Why are scanned PDFs not showing up in our internal search?

Scanned PDFs often don’t show up in internal search because the search system can only index text, and scanned PDFs usually contain images of text, not actual selectable text.

Common reasons:

1. No OCR performed

   • The PDF is just a series of page images.
   • Search engines can’t read the words unless Optical Character Recognition (OCR) has been run.

2. OCR is missing or poor quality

   • If OCR wasn’t applied, or the scan is blurry/rotated/low contrast, the extracted text may be empty or inaccurate.
   • That means the document won’t match search queries well.

3. Indexing pipeline skips PDFs

   • Some internal search systems only index certain file types, or they may not process large/complex PDFs reliably.
   • Permissions or ingestion failures can also prevent indexing.

4. Text is embedded oddly

   • Even “digitally created” PDFs can have text in unusual layers/encodings that your indexer doesn’t extract correctly.

5. The document isn’t in the search scope

   • It may not be in the indexed folder/repository, or it may be excluded by rules like file size, page count, or restricted access.

What to check:

• Open the PDF and try selecting text. If you can’t, it likely needs OCR.
• Verify your ingestion/indexing logs for errors on that file.
• Confirm the search tool supports OCR for scanned PDFs.
• Reprocess the PDFs with OCR and reindex them.

If you want, I can help you troubleshoot this based on your search stack (e.g., SharePoint, Elasticsearch, OpenSearch, Azure AI Search, Google Drive, etc.).