Skip to main content
Data Tables (/platform/data-tables) is the workspace for comparable metric extraction across a watchlist. You write one prompt describing the metrics you want, and the platform runs it against every entity, normalising the results into a sortable, filterable matrix.

​
When to use Data Tables vs other surfaces

The difference from Deep Dive is the shape of the interaction: Deep Dive is a conversation where you ask follow-ups and refine the answer; Data Tables is a one-shot extraction across a whole list.

​
The workspace

There’s one workspace per prompt. The sidebar lists every prompt you’ve written under My tables, and anyone else’s under Shared with me. The main area has three parts:
  • Prompt panel: the prompt body, editable inline. It collapses while you work on the matrix.
  • Matrix: rows and columns of extracted values. Every value is traceable: click a cell for its reference, source document, pages and notes, or click an entity name for a panel with two tabs:
    • Documents: the source filings used for that entity.
    • Notes: caveats, ambiguities, methodology and restatements flagged during extraction, with links to the cited pages.
  • Toolbar: the view selector, the Pivot button, filters, copy, and Excel export.

​
Creating and running a table

New Table opens one window covering everything a run needs: a name, the watchlist to run against, the instructions, how far back to search, and any documents of your own you want included.
1

Write the prompt

Describe the metrics you want, the periods to cover and any sector-specific rules. You can load the instructions from one of your existing tables or tear sheets as a starting point, then edit from there.
2

Check the layout

The window shows the rows and columns it read out of your prompt, before extraction starts. Rename anything that came through awkwardly, drop a row or a period you didn’t want, set a unit on a line item, and reorder them.
3

Run on a few CUSIPs

Pick a handful of representative CUSIPs. Extraction covers only your selection, so trying a prompt out is cheap.
4

Tune the prompt

Adjust the prompt body until the cells look right. Re-running an edited prompt replaces the table with a fresh generation.
5

Extend to the rest of the watchlist

Once the prompt is dialled in, click Run again and select the remaining CUSIPs. Already-extracted entities are locked and keep their cells; only the newly selected ones are extracted, and the matrix fills in as they come back.
Between runs, the scheduled sweep refreshes only the entities the table already covers, so a partial table never grows on its own. Adding a CUSIP to the watchlist does extend a table that was applied to the whole list.

​
Options

Alongside the prompt, a few choices control what the extraction runs over:

​
Views

The same extracted cells can be sliced three ways. Switch between them from the toolbar: Pivot, next to the view selector, swaps rows and columns in whichever view you’re in. It doesn’t change the saved table, and it stays in the URL, so a refreshed or shared link opens the same way you left it. Below those, filters narrow the matrix by status, value type, period, entity and audit status.

​
Getting the data out

  • Copy puts the matrix on your clipboard, laid out as you see it, ready to paste into a spreadsheet.
  • Export to Excel has two options:
    • Values only (.xlsx): numbers written as real numbers with unit-aware formats, so they stay sortable and summable.
    • With links to notes (.xlsm): the same workbook plus a right-click View Extraction Info on any data cell, which opens that value’s notes in your browser. Excel blocks macros by default, so you’ll need to click Enable Content when you open it.
Either way you choose which periods, entities or line items become workbook tabs, up to 100 at a time. Search the list, tick the ones you want, and download only those.

​
Tips

  • Be explicit about units and time bases. “Revenue in USD millions” and “Fiscal year ending June 30” beat “operating revenue” for sortable comparability.
  • Pin sector idiosyncrasies in the prompt. For example, exclude entrance fees from DSCR for charter schools, or specify GASB vs FASB.
  • Start on a small selection. Run the prompt on a few CUSIPs first, then tighten definitions, units, and output shape before extending it to the rest of the watchlist.
  • Read the layout before you run. A row list that doesn’t match what you meant is faster to fix in the run window than by rewording the prompt and running again.

​
See also