- Moderated usability testing
- The default. Watch somebody attempt the real task, count where they stop. Ran remotely over Zoom with screen sharing on the McDonald’s study, in person on ioMoVo and Toronto Cupcakes.
- Heuristic evaluation
- Fast, cheap, and best used after a live session rather than before one — an audit tells you what violates a principle, not what actually stops a person.
- Surveys
- For reasoning at a scale interviews cannot reach. Useful for “how often” and “how many”; unreliable for “why” without a session behind it.
- A/B testing
- Turns a recommendation into a measurement. Only worth setting up once you have named the threshold that counts as success.
- Analytics review
- Google Analytics to check whether a qualitative finding happens often enough to prioritise. Frequency, not cause.
- Session coding and affinity mapping
- Recordings reviewed afterward so the same observation is counted the same way across every participant. It is what turns eight anecdotes into “seven of eight”.
- Task scenarios and screeners
- Writing the task so it does not contain its own answer. The hardest part of the whole method and the least discussed.
02 — Design
Making it legible.
Figma for everything below, with real content in the frames rather than placeholder text.
- Information architecture
- Naming things the way people ask for them instead of the way the database stores them. Three of the seven case studies turned out to be this problem wearing a different hat.
- Interaction design and prototyping
- Clickable enough to test, not so polished that feedback becomes about the visuals. Mobile and web flows kept consistent across both.
- Design systems
- Type scale, spacing scale, one accent, states defined once. This site runs on a nine-step spacing scale and a single set of component rules — no page invents its own pattern.
- Accessibility
- Contrast checked against WCAG AA, full keyboard operability, visible focus, reduced-motion respected. Treated as a requirement, not a pass at the end.
- Content and microcopy
- Plain language in the interface, especially in errors. Most “confusing UI” is a labelling problem wearing a layout costume.
03 — Build
Shipping it.
Enough front-end to build and ship the real thing, which changes what I hand a developer.
- HTML, CSS, JavaScript
- Semantic markup, modern CSS, vanilla JavaScript. This site is hand-written, has no build step, and no framework — including the landing animation and the search palette.
- Static hosting and edge
- Deployed on Netlify with edge functions and serverless endpoints where a page needs something the browser cannot do alone.
- Prototyping in the browser
- When an interaction depends on timing or momentum, a coded prototype answers the question and a Figma frame guesses at it.
- Handoff
- Specifications in the units engineering uses, with states and edge cases written down. Knowing what is expensive to build changes which design I argue for.
04 — Camera
Watching closely.
The practice that trained the observation habit the research runs on. Details on the film page.
- Cinematography
- Documentary, automotive and brand work. Handheld and locked-off, available light first.
- Photography
- Portrait, editorial, event and product. Short sessions, because people stop being themselves after about forty minutes.
- Colour and edit
- Grade and cut in-house. One revision round inside the agreed turnaround, delivered at the sizes the destination needs.
- Directing a subject
- Getting somebody comfortable enough to be themselves on camera is the same skill as getting a participant to think out loud in a session.
05 — Limits
What I am still bad at.
A toolkit page that only lists strengths is a sales document. These are the four honest gaps, and what I am doing about each.
I am confident with eight participants and a coded transcript. Statistical significance on a sample of ten thousand is a different discipline and I bring somebody in for it.
I can specify and code an interaction. I do not build illustrated animation, and I stop before pretending otherwise.
Comfortable with static hosting, serverless endpoints and APIs. Database architecture and scaling are not mine.
My instinct on a new brief is to research more than the decision requires. I now write the decision the study has to serve at the top of every plan, and cut back to it.