Every Google Sheets limit, in one table

Google Sheets has three kinds of limits that matter for automation. The first is size: as of September 2026, a spreadsheet can hold up to 20 million cells or 100 MB. Google doubled the cap from 10 million that month (Rapid Release domains from September 10, Scheduled Release from September 28), so many guides and AI answers are now out of date. The second is IMPORT functions. IMPORTXML, IMPORTHTML, IMPORTDATA, IMPORTFEED and IMPORTRANGE share a traffic budget that is enforced on the spreadsheet's creator. IMPORTRANGE is capped at 10 MB of received data per request. The third is the Sheets API: 300 read and 300 write requests per minute per project, and 60 of each per minute per user per project. That per-user figure is the source of the 429 errors that integrations report.

Every value here comes from Google's help center, developer documentation or the Workspace Updates blog, verified November 12, 2026. Google does not currently publish a column limit or a per-tab limit on the official file-size page, so this page does not state one. For the error messages these limits produce, see Google Sheets import errors. Script-based automation has its own quotas, covered under Apps Script quotas.

The official limit is per spreadsheet file, counted across all tabs: 20 million cells or 100 MB for spreadsheets created in or converted to Google Sheets. Empty cells in the grid count, so deleting unused rows and columns frees capacity. The 100 MB figure also applies to imported files. Google's September 2026 announcement says it plans to raise the byte-size limit for imported spreadsheets later. Until then, a large CSV or .xlsx file can fail on bytes even when it is under 20 million cells.

Google does not publish a fixed number of IMPORT formulas per sheet. Instead it enforces a traffic limit on the creator of the document, summed across all open documents that person created. A collaborator's edits can also count against the creator's quota. When the budget is exhausted, cells show Loading... with the error quoted below. Refresh timing is documented, and so are a few hard rules: no volatile inputs, and a one-time Allow access prompt for external URLs.

The Sheets API counts requests per minute, with separate read and write buckets. There is no daily limit as long as you stay inside the per-minute quotas, which refill every minute. A batch request counts as one request regardless of how many subrequests it contains, so batching is the most effective lever. Integration platforms that use one Google Cloud project for all their customers share the project-level 300-per-minute pool. Per-user quotas apply to each connected Google account.

Size: edits, pastes or API calls that would push the file past the cell limit are rejected. The API error begins This action would increase the number of cells in the workbook above the limit of followed by the cell count.

IMPORT functions: over-budget formulas show Loading... with Error Loading data may take a while because of the large number of requests. Try to reduce the amount of IMPORTHTML, IMPORTDATA, IMPORTFEED, or IMPORTXML functions across spreadsheets you've created. Oversized IMPORTXML queries return Result too large.

Sheets API: exceeding a per-minute quota returns HTTP 429: Too many requests. In practice the message reads Quota exceeded for quota metric 'Read requests' and limit 'Read requests per minute per user' of service 'sheets.googleapis.com' (or the 'Write requests' equivalent). Quotas refill every minute, so a retry after up to 60 seconds succeeds.

Batch API calls. One spreadsheets.values.batchGet or batchUpdate counts as a single request, which turns hundreds of cell writes into one call. Google also recommends truncated exponential backoff (wait 2^n seconds plus random jitter, capped around 32 to 64 seconds) on 429s.

Request a quota increase for the per-project limits in the Google Cloud console if your own project is the bottleneck.

Shrink the grid. Delete unused rows and columns, archive old data to a second spreadsheet, or move large datasets to BigQuery and use Connected Sheets.

Reduce IMPORT churn. Aggregate in the source sheet and IMPORTRANGE only the summary. Avoid chains and cycles of IMPORTRANGE, and replace hundreds of IMPORTXML calls with a scheduled fetch that writes static values.