your own design mid-session: leading questions, rescuing a stuck participant, hearing politeness as proof. Write the task script in advance, stay quiet while people work, and ask a colleague to review the notes when the stakes are high.
Courses in the directory
24 courses are filed here; the top 6 by our ranking, details and the provider link on each course page.
Last reviewed 26 September 2026 · Getting Digital