How to extract text from a PDF on a Mac
By Himanshu Ranjan · Updated 8 November 2026
To extract text from a PDF on a Mac, select it and run Convert › PDF to Text in FileDial for a PDF with real text, or OCR › Extract text with OCR for a scan. In Preview you can select text with Tools › Text Selection and copy it. All of it works offline.
First, find out what kind of PDF you have
The right method depends on one question: does the PDF contain text, or only pictures of text? Open it in Preview and try Tools › Text Selection, then drag across a line. If the line highlights, the file has a text layer and extraction is straightforward. If you can only drag a rectangle, the page is an image and the text must be read by optical character recognition, which turns the picture into characters.
Plenty of PDFs are a mix: a typed report with a scanned appendix, or a scanned contract with a few typed stamps. Decide per file, and when unsure, try the plain extraction first because it is exact, then fall back to OCR for pages that came out empty.
Extract real text with PDF to Text
For PDFs that already hold text, Convert › PDF to Text writes a plain-text copy beside the original. Plain text keeps the words and drops fonts, columns and images, which is the point if you want to search, paste or feed the words into another tool.
- Select the PDF in Finder, or select many.
- Hold the right Option key to open the dial.
- Choose the Convert category and release on Convert › PDF to Text.
- Open the .txt file beside the original in TextEdit or any editor.
Extract text from a scan with OCR
If the PDF is an image, use OCR › Extract text with OCR, which reads every page as an image and writes a .txt file. A 42 MB PDF takes about 2 seconds to OCR in FileDial's measured test; your time depends on page count and your Mac.
OCR is never perfect. Letters like l, 1 and I, or O and 0, get confused, and tables and columns can read in the wrong order. Accuracy improves with flat, well-lit, 300 dpi scans. If you also want to keep the PDF itself searchable, run OCR › Make searchable, which adds an invisible text layer to pages without text, instead of making a separate .txt.
- Typed PDF: Convert › PDF to Text.
- Scanned PDF: OCR › Extract text with OCR.
- Keep the PDF and make it searchable: OCR › Make searchable.
- Need structure: Convert › PDF to Markdown or PDF to Word.
Choosing between text, Markdown, Word and RTF
A .txt file loses structure. If headings matter, Convert › PDF to Markdown guesses headings from font size, which is a heuristic rather than a guarantee. If you need paragraphs, tables and pictures where they were, Convert › PDF to Word is closer. Convert › PDF to RTF keeps text and styles, though layout may shift.
Pick the lightest format that keeps what you need. Plain text is the most portable and the easiest to search or diff. Word and RTF carry more but cost layout surprises.
What copying in Preview gives you
Apple's guide to copying text in Preview: open the PDF, choose Tools › Text Selection, drag across the text, then Edit › Copy and paste it somewhere. Holding Option while you drag copies a vertical strip, such as one column of a table.
This is ideal for a paragraph or two. For a whole document it is tedious, and line breaks usually come through as hard returns that you must clean up. A bulk extractor gives you one tidy file.
A decision guide by document type
A typed contract with a text layer: use Convert › PDF to Text and expect near-exact words. A scanned lease: use OCR › Extract text with OCR and proofread amounts and dates. A report with multi-column layout: extract and expect jumbled reading order in places, then fall back to Convert › PDF to Word if you need paragraphs. A slide deck PDF: extraction gives titles and bullets in page order, which is usually enough for notes.
One quick test saves time: extract the first three pages only (Organize › Extract pages…) and read the result before you run a hundred-page file.
Cleaning the extracted text
Extracted text often has a hard line break at every line end, hyphenated words split across lines, and page headers repeated through the body. Search for page numbers and repeated headers and delete them before you reuse the text. If a table came out as a column of values, you are better off using Convert › Table to spreadsheet or PDF to Excel.
If you plan to summarize or ask questions of a document, FileDial has Convert › Summarize and Convert › Ask a question…, which use Apple Intelligence on this Mac and write a .txt beside the file.
The documents people extract text from are often legal, medical or financial. Running locally with Wi-Fi off means the file never reaches another server. For why that matters, see our guide to online converter safety.
Questions
How do I copy text from a scanned PDF on a Mac?
A scan has no text layer, so run OCR first. In FileDial, OCR › Extract text with OCR writes a .txt, and OCR › Make searchable adds a hidden text layer to the PDF so Preview can select it. Then copy as usual with Tools › Text Selection.
Why does extracted text have strange line breaks?
PDFs store lines of text positioned on the page, not paragraphs, so extraction often keeps a break at each line end. Hyphenated words can also appear split. Use find and replace to join lines, or use Convert › PDF to Word if you want paragraphs rebuilt.
Is PDF to Text accurate?
For a PDF with a real text layer it copies the words exactly, so the main issues are reading order and line breaks. For scans it depends on OCR, which can misread similar characters, so proofread numbers and names against the page before relying on them.