Convert HTML to Word
HTML and a Word document describe the same things in different vocabularies — headings, paragraphs, tables, lists, emphasis. That correspondence is why this conversion works well. What has no counterpart is everything a stylesheet says, and that is why the result arrives looking nothing like the page.
Use this without the search next time. Prathom Workbench puts Prathom's tools in your toolbar.
Add to Chrome — freeDrop your HTML file here, or click to browse
Up to 50 MB. Deleted automatically after 30 minutes.
What it does
- Headings become Word heading styles, so the navigation pane works
- Tables become real tables, lists become real lists
- Link addresses preserved, not flattened to their text
- No watermark and no sign-up
How to use HTML to Word
- 1
Add the HTML file
A page saved from a browser, an export from a CMS, an email template, or documentation generated by a build tool. One file — this does not fetch anything the page refers to.
- 2
Convert
The markup is parsed into a document model and written out as OpenXML. The stylesheet is not consulted.
- 3
Restyle it in Word
Because the structure is real, applying a Word theme or a house template restyles the whole document at once. That is the intended workflow, and it is faster than it sounds.
How it works
The markup is parsed into a document model — headings, paragraphs, emphasis runs, lists, tables, links — and that model is written out as OpenXML.
There is one implementation detail worth recording, because it is the kind of thing that silently breaks. An HTML file loads into the word processor's web module rather than its document module, and that module's export filters are a different set: asking for a .docx the ordinary way fails outright with "no export filter found". The filter has to be named explicitly. It affects .docx, .doc and .rtf, and not .odt or .txt, for no reason visible from the outside.
What maps, and what does not
Maps cleanly: headings to heading styles, paragraphs, ordered and unordered
lists, tables with their cells, strong and em to the Strong and Emphasis
character styles, and links with their addresses intact.
Does not map: everything in the CSS. Fonts, colors, spacing, borders,
backgrounds, widths, floats, grid and flex layout, anything responsive. Also
scripts, forms, video, and any layout produced by nested div elements, which
have no meaning to convert.
Maps into a broken state: images, as described above.
The dividing line is not arbitrary. It is exactly the line between semantic markup and presentation, which HTML has spent thirty years separating.
The workflow this is good for
Take a page, convert it, and then apply your own template. You get the content with a correct outline in a document where the styles are yours rather than a website's — which is usually what someone asking for "this page as a Word document" actually wants.
It is much faster than the alternative most people reach for, which is selecting the rendered page and pasting it. That route brings the CSS across as thousands of pieces of direct formatting welded onto the text, and stripping it afterwards takes longer than the conversion saved.
When to convert to PDF instead
If the point is to preserve how the page looks, this is the wrong tool entirely — the appearance is precisely what does not survive. Print the page to PDF from the browser instead, which renders the CSS and captures the images because the browser already has them.
Convert to .docx when you want the words and the structure in something editable, and accept that the design stays behind.
Examples
A briefing page saved from a browser
Verified by unzipping the result: one Heading 1, two Heading 2, a real table element, a genuine list, and the link's address preserved as a document relationship. The document is ten times the size of the source because a .docx carries a style catalog and the HTML carried none.
A page with a stylesheet and an image
The font and the color are gone, because they lived in the CSS. The image is worse than gone: the document keeps a reference to a file it cannot reach, so Word shows a placeholder where the picture was. Only the .html file is uploaded, so there was never an image to embed.
Frequently asked questions
Why does the result not look like the web page?
Because a stylesheet is not part of the document in the sense Word understands. HTML separates what the content *is* from how it *looks*, and only the first half has a Word equivalent. A heading arrives as Heading 1, which Word then renders in whatever your template says a Heading 1 looks like — not in the font the site used. That is a faithful conversion of the structure and a total loss of the design.
What happens to images?
They are not carried across. A web page refers to its images by address rather than containing them, and only the HTML file is uploaded here, so there is nothing to embed. The reference survives and points at a file that is not there, which Word displays as a broken-image placeholder. Delete the placeholders, or insert the images yourself afterwards.
Do bold and italic survive?
Yes, and in a better form than you might expect. A `strong` tag arrives as Word's Strong character style and an `em` tag as Emphasis, rather than as direct bold and italic formatting. The text renders the same way, and because it is a style rather than a manual override, changing the style changes every instance at once.
Can I convert a live web page by giving you its address?
No — this takes a file, not a URL. Save the page first: in most browsers, Ctrl+S or Cmd+S with "Web page, HTML only" selected. Choosing "complete" instead gives you a folder of assets that this cannot use anyway, since the images still would not travel with the single file you upload.