FileDial › Guides › Documents, spreadsheets and slides
How to extract text from a DOCX file on a Mac
By Himanshu Ranjan · Updated 8 November 2026
Select the DOCX files in Finder, open FileDial and choose Convert › Convert to plain text. You get a TXT copy beside each original with the formatting stripped. It runs offline on your Mac and handles a whole folder in one pass.
When plain text is what you want
A DOCX is a ZIP of XML files that carries styles, images, headers and revision data. If you only need the words, say for a search index, a diff, an import into a notes app or a quick read on a device with no Word, plain text is smaller and opens anywhere.
FileDial's Convert › Convert to plain text writes a new copy beside the original. Its note says nothing more than that, so expect words and paragraph breaks and no formatting.
Steps
These steps work on one document or on every document in a folder.
- Select the .docx files in Finder.
- Hold the right Option key and release on Convert, then Convert to plain text.
- Open one TXT to confirm the paragraph breaks look right.
- If the text will go into another program, check the encoding; Developer › Convert to UTF-8 makes a UTF-8 copy and detects the original encoding.
Related text tasks
Sometimes plain text is a step, not the end. Convert › Summarize writes a summary text file using Apple Intelligence, on this Mac. Convert › Ask a question… writes an answer file with the sentence it came from. Both are listed for documents and run on the Mac.
If you need to compare two versions of a document, convert both to text and use Combine › Compare two files, which writes a unified .diff and a side-by-side .html. That is a neat way to see what changed between drafts when tracked changes were not switched on.
Counting words
FileDial's catalogue has no dedicated word-count tool, so we will not pretend otherwise. A practical route is to convert to plain text and read the count from your editor or from Terminal's wc command. That works for one file or, with a wildcard, for a folder of text files.
Built-in alternative
TextEdit and Pages can both open a DOCX and save plain text. In Pages, File > Export To > Plain Text may appear in recent versions. For a few files this is fine. For a folder of files it is slow, and that is where a batch converter pays off.
Using the plain text afterwards
Plain text works with the simplest tools. Open the TXT in any editor and search across it, paste it into a form that rejects formatting, or feed it to a script. For several documents, put the TXT files in one folder and search them together with Finder's search or Spotlight.
If the text will be processed by software, check the encoding. CSV and TXT files do not declare theirs, and a wrong guess shows up as garbled accents. Developer › Convert to UTF-8 detects the original encoding and writes a UTF-8 copy, and Developer › Convert line endings to LF or CRLF matches the system you are sending to.
Common mistakes and when FileDial fits
If you need to keep headings, use Convert › Convert to Markdown, which keeps headings, lists, bold, italic and links and drops tables and images. If you need to keep tables, a spreadsheet is a better home. If you need a faithful read-only copy, use To PDF › Convert to PDF.
For scanned documents, plain text conversion has nothing to read, because a scan is an image. In that case run OCR first; our OCR PDF guide covers it.
Plain text throws away headings, lists and emphasis. If structure matters, choose the converter that keeps it. Convert › Convert to Markdown keeps headings, lists, bold, italic and links. Convert › Convert to HTML keeps more and gives you a web page. Convert › Convert to plain text is the right one when you want nothing but the words, for example to count or search them.
A DOCX is a ZIP package of XML files, so in principle you can unzip it and read document.xml. In practice that gives you markup and fragments, which is a poor way to read text. A converter that understands the format is much less error-prone.
Expecting text from a scanned document is the most frequent. A scan is a picture, and a document made of scanned images has no words to extract until OCR has been run. Another is converting a document with tracked changes and being surprised by what is included: run Privacy › Accept changes and remove comments first if you want the final text only. And do not assume that headers, footers and text boxes appear where you expect; read the result.
If the text goes into another system and accents look wrong, the cause is encoding, so use Developer › Convert to UTF-8 on the result.
Questions
Does extracting the text change my DOCX?
No. The TXT file is saved beside the original, which is not modified. Every result also has Undo, so if the output is not what you expected, you can remove it and try another tool such as Convert › Convert to Markdown, which keeps headings and lists.
Will images and tables come through in the text file?
Plain text holds neither images nor table layout. You get the words and paragraph breaks. If you need table data, copy the table into a spreadsheet and use Convert › Convert to CSV, or choose Convert › Convert to Markdown if you only need headings and lists.
Can FileDial count the words in my documents?
There is no dedicated word-count tool in the catalogue, so we do not claim one. A practical route is to convert to plain text with Convert › Convert to plain text and then use your editor's word count, or Terminal's wc command on the TXT files.