Glossary

UserTold is a research operating system: it turns real user interviews into evidence your agent can ship against. This is the vocabulary — and the loop it describes.

The loop

Interview → Evidence → Work.

You interview real users inside your product. UserTold extracts an evidence card from each proven moment. Evidence becomes work your agent ships — then completion resolves the evidence, and new interviews surface recurrence. Today is the briefing over the whole loop; Studies (and their Intake) are the setup beneath it.

Terms

Interview

A conversation with a real user, captured inside your product — screen, voice, transcript, and context. The raw material of research, billed per interview. (Browse them under Interviews.)

Evidence — an evidence card

A single proven user moment extracted from an interview: a typed, graded quote you can act on, shown as an evidence card. Each card carries its type (struggling moment, desired outcome, …), a confidence grade, the verbatim quote, and a link back to its source interview. (Browse them under Evidence.)

Work — a work item

An evidence-backed item to ship. Pushed to Linear or GitHub with its source evidence attached, so your agent knows why it exists. When the linked issue completes, UserTold resolves the evidence behind it and watches future interviews for recurrence. (Browse it under Work.)

Study

The research you're running: goals, script, and intake that decide what gets asked and of whom. A study produces interviews; interviews produce evidence.

Intake

The qualification step that decides who gets interviewed — a study's front door. Intake lives inside its study.

Today

The project briefing: the live state of the loop and what needs your decision right now — which studies are collecting, what evidence is awaiting triage, and what work is moving.