Studies
cursor · paginated
The studies in your workspace, newest first. Each entry carries title, topic, research objectives, status, modality, languages, duration, group size, participant count, and
hasAnalysis — the flag that tells your assistant whether findings exist yet.Returns 25 studies by default, up to 100 per page. A non-null nextCursor means there’s another page.study_id
One study in full: everything from the list, plus the discussion guide itself — each question’s script and goals, any stimulus shown — and session counts. Useful as context before reading findings, since it shows what participants were actually asked.The analysis tools read the guide as it stood when the analysis ran. Each question here carries
analysisGuideSectionId, its id in those tools. It reads null when the analysis has no results for the question, such as one added later.Findings
These need a completed analysis. Before that, they return a message saying so rather than empty data. Every quote keeps the participant’s own words and names the language they spoke. A quote from a session in another language also carries its translation into your workspace language.study_id
The main results bundle: overview, key findings with supporting quotes, recommendations, suggested follow-up studies, and segment insights. The usual starting point for “what did we learn?”A finding carries its counts when the analysis recorded them, such as “9 of 23 who answered”. Survey findings have none. Each count names its base: who spoke, who answered, who compared the concepts, or who saw one. It never lists the people behind the count.
study_id
The concept test results the report shows, one entry per concept-test question. Each entry has the verdict and each concept’s read: what drew people in, what held them back, and example quotes. It never includes a concept’s image or video link.The tallies count everyone and each segment. Their picks appear only when the report draws the pick chart. The picks’ base counts only participants who answered. Separate counts cover who declined every concept, who named more than one, and who gave no answer. A concept that people rated on a scale also lists those answers, such as interest levels. Each level counts out of who gave one.A study with no concept test returns an empty list.
study_id
Every theme the study’s findings draw on, from the latest analysis, with prevalence — how many participants out of the analyzed base expressed each one.
study_id · theme_id
One theme in depth: its definition, prevalence, and the verbatim participant quotes behind it.
study_id
The discussion guide question by question: per-question takeaway, summary, supporting quotes, and — where a stimulus was shown — how participants evaluated it.It carries no answer counts. For each participant’s answer and the counts, use
get_question_coding.Each question also carries currentPosition, where it sits in the study’s guide today. After a researcher moves a question, it differs from position. It reads null once the question is gone. get_participant_responses and get_emotion_overview carry it too.study_id · question_id? · guide_section_id?
List the saved single-priority coding questions. Supply
question_id to inspect every analyzed participant’s recorded answer and count membership.Exclusive choices, combined choices, and inclusive counts stay separate. Historical categories retain unknown exclusivity.
Each question names the guide question it measures when the analysis recorded one. Missing summaries never remove participants. Source passages include recorded times and checked status when live evidence supports them.
Participant row keys stay stable within a study, differ across studies, and omit internal participant identifiers.Supply a guide question’s id from get_study, get_question_breakdown, get_participant_responses, or get_emotion_overview as guide_section_id to inspect that question’s summaries. summaryContext.association reads linked when it is the coding question’s recorded guide question, and not_linked otherwise. Analyses saved before links existed read not_recorded.study_id · participant_id?
The participant-by-question grid — each participant’s answer summary for each question, tagged with the themes it supports. Pass
participant_id to narrow to one person. This is what makes “compare how participants differed on X” answerable.A question’s summaries cover only the participants who have a summary for it. A tally of them can miss answers. Count answers with get_question_coding.study_id
How participants felt at each question: per-question distributions over the seven Ekman emotions. Each count is a number of participants, not utterances. A participant who showed several emotions counts once under each. A second distribution counts each participant once, under the emotion they showed most. Aggregate only, with no per-participant or line-by-line detail.
Screener
study_id
Answer distributions for the closed-ended screener questions — multiple choice, yes/no, rating scales, matrix, ranked choice, MaxDiff, and image concept — with counts.Available once sessions have completed. Counts only people who took part in a completed session.
Search
query
Keyword search across the whole workspace: study titles and objectives, theme names and definitions, and participant quotes. A quote matches in the participant’s words or in its translation. Returns matching records with ids that
fetch can expand.id
The full record behind a search result. A study id returns the study plus its analyzed summary; a theme id returns the theme with its supporting quotes.
Open questions
study_id · question
Ask anything about one study and get an answer grounded in the full data, including complete session transcripts. This is the same engine behind Explore in the dashboard.It reads far more than the structured tools do, so it’s slower and costlier. Your assistant should prefer
get_study_summary or get_theme when those answer the question, and reach for this when the answer isn’t in the analysis — something a participant mentioned in passing, a tangent nobody coded, the exact phrasing someone used.For a “how many” question, it cites the recorded count and its base from the study’s answer coding. A number it works out from transcripts alone comes labeled as an estimate.Capped at 50 questions per workspace per day, resetting at midnight UTC. Hitting the cap doesn’t affect any other tool.Getting better answers
- Name the study. “In the checkout study, what drove abandonment?” saves a round of searching.
- Ask for the evidence. “…and quote the participants” pushes your assistant toward
get_theme, which carries verbatims, over a summary-level answer. - Ask across studies deliberately. “Compare the themes from our last three pricing studies” is the kind of question this connection is genuinely better at than the dashboard.
- Push back on prevalence claims. Themes carry real participant counts. If an answer says “most participants,” ask for the number — it’s there.