How to Extract Text From a PDF: Text Pages vs Scanned Pages

A PDF page can contain text or only a picture of text

A PDF describes how a page should look, but it does not guarantee that the visible words are stored as characters. A document exported from Word, Google Docs or a publishing application usually contains text objects that can be selected, searched and extracted. A scanner often produces one large image for each page instead.

That distinction explains why two nearly identical PDFs can behave differently. If dragging over a sentence selects individual words, embedded text is probably present. If the whole page behaves like one picture, the document probably needs optical character recognition, or OCR.

• Text-based PDF: fast extraction, searchable words and usually more accurate results.
• Scanned PDF: page images must be interpreted with OCR before the words become editable.
• Mixed PDF: some pages contain text while attachments, signatures or scanned inserts do not.
• Outlined text: letters may be stored as drawing shapes, so they look sharp but cannot be extracted as characters.

Extract the text while keeping a useful connection to the source

Work from the original PDF and create a separate TXT result. Plain text is convenient for searching, quoting and drafting, but the PDF should remain the visual reference for page layout and final verification.

1. Choose the pages you actually need

   Enter one range or several comma-separated ranges. A smaller selection is easier to review and avoids mixing unrelated sections.

2. Keep page headings when references matter

   Page separators make it easier to trace a quotation or number back to the corresponding PDF page.

3. Extract the embedded text

   The browser reads text objects from the selected pages locally and builds an editable result without uploading the document.

4. Review pages reported as blank

   A blank result does not always mean the page is empty. Open the source page and check whether it is a scan, outlined lettering or a restricted document.

5. Copy or download the result

   Use the TXT version as a working copy and keep the PDF as the authoritative visual source.

Why extracted text does not perfectly preserve the layout

PDF text is often stored as positioned fragments rather than paragraphs. A heading, table cell or individual word may have its own coordinates. Extractors infer reading order from those objects, so multi-column pages, sidebars, footnotes and complex tables can appear in an unexpected sequence.

A TXT file also has no concept of fonts, page geometry or floating boxes. Use it when the wording matters more than the design. If the exact layout matters, keep the PDF or extract individual pages as images for visual reference.

Use OCR for scans, then proofread the result

When no embedded text exists, convert the necessary PDF pages to clear images and run those images through OCR. Use a sufficiently high resolution, keep the page upright and crop unnecessary borders when possible.

OCR predicts characters from pixels, so it can confuse similar shapes such as 0 and O, 1 and l, or 5 and S. Names, totals, dates, account numbers, legal wording and citations always deserve comparison with the original page.

• Use direct extraction first because it is faster and normally more accurate for text-based PDFs.
• Use OCR only for pages that contain images of text or inaccessible lettering.
• Never assume a clean-looking OCR result is error-free.
• Keep sensitive files and extracted text in appropriately protected storage after the browser work is finished.