Skip to main content

Spreadsheet Generation in Glean

Glean can turn a chat prompt into a spreadsheet, or take a spreadsheet you already have and work with it in Canvas. Iterate on the rows and columns with Glean in Canvas, then export to Excel, Google Sheets, or CSV — or copy the table straight into another tool.

Prerequisites​

Spreadsheet generation is available wherever Glean and Canvas are enabled. There is no admin template or separate setup step.

If you don't see the option to create a spreadsheet, the most common causes are:

  • Canvas is not yet enabled for your organization.
  • You need to hard-refresh the page after recent feature rollouts.

How spreadsheet generation works​

Spreadsheet generation uses Canvas as the working surface. You describe the spreadsheet you want in chat, refine it with follow-up prompts to Glean, and then export it in your preferred format.

Start from chat​

1

Ask Glean for a spreadsheet

In Glean, describe the spreadsheet you want. For example: "Build a spreadsheet comparing our top five competitors across pricing, target market, and key features."

2

Iterate with Glean in Canvas

Glean opens Canvas with a draft spreadsheet. Use follow-up prompts to add or remove columns, change which rows appear, restructure the data, or pull in additional sources until the spreadsheet matches what you need.

3

Export the spreadsheet

Use the export control in Canvas and choose your format:

  • Excel — downloads a .xlsx file you can open in Excel or any compatible application.
  • Google Sheets — exports the spreadsheet directly to your Google Drive.
  • CSV — downloads a plain .csv file for use in any tool that accepts CSV input.
  • Copy to clipboard — copies the table contents so you can paste into another spreadsheet, doc, or message.
4

Open and share

Open the exported file in Excel, Google Sheets, or your tool of choice to continue editing and share with your team.

Start from an existing Canvas​

If you already have a Canvas open — for example, a brainstorm or research summary — you can ask Glean to turn the relevant content into a spreadsheet from the same Canvas without starting a new conversation.

Start from an existing spreadsheet​

You can also bring an existing spreadsheet into Glean. Upload a file in chat and ask Glean what you want to do with it — for example, summarize it, restructure it, or pull additional data into new columns. Glean opens the spreadsheet in Canvas so you can iterate with follow-up prompts and then export the result.

What Glean generates​

When you generate a spreadsheet, Glean produces:

  • A structured table with column headers and rows based on the prompts you've given Glean.
  • Content drawn from the company knowledge and sources referenced in your chat — for example, documents, tickets, and search results that Glean used to answer your prompt.
  • A Canvas representation you can keep iterating on with Glean before exporting.

Use cases​

Spreadsheet generation is designed to help with common tabular tasks, including:

  • Competitive comparison tables — compare vendors, products, or features across consistent attributes.
  • Project trackers — compile status, owners, and dates from project tools and chat into a single sheet.
  • Metric roll-ups — aggregate numbers referenced across multiple documents or reports into one view.
  • Catalog and inventory extraction — pull structured lists out of long-form documents or search results.
  • Tool-item tables — turn meeting notes or decisions into a tracked list of follow-ups.

Tips for better results​

  • Describe the columns you want. Naming the columns up front (e.g., "columns for vendor, pricing tier, region, contract end date") gives Glean a clearer target.
  • Iterate with follow-up prompts. Ask Glean to adjust columns, filter rows, or pull in additional sources until the spreadsheet looks right before you export.
  • Reference internal sources. Ask Glean to pull from specific documents, projects, or teams so the rows are grounded in real data.
  • Review after exporting. Treat the exported file as a starting point — verify cells against source-of-truth systems before sharing widely.

Frequently asked questions​

See also​