Qualitative data is everything you capture during a usability test that is not a number: what participants say as they think aloud, where they hesitate, the workarounds they invent, the confused facial expression before a misclick, and the explanations they give in a post-test interview. Where quantitative measures tell you that a task took 94 seconds or failed 4 times out of 10, qualitative data tells you why it happened and what the user believed was going on. Both are needed; neither is enough alone.
In this topic we look at how to capture that non-numerical evidence in a disciplined way. Structured observation and note-taking keep your records consistent across participants instead of dependent on memory. Post-test interviews and debriefs let you probe moments you noticed without leading the participant. Verbatim quotes and critical incidents (the specific breakdowns, errors, and moments of delight) become the evidence you cite in a findings report. Affinity diagramming is then the synthesis step: clustering dozens of raw observations into themes so that patterns, not anecdotes, drive your recommendations.
The three videos below come from Nielsen Norman Group, a long-standing UX research practice. They do not walk through a test session start to finish; instead they address the synthesis and interpretation side of qualitative work — how to organize what users say, think, feel, and do into a shared artifact, and how to classify the problems you observe. While watching, pay attention to the vocabulary used for categorizing raw observations, and think about how each technique would change the notes you take during a session. Ask yourself: what would I have to write down in the moment for this analysis method to work afterward?
A short explainer on the empathy-mapping technique, which organizes what you know about a user into quadrants covering what they say, think, do, and feel. The video describes how the mapping process distills scattered research knowledge into one place and how the resulting artifact serves as a quick reference for a team. Expect guidance on how the map is built and what each section is meant to hold.
The says/thinks/does/feels structure is a practical taxonomy for the types of qualitative data you collect in a usability session, and the clustering process it describes is closely related to affinity diagramming as a synthesis method.
This video breaks user pain points into three levels: interaction level, journey level, and relationship or organizational level. It explains how problems a user encounters can range from a small friction on a single screen to a larger breakdown across an entire experience. The framing helps you label the severity and scope of the problems you observe.
Documenting critical incidents requires a vocabulary for describing what kind of problem you just witnessed, and this gives you a clear, widely used classification to apply to your observation notes.
A companion piece focused on timing rather than technique: when in a project an empathy map is worth building. It covers using maps to plan an upcoming research study, to capture insights while research is underway, and to communicate findings to a wider team afterward.
It connects a synthesis artifact to the research workflow, which is exactly the decision you face when planning how to record and later report qualitative data from a controlled test.
These videos mainly support MO2 by sharpening how you treat the qualitative half of your data set — think-aloud utterances, observed behaviors, and inferred feelings — and by offering structures (empathy map quadrants, pain-point levels) for organizing raw notes into reportable findings. They indirectly support MO1, since knowing what you intend to capture shapes how you observe and what you probe during a debrief, and MO3, because classifying an observed problem by level forces you to ask whether it reflects a real interface issue or an artifact of the session setup. Pair them with the module readings on note-taking protocols and post-test interviewing, which cover the in-session mechanics the videos do not.
These videos were chosen from YouTube on October 3, 2026. Each has captions; the player turns them on. Videos on YouTube can be changed or removed — if one no longer plays, tell your instructor. If a player does not appear, use its Watch on YouTube link.