DOCX formatting cleanup — FAQ

No. The cleaner runs entirely inside your browser tab — there is no backend and no upload step. You can even switch off your network after the page loads and it still works. The privacy page explains how to watch the Network tab and verify it yourself.

Never. DocxCleaner only removes formatting properties, empty paragraphs (with no text), and style references — it does not edit, move or re-order text. Before the download is offered, the cleaned document is re-opened and every paragraph is compared character-for-character against the original; if anything differs, you get your original file back instead.

Formatting that was applied to a specific stretch of text rather than to a style: per-run bold, italic, underline, strikethrough, color, font, size, highlight and small caps. The cleaner removes these in paragraphs that use a named style (Headings, Title, Quote…) because the style should be in charge there. Direct bold and italic inside Normal paragraphs are kept — those often carry real emphasis.

In styled paragraphs — headings, titles, quotes — per-run character formatting is treated as leftover cruft and stripped, so the style shows through. If your headings use italics meaningfully, turn off “Strip direct character formatting”. Normal-paragraph emphasis is never touched by the default rules.

Yes. Every file inside the .docx other than word/document.xml is copied byte-for-byte: headers, footers, footnotes, images, macros and link relationships included. Tables, drawings and hyperlinks inside the main document are preserved too — the cleaner only removes formatting properties around them, and the upload never includes anything that carries content.

They are left in place. The cleaner does not accept or reject revisions, and it deliberately skips deleted-revision text when cleaning. If a document uses heavy tracked changes, they will still be there in the cleaned output, ready for the next step in your editorial process.

It works on real .docx files — the zipped XML format used by Word 2007 and later, and also exported by Google Docs (“Download as .docx”), LibreOffice and most writing apps. Old .doc binaries, PDFs and .txt files are not supported; you will get a friendly error instead of a broken file.

Browser memory. Cleaning happens in your tab, so very large manuscripts can hit memory limits, especially on phones. Up to 25 MB is comfortably safe; if your book is bigger, split it into parts, clean each part, and merge afterwards — or run the tool on a desktop browser.

Yes — free, no account, no payment, no watermark. The tool is static and serverless, so running it costs almost nothing, and it stays available as a small utility for editors.