Open the Documents page and look at the OCR Status column. The status tells you what to do next.
| OCR Status | What it means | What to do |
|---|---|---|
| Queued | The document is waiting its turn. | Keep Excel open and wait. |
| Processing | Text is being read. | Keep Excel open and stay on the Documents page. |
| Not Started or OCR required | OCR never ran, or it was interrupted. | Click Start OCR on the document. |
| OCR failed | Reading failed. | Click Re-run OCR on the document. |
| Complete, but "No OCR text detected" when you open it | OCR ran and found no words. | Click Re-run OCR. If it finds nothing again, see "Scans with no text found" below. |
OCR seems stuck
Keep Excel open and stay on the Documents page while OCR runs. If you leave the page, close the pane or close Excel during the import, the OCR result can be lost.
- Go back to the Documents page and wait a few minutes with Excel open.
- If the status still has not moved, close and reopen the Blast Audit pane, then open Documents again.
- For any document now marked OCR required or Not Started, click Start OCR.
- For a document still stuck on Processing, click the ⋮ button at the end of its row, then Re-run OCR. To restart several at once, select them and use ⋯ > Re-run OCR in the selection bar.
You can still open a document whose OCR is empty or stuck: the viewer shows a banner with a button to run OCR again.
OCR is slow
PDFs that already contain text are read directly from the PDF, which takes seconds. Scans and photos go through visual OCR, which takes longer and depends on the number of pages.
To speed things up:
- Import large batches in smaller groups.
- Split very long scanned files before importing them.
- Keep the Documents page open until every status reads Complete.
Messages about limits
| Message starts with | What to do |
|---|---|
| Your OCR quota is reached | Check your organization's usage with your administrator. Then click Start OCR on the documents marked OCR required. |
| Your organization already has the maximum number of documents waiting for OCR | Wait until some documents finish, then start OCR on those marked OCR required. |
| Your organization already has too much volume waiting for OCR | Same as above. |
| The OCR service is busy | Wait a few minutes, then start OCR on the documents marked OCR required. |
| OCR could not be re-run. The previous OCR result was kept. | Try again later. Your earlier result is still usable. |
Scans with no text found
"No OCR text detected" means OCR finished but found no words, so you cannot snip that document.
- Re-run OCR once. A temporary error can cause an empty result.
- Open the document and check the scan: a blank page, a very faint scan or a photo taken at an angle gives little or no text.
- Rescan the paper at a higher resolution, or ask the client for the original PDF, then import the new file.
Snips look wrong on a PDF that has text
Some PDFs carry a text layer that does not match what you see on the page. If snips on such a document look wrong or miss values, click the ⋮ button at the end of its row on the Documents page, then Run visual OCR. Blast Audit then reads the page images instead of the PDF text.
Test whether the problem is the document or your setup
- Open a new blank workbook.
- Import one small PDF, for example a one-page invoice.
- Wait for the status to reach Complete.
If that works, the problem is linked to the original documents or their size. If it also gets stuck, the problem is in your Excel or network setup.
If it still fails
Contact support with:
- The OCR Status shown and any message, as a screenshot.
- Your Excel: Windows, Mac or web, and the version.
- The number of documents and roughly how many pages, and when you imported them.
- The result of the blank-workbook test above.
Do not send the client documents unless support asks for one to reproduce the problem.