AI Summaries & Q&A

Ask a report a question in plain language, or have it write the summary for you.

ℹ️ Requires AI Access AI features are granted per user on the Users screen — AI Access for summaries and Q&A, Adv AI Access for the agentic features. They also have to be enabled on the instance.

Summarize a Visual

AI Summarize on a visual's menu produces a written description of what that chart or table is showing — the shape of the data, what stands out, and where the notable values sit. It reads the visual's current result, so it reflects whatever filters are applied.

Ask a Question of a Report

Ask in plain language and get an answer drawn from the report's own data. The conversation is kept, so you can return to a report and pick up where the questioning left off rather than starting over.

💡 Ask What the Report Can Answer Answers come from the data on the report. A question needing a field the report doesn't include can't be answered from it — add the field, or ask on a report that has it.

Agentic Answers

Where a question needs more than one look at the data, the agentic mode works through it in steps — querying, reading the result, and querying again — and streams its reasoning as it goes, so you can see how it reached an answer rather than just being handed one.

Across Every Report

Cross-report Q&A takes a question when you don't know which report holds the answer. It routes to the right report first, then analyses within it.

ℹ️ Scoped to Your Access Cross-report search only considers reports you already have access to and that have AI enabled. It cannot surface data from a report you couldn't otherwise open.

Executive Deep Summaries

The most involved of these produces a full executive summary as a presentation, with generated observations and rendered visuals. See Executive Deep Summaries.

Guiding the Agent

Where an AI step runs as part of a scheduled delivery, you can supply written instructions for the agent — focus areas, boundaries for exploration, exceptions to ignore, and business logic it should take into account.

💡 Instructions Are Where the Domain Knowledge Goes "Ignore the reversal entries in October" or "treat anything under ₹500 as noise" is context the data doesn't carry on its own. Putting it in the instructions is the difference between a summary that reads as informed and one that flags things everyone already knows to disregard.

Comments on Rows

Not AI, but adjacent, and often used alongside it: rows in a table can carry comments, so the human interpretation of a number lives next to the number rather than in a separate thread. See Working with Tables.

⚠️ Check Before You Circulate Generated summaries and answers are a starting point, not a signed-off analysis. Read them against the underlying visual before forwarding them onward — particularly where a figure will be quoted.