Extract text
How to count the words in a PDF
By DeckleUpdated
To count the words in a PDF, extract its text with PDF to text, then open the TXT file in a program that counts words, such as Word, Google Docs or any text editor with a word count. PDF viewers rarely show a word count, but once the text is in a plain file, the count takes seconds. Choose only the pages you want counted, and leave out covers, contents and references if your limit excludes them.
Pages become plain text
Selected PDF pages
Order chosen in the review
UTF-8 text
Words without page formatting
Page boundary: U+000C or a blank line
How to count words in a PDF
- Open PDF to text and choose your PDF, or drop it onto the page.
- Under Pages and order, keep All pages for the whole document, or choose Choose pages and type the pages that count, such as
3-18. - Under Between pages, choose One continuous document so no page-break characters end up in the count.
- Choose Extract text.
- Choose Copy text on the result screen. If the document is too long for the preview, choose Download text instead, open the file and copy all the text.
- Paste it into a word counter: in Google Docs, choose Tools and then Word count; in Word, the count shows in the status bar at the bottom of the window.
The result screen tells you how many pages were read. If the Verified: line says there was no selectable text to compare, the PDF is a scan and the count would be zero; it needs OCR first.
Count only the words that matter
Word limits for essays, applications and articles usually exclude some parts of the document. Because PDF to text lets you pick pages, you can leave those out before counting.
| Part of the document | Usually counted? | How to leave it out |
|---|---|---|
| Title page | No | Start your page list after it, such as 2-20 |
| Contents and abstract | Often not | Skip those pages in the list |
| Main text | Yes | Include |
| Footnotes on the same page | Depends on the rules | Delete them from the TXT file before counting |
| Reference list or bibliography | Usually not | End the list before it |
| Appendices | Usually not | End the list before them |
Example: a 24-page dissertation chapter has a title page, a one-page abstract, main text on pages 3–21 and references on 22–24. Type 3-21 in Page numbers in output order and the count covers the main text only.
Why counts differ between programs
Different counters disagree by a few per cent on the same text. The differences come from small rules:
- Hyphenated words such as
well-knowncount as one word in most programs and two in some. - Numbers and symbols like
£1,020or§ 4.2are counted differently. - Headers, footers and page numbers repeat on every page and are part of the extracted text. On a 20-page document that can add 60 or more words. Delete them from the TXT file before counting if your limit excludes them.
- Words split by a line-end hyphen in the PDF, such as
docu-andment, may count as two.
If your limit is strict, ask which program the person setting the limit counts with, and use the same one.
When the count looks wrong
A count far lower than expected usually means some pages are images. Scanned pages, photographed tables and screenshots contribute nothing, because there is no text to extract. Why is text missing from some pages shows how to find them.
A count that looks high may include repeated headers and footers, or text from tables, figure labels and captions. Skim the TXT file before trusting the number.
Tips
- Count from the source if you have it. A Word or Google Docs original gives the exact count without extraction.
- Remove repeated running heads. Search for the header text in the TXT file and replace it with nothing.
- Check a page you know. Count one page by hand or against the source to make sure the extract is complete.
- Use page ranges for chapter counts. Extract each chapter's pages separately to get a count per chapter.
Frequently asked questions
Can I see the word count of a PDF without converting it?
Most PDF viewers do not show one. Extracting the text first is the reliable way to get a count.
Why is the word count zero?
The PDF is a scan or image, so it has no text to count. Run OCR to add a text layer, then extract again.
Does the count include footnotes?
Yes, if they are on the pages you extract, because they are part of the page text. Remove them from the TXT file if your rules exclude them.
Can Deckle count characters as well?
Deckle does not show a count itself. Google Docs and Word both report characters alongside words once you paste the text.
Related tools
- PDF to text to extract the pages you want to count
- Split PDF to save the counted section as its own PDF
- Organize PDF pages to remove pages you do not want counted