Frame the Real Work
Listen for the structure inside the language.
User research does more than reveal pain points. It gives us the raw material for the conceptual model: the things people work with, the actions they take, the states they recognize, and the distinctions they rely on to make progress.
A practical way to begin is to listen for nouns, verbs, and adjectives. Nouns can reveal objects. Verbs can reveal operations. Adjectives can reveal attributes and states. This is not a mechanical formula, and not every word deserves to become a product concept. It is a disciplined way to notice the user’s world before imposing the team’s vocabulary on it.
Nouns
Reveal the objects people create, exchange, approve, protect, and return to.
Verbs
Reveal the operations people perform and the outcomes they are trying to produce.
Adjectives
Reveal the attributes and states that change what people can decide or do next.
Watch how the work actually moves.
Language alone is not enough. Observe the spreadsheet running the department, the policy no one can explain, the unofficial handoff held together by two trusted employees, and the notes people keep because the official system forgets what matters.
These are not distractions from the “real” process. They are evidence of the real process. They reveal missing context, hidden authority, brittle dependencies, and the gap between the official model and the operation people must sustain every day.
A workaround is often the clearest available diagram of an unmet capability.
Name the capability problem before the feature.
A request for a new screen may actually be a request for shared visibility. A request for automation may be a request for consistency. A request for AI may be a request for access to judgment that currently lives in one expert’s head.
Framing the problem at the capability level changes the solution space. It lets the team consider process, policy, roles, data, training, and technology together. Sometimes the right answer is an interface. Sometimes it is a clearer decision rule or a redesigned handoff. Often it is a combination.
Research. Model. Interface.
The order matters: research how people actually work, develop the conceptual model, then design the interface. Starting with the interface risks encoding the team’s assumptions before the team understands the user’s world.
Once those assumptions become screens, they become surprisingly difficult to question. A grounded model keeps the design open long enough for the operation to become visible.
The user’s language is a doorway into the system.
The next field note continues the BFL framework from evidence to lasting capability.