Skip to main content
Analyse is where responses become results: indicator scores, themes drawn from open-ended answers, and the quotes behind both.

​
Layout

Analyse is a single feed rather than separate tabs. A bar at the top holds a filter button, a search box, a running count of respondents, and a refresh control. Below that comes an executive summary, then your indicators and questions one after another. A Group by toggle switches between Interview order, where Analyse opens, and Study focus, which reads Outcomes when your study focus has no further topics. When your responses come from more than one interview version, a line above the feed says so, and opening it shows how many responses each version has. On a wide enough screen, side panels show which outcome or demographic each part of the feed relates to, and let you jump straight to a specific section.

​
Indicator scores

Every indicator you set up in Plan is calculated automatically from responses and shown as a chart as soon as enough responses come in, in whichever form fits the data: a gauge, a ring, a bar, a histogram, or a set of grouped bars. Opening an indicator shows its concept definition, its source, how it’s calculated, and, where the indicator defines them, Poor, OK, and Good bands so you can read the score at a glance. A Show underlying questions option reveals every question feeding that indicator, each with its own chart and quotes, so nothing is a standalone number. Under the card, short lines say how many answers the score used when some were left out, and how many older answers it scored with a changed scoring key. An answer option that keeps its code and label counts as the same option after its score changes, and no stored answer is ever altered. See How indicators work for where the indicators themselves come from.

​
Themes from open-ended answers

Every open-text answer is grouped into themes automatically, so instead of reading through hundreds of individual responses, you see the topics that came up most. You can switch how themes are displayed for a given question, between a simple bar list, a word cloud, or badges, whichever reads most clearly for that data. Every theme links back to the original answers and quotes it was built from. Opening a quote plays back its voice or video recording directly, if the respondent answered that way.

​
Breaking results down by group

Open the filter panel to disaggregate any result. Pick a dimension first (any closed question in your study, plus a built-in language filter when your respondents answered in more than one language), then pick the values or range you want within it. You can combine more than one dimension at once; each one you’ve applied shows as its own removable chip, so it’s clear at a glance exactly what you’re looking at.

​
Mixed-language responses

Where an answer isn’t in English, a toggle on it switches between the original wording and an automatic English translation, the same way it works in Review. This runs through the same AI gateway as the rest of the analysis.

​
Getting a written summary

An executive summary sits at the top of the feed, generated on demand. It follows a fixed structure, Overview, Key findings, Group differences, and Limitations, skipping any section that doesn’t apply, and its findings link back to the specific indicators and questions behind them, so you can click straight through to the evidence. If you’ve applied a filter, you can generate a summary scoped to just that group instead of the whole study. For a conversational way into the same data, see Explore. To turn your analysis into a document, see Report.