Tabular review lets you review multiple documents against a set of questions in a table, rather than one document and one question at a time. You pick a starting point — a playbook or a saved question list — and it builds a table where each row is a document and each column is a question or playbook rule, with AI-generated answers populated in the cells.
Key capabilities:
A starting point from your Playbooks or question lists, then open a detailed table view.
Filter rows by tags or review status, select individual cells or ranges to mark as reviewed or tag in bulk.
Multi-select across several results to act on them at once.
Search tags by name or create new ones on the fly from the tag picker.
Columns generated from a playbook rule show which playbook they came from.
Each cell links back to its source: selecting a cell's evidence jumps to and highlights the exact passage in the document that supports the AI's answer.
Export the table to CSV or XLSX, with review markers, tags, and a reference sheet of supporting evidence included.
You can switch to other Ask conversations while viewing a tabular review and come back to it later.
In this section,