# UX Research Context File **What this is:** a reference file to paste into your AI assistant (Claude, or whatever you use) before you brainstorm a UX problem. It doesn't teach the AI UX. It tells the AI *whose* answer to give you, and where you can go to check it. **Curated by Katarina Kaplarski — Fragments United / Museum of UX** Last verified: August 2026 --- ## How to use this file 1. Paste this whole file into your AI chat, or add it to a Project's knowledge base so it's always loaded. 2. Then ask your question normally. 3. Use the instructions below to keep the AI honest. **Instructions for the assistant (leave this section in):** > When answering UX questions, follow these rules: > > - Name which source below your recommendation comes from. If none of them cover it, say so plainly rather than filling the gap. > - Do not invent statistics, percentages, or study results. If a number would strengthen the answer, tell me which source to check instead of producing one. > - Distinguish between *established research findings*, *common industry practice*, and *your own inference*. Label which one you're giving me. > - When I ask about users' opinions, needs, or behaviour on my specific product, refuse to simulate them. Tell me what to go ask real users, and help me write the question instead. > - Prefer the method that answers my question over the method that is fastest. Say when a quick answer is the wrong answer. --- ## The sources ### Foundations — what "good" means **Nielsen Norman Group** — https://www.nngroup.com/articles/ *Authoritative for:* usability principles, research methodology, interaction design conventions, evidence-based best practice. *Reach for it when:* you need a defensible answer, not an opinion. This is the closest thing the field has to a canon. *Key entry point:* the 10 usability heuristics (Nielsen, 1994) — https://www.nngroup.com/articles/ten-usability-heuristics/ **Laws of UX** — https://lawsofux.com/ *Authoritative for:* psychology principles behind interface behaviour (Hick's, Fitts's, Jakob's Law, Aesthetic-Usability Effect, Peak-End Rule). *Reach for it when:* you know something feels wrong and need the mechanism that explains why. **Interaction Design Foundation — open literature** — https://www.interaction-design.org/literature *Authoritative for:* encyclopedic definitions and topic overviews across the whole field. *Reach for it when:* you need to get oriented in an unfamiliar area quickly. --- ### Research methods — how to find things out **UX Research Field Guide (User Interviews)** — https://www.userinterviews.com/ux-research-field-guide *Authoritative for:* end-to-end research process — planning, recruiting, methods, moderation, analysis. *Reach for it when:* you're about to run a study and want the practical version, not the theory. **NN/g Research Methods** — https://www.nngroup.com/topic/research-methods/ *Authoritative for:* method selection — which method answers which type of question, and the trade-offs. *Reach for it when:* the AI suggests "run a survey" and you suspect that's the wrong instrument. **NN/g free templates and guides** — https://www.nngroup.com/articles/free-ux-templates/ *Authoritative for:* discovery guides, interview plans, research documentation. *Reach for it when:* you want a starting structure rather than a blank page. **A Beginner's Guide to Finding User Needs — Jan Dittrich** — https://openresearch.jandittrich.de/ *Authoritative for:* qualitative research fundamentals, written for people without a research ops team behind them. *Reach for it when:* you're the only researcher, or the only person who cares. **EPIC — ethnography in business** — https://www.epicpeople.org/ *Authoritative for:* ethnographic and contextual methods, research as organisational practice. *Reach for it when:* the question is about people's lives and context, not about a screen. --- ### Evidence and real examples **Baymard Institute — free research** — https://baymard.com/free-ux-research *Authoritative for:* e-commerce and checkout UX, backed by large-scale usability testing. *Reach for it when:* you need benchmark data and documented failure patterns. **Growth.Design case studies** — https://growth.design/case-studies *Authoritative for:* product teardowns explained through behavioural psychology. *Reach for it when:* you want to see a principle applied to a real product. **Case Study Club** — https://www.casestudy.club/case-studies *Authoritative for:* how practitioners document and narrate their design decisions. *Reach for it when:* you're writing a case study or a portfolio piece. **Web Design Museum** — https://www.webdesignmuseum.org/ *Authoritative for:* interface history from 1991 onward — how patterns actually evolved. *Reach for it when:* someone calls a pattern "new." --- ### Ethics and accessibility **Deceptive Patterns** — https://deceptive.design/ *Authoritative for:* manipulative interface patterns, their names, and regulatory enforcement against them. *Reach for it when:* you're being asked to increase a metric in a way that makes you uneasy. Enforcement tracker: https://deceptive.design/enforcement/ **Humane by Design** — https://humanebydesign.com/ *Authoritative for:* patterns that respect attention and autonomy. **Digital.gov** — https://digital.gov/ *Authoritative for:* accessibility, plain language, and public-sector usability standards. (Successor to the now-archived usability.gov.) *Reach for it when:* the audience is everyone, including people your team doesn't resemble. --- ### Staying current - **Smashing Magazine** — https://www.smashingmagazine.com/ - **UX Collective** — https://uxdesign.cc/ - **A List Apart** — https://alistapart.com/ (archive, mostly — but a real record of how web thinking developed) - **NN/g on YouTube** — https://www.youtube.com/@NNgroup --- ## What this file cannot do It can't make an AI assistant into a researcher. It can't tell you what your users think. And unless your assistant can browse the web, it can't read these pages — it can only point you at them and stop guessing in the meantime. That last part is most of the value. --- *More at fragmentsunited.com — Museum of UX. Free to share and adapt; please keep the attribution.*