Strategic Note-Taking for UX Research & Better AI Prompts

What if the most powerful research tool you have isn't software — it's your own attention? This lesson introduces metacognition as a practical research skill: the ability to notice what's happening inside you while you're listening to someone else. You'll learn why your surprise, boredom, and gut reactions are valuable signals worth capturing — not because they're always right, but because they give you something concrete to test. You'll also try a simple 3-minute pre-interview practice designed to sharpen your awareness before a session, so you walk in ready to notice what matters.

You noticed something during that last interview — a flicker of surprise, a moment that felt off. But what do you do with it before it fades? This lesson gives you a simple, repeatable 4-step method you can use after every session: tune into your body, name the signal, write it down, and then bring in AI. You'll see an example of what a 5-minute post-session note dump looks like, learn how to turn raw signals into testable hypotheses, and discover why the order you do things in changes everything about the quality of your AI prompts.

So you know how to capture signals after a session — but what about during the interview, when things are moving fast? This lesson introduces a lightweight notation toolkit: a small set of marks that help you tag what matters in real time without losing the thread of the conversation. You'll learn three "buckets" of notations — one for capturing your thinking, one for steering the session, and one for setting up sharper AI prompts later — and you'll pick just one or two that fit your style. By the end, you'll see how a few simple marks can turn messy live notes into structured, actionable inputs.

During a live session, the most valuable signals are often the ones that flash and disappear — a moment of surprise, a flicker of confusion, something that confirms exactly what you suspected. This lesson gives you three lightweight markers you can drop into your notes in under a second: !! for surprise, ?? for confusion, and →→ for strong confirmation. You'll practise using them in real time, learn the common mistakes that turn markers into premature conclusions, and see how these breadcrumbs become testable hypotheses when you bring AI into the picture.

Research notes often flatten an experience into a list of quotes — but the real insight lives in the shape of the journey. Where did energy rise? Where did it crash? What happened right before the drop? This lesson teaches you to track emotional arcs in real time using four simple marks. You'll learn to name the trigger behind each shift in one sentence, spot patterns across participants, and hand those moments to AI for deeper evidence retrieval.

Your brain fires off follow-up questions the moment something interesting lands — but chasing every thread in real time turns an interview into chaos. This lesson gives you a two-tag system to capture your curiosity without losing the participant's flow and see how AI can turn your unanswered follow-ups into a structured guide for the next round.

Transcripts capture what people said — but they quietly erase why they said it that way. Was the participant rushing because they were disengaged, or because a child just walked into the room? Were the short answers a sign of low interest, or were they whispering in an open-plan office? This lesson teaches you to capture those invisible conditions using [CONTEXT] anchors: quick bracket notes that record constraints and changes like environment, device, timing, and who else is present. These details protect your analysis from confident-sounding misreads and give AI the information it needs to interpret data honestly.

It's the most natural thing in the world: you see someone hesitate and write "confused by the layout." But that's not what you saw — that's what you think it means. And once an interpretation lands in your notes unmarked, nobody questions it, least of all AI. This lesson gives you a clean two-prefix system for what you observed and what you think it might mean — so your facts stay solid and your guesses stay testable. You'll practise the camera test, learn to write interpretations as hypotheses with confidence levels, and see how keeping the two apart lets AI challenge your thinking instead of just confirming it.

People don't always do what they say they do — and that's not dishonesty, it's human nature. Someone calls a feature "easy" but never touches it. Someone says they check their account weekly but hasn't logged in for months. These gaps between words and actions are some of the richest material in research, and transcripts flatten them completely. This lesson teaches you to catch and record those contradictions in real time. You'll learn to tell behaviour contradictions from tone contradictions, keep your notes curious rather than judgmental, and use AI to find where the same tensions repeat across participants.

AI can sound confident even when it's wrong — and if you're not careful, that confidence can quietly become yours. This lesson teaches you a 4-step validation loop — draft, critique, verify, document — that turns any claim into something you can actually stand behind. You'll see how to pressure-test a finding by asking better questions, hunting for real evidence in your data, and building a clear trail from signal to conclusion. This is the habit that separates a polished-sounding insight from one that's genuinely trustworthy.

You've been capturing signals and building hypotheses — now it's time to put AI to work on your terms. This lesson gives you four prompt patterns designed to test your thinking rather than replace it: pressure-test a hypothesis from both sides, retrieve exact quotes, hunt for counterexamples, and compare how someone's experience shifted over time. You'll also learn what makes a prompt fail and how to build a personal prompt library that encodes your best research habits for reuse.

You've got a finding you believe in — but could you prove it to someone who wasn't in the room? This lesson shows you how to build an evidence trail: a simple table that connects every claim to the exact quotes, moments, and context it came from. You'll learn how to assess your own confidence honestly (based on coverage and contradiction, not gut feel), and how to use AI safely for quote retrieval without falling for fabricated evidence. The result is research that holds up — whether it's being reviewed by a stakeholder tomorrow or by future-you three months from now.

You've been building your own observations and letting AI expand your view — but how do you know where each of you is strong and where you're both missing something? This lesson introduces the calibration matrix, a simple 2×2 grid that compares what you noticed against what AI picked up. You'll walk through a worked example, learn to ask the honest question — is this my edge, or my bias? — and set up a quick weekly ritual that sharpens your self-awareness as a researcher over time. This is how you keep the human + AI partnership honest.

AI can help you move faster — but who's responsible when something goes wrong? This lesson makes ethics practical. You'll learn what's safe to share with AI tools (and what should never be pasted), how to disclose AI use clearly to both participants and stakeholders, and how to spot when AI quietly flattens messy human experiences into one smooth, misleading story. You'll also pick up lightweight habits — from a simple pre-upload checklist to an AI usage log — that keep your work credible, reproducible, and respectful of the people behind the data.

You've learned the notations, the prompts, the evidence workflow, and the calibration habit — now it's time to run the full cycle from start to finish. This capstone lesson walks you through a complete practice project: taking human-first session notes, writing validation prompts, building an evidence table, and comparing your observations against AI with a calibration matrix. You'll also get a downloadable template pack so you can keep using everything you've learned in your own research. The standard is simple: could someone else follow your reasoning and check your evidence?

You've built a full toolkit — notations, prompts, evidence trails, calibration, ethics habits. Now what? This final lesson brings it all together and points you forward. You'll get a quick recap of the complete workflow, a concrete plan for your next research session (one notation, one prompt, one calibration check), and a simple way to track your growth over time. Most importantly, you'll walk away with a clear sense of the researcher you're becoming: someone whose perception steers the work, with AI expanding what's possible.

Join The Research Lab waitlist: https://ux.universalmind.coach/lab

The Research Lab is a community for UX researchers working AI into their practice. Beta cohort opens soon. Spaces are limited!

What's inside:

• Prompts and playbooks pulled from real research projects

• Monthly Office Hours to think out loud with peers

• Quarterly Research Sprints, starting September 2026 with interviews that go somewhere