Why some PDFs have no selectable text

A PDF can look like a document while containing only page images. This happens often with scans, photographs, and faxed records. The letters are visible to a person, but a computer sees pixels rather than words.

A digital PDF exported from a word processor usually contains text objects. That text can be selected, searched, copied, and used by assistive technology. An image-only PDF needs an additional recognition step.

How optical character recognition works

Optical character recognition, commonly called OCR, analyzes an image and estimates which shapes represent letters, numbers, and punctuation. The result is often stored as an invisible or visible text layer over the original page image.

OCR is useful, but it is not identical to the source document. Accuracy can be affected by blur, skew, low contrast, unusual typefaces, handwriting, and complex tables.

Choose the right output

If you need to preserve the original page appearance while adding searchable text, use the OCR to text tool as part of the document processing workflow. If the goal is a clean text file for analysis, the PDF to text tool is a more direct fit.

For a document that will be edited or published elsewhere, decide whether plain text is enough or whether structure such as headings, lists, and tables must be preserved.

How to check OCR results

Search for a phrase that appears on the first page.
Compare names, dates, and numbers with the original scan.
Review columns and tables because reading order may change.
Check punctuation and symbols in technical or legal content.

OCR is an assistant, not a final proofreader

For low-risk archives, OCR output may be sufficient for discovery and search. For contracts, invoices, medical records, or identity documents, a human review is essential. A single misread digit can change the meaning of a record.

The best workflow preserves the original scan, stores the recognized output separately, and records which version was reviewed.