Tag quotes to find meaning
Use tagging as analysis — turning a set of quotes into themes and findings.
Pressing t puts a tag on a quote. This page is about the part that matters more: what to tag, and how tagging turns a set of quotes into findings. For the mechanics — keys, installing codebooks, AutoCode — see Use codebooks and AutoCode.
Tagging is coding
In qualitative research, attaching a label to a piece of data is coding. A tag in Bristlenose is a code. The point is not to file quotes tidily; it is to name the meaning in each one so that patterns across participants become visible.
Two ways to build your codes
- Let codes emerge (inductive). Read the quotes and create tags for what you actually see — a recurring frustration, an unexpected workaround, a mental model that keeps surfacing. This is the default in exploratory research, and it is how Bristlenose forms themes.
- Apply a framework (deductive). Start from an existing scheme and tag against it — Norman's usability principles, Nielsen's heuristics, the others in Codebook frameworks. This is useful when you have a specific lens or need to compare against a standard.
Most studies use both: a framework for structure, plus your own tags for what the framework misses. You can tag the same quote through several lenses.
Tag the meaning, not the topic
"Onboarding" is a topic. "Gave up before finishing onboarding" is a finding. Code for what the quote tells you — a need, a barrier, a motivation, a moment of confusion or relief — not just the subject it mentions. Codes that name meaning are the ones that turn into themes.
Keep your codes consistent
Patterns are only real if the codes are applied consistently. As you go, merge tags that mean the same thing, rename ones that have drifted, and group related tags. Bristlenose lets you drag, rename, and merge tags in the codebook panel; doing this as you code keeps the later analysis honest.
From codes to findings
As you tag, Bristlenose shows where codes concentrate and how widely they are shared, as signal cards (see Sentiment and signals). A code that recurs across many participants, strongly, is a finding worth reporting; one that appears once is a note, not a pattern. Use the signals to decide where to look, then read the quotes themselves before drawing a conclusion.
Tagging is iterative. Expect to revisit and refine your codes as your understanding of the data grows — that revision is the analysis.