` element, then from the first `h1`, then from the file name, and finally with "Document". The value is written into the Word file's properties, so the document carries a name of its own and not only a file name. ### Step 4: Read the Conversion Notes The notes name what the converter left out, and images are the usual entry. A note counts images with remote addresses separately from embedded images in a format the converter cannot place. Read the notes before you send the file on. ### Step 5: Download the .docx Click Download. The file is named after the source file's stem, so `page.html` becomes `page.docx`, and pasted markup with no file behind it becomes `document.docx`. The body text is Calibri at 11 points, which is the default a reader expects to see. ## Which HTML Elements Become Which Word Parts? Each supported element has one Word result. The table below is the whole mapping, and anything absent from it is dropped or treated as plain text. | HTML | Word result | |---|---| | `<h1>` to `<h6>` | Heading 1 to Heading 6 styles | | `<p>` | Paragraph | | `<ul>`, `<ol>` | Bulleted and numbered lists, nesting kept to eight levels | | `<table>` | Full-width table, header row shaded and bold | | `<a href="https://...">`, `mailto:` | Word hyperlink | | `<b>`, `<strong>` | Bold | | `<i>`, `<em>`, `<cite>` | Italic | | `<u>`, `<ins>` | Underline | | `<s>`, `<del>` | Strikethrough | | `<code>`, `<kbd>`, `<samp>` | Courier New text | | `<pre>` | Shaded monospace block, one paragraph per line | | `<blockquote>` | Indented italic paragraph with a left border | | `<hr>` | Horizontal rule | | `<br>` | Line break inside the paragraph | | `<figcaption>` | Italic paragraph | | `<dt>` | Bold paragraph | | `<img src="data:image/png;base64,...">` | Picture, capped at 550 px wide | | `<img src="https://...">` | Left out and counted in the notes | | `<script>`, `<style>`, `<iframe>`, `<svg>`, form controls | Dropped | | `class`, inline `style` | Ignored | Elements hidden with the `hidden` attribute or with `aria-hidden="true"` are dropped along with their text. Content inside `nav` and `footer` is kept, because a pasted fragment's footer is usually part of what the reader wants. Tables get two extra touches. A row with fewer cells than the widest row is padded with empty cells, so the table stays rectangular in Word, and the header row is marked as a header so Word repeats it when the table runs past a page break. ## What Happens to Links? A link survives when Word can follow it. An `http` or `https` href and a `mailto:` href become real Word hyperlinks, clickable in the document and recorded in the file's relationships. A relative link such as `/pricing` becomes plain text, because a Word file has no base address to resolve it against. A `javascript:` href becomes plain text for the same reason a document should not carry executable fragments around. Several pieces of text that share one href become a single hyperlink rather than several, so a link with bold words inside stays one clickable link. Make the links absolute in the source markup when the page you copied uses relative paths, or add them in Word afterwards. ## Why Are Remote Images Left Out? Only images already embedded in the markup are placed in the document. An `<img>` whose `src` is a `data:` URL in PNG, JPEG, GIF, or BMP form becomes a picture. Every other image is left out and counted. An image with a remote address stays out because the page fetches nothing during a conversion. Downloading images would mean sending requests on your behalf while you convert a private document, and the page does not do that. You can watch the browser's network panel during a conversion and see that it stays quiet. A placed image keeps its pixel size, read from the file header, capped at 550 px wide with the height scaled to match. When the header cannot be read, a default size is used and the picture can be resized in Word like any other. The `alt` attribute is carried into the picture's alternative text. An embedded image in a format the converter does not place, such as SVG or WebP, is counted in its own note. Convert those to PNG first when the picture matters, then embed the result and convert again. ## What Does CSS Do During the Conversion? CSS does nothing. Classes, inline `style` attributes, floats, grids, multi-column layouts, and media queries have no effect on the Word file, and a `<style>` block is dropped along with its rules. The reason is a difference in kind. A stylesheet describes a page that reflows at any width, while a Word document is a fixed page built on named styles such as Heading 1, Normal, and List Paragraph. No mapping between the two is faithful, so the converter records what each element means instead of guessing at how it should look. A document built from real Word styles is also the document a colleague can restyle. Changing Heading 2 once updates every Heading 2 in the file, which is impossible when headings arrive as loose bold text with a hardcoded size. When you need the page's appearance rather than its text, take a screenshot or print the page to PDF. A Word file is the wrong container for a visual copy of a web page. ## How Do You Get Cleaner Input? Feed the converter the article, not the website. The page reads tags, so markup that names its structure honestly produces a document that needs no cleanup. - **Paste the body, not the whole page.** A saved page carries its navigation and its footer into the document, because those are content to the converter. Copy the `<article>` or the main column instead. - **Export from the CMS.** A content editor that can show the stored HTML of a post gives you the cleanest possible input, with no site chrome around it. - **Rebuild layout tables.** An email template built from tables for layout converts into Word tables, one per layout box. Lift the message body out first, or accept that the tables need deleting in Word. - **Use real headings and lists.** A `<div class="title">` is a paragraph to the converter. An `<h2>` is a heading. The difference decides whether Word's navigation pane can see the structure. - **Tidy the markup first.** [HTML Formatter](/developers/html-formatter/) indents the source so the structure is visible before you paste it. Broken markup still converts, because unclosed tags are closed the way a browser would close them, but a nesting mistake is easier to spot in formatted source. ## FAQ ### Is HTML to DOCX free to use? Yes. The converter is free, needs no account, and converts one document at a time in the browser. ### Is my HTML uploaded anywhere? No. The markup is parsed and the Word file is built inside the tab. Nothing is uploaded, nothing is fetched, and you can confirm both in the browser's network panel while you convert. ### Can I drop an .html file instead of pasting? Yes. Drop an `.html` or `.htm` file, or pick one from the file dialog. The download is named after that file, so `report.html` becomes `report.docx`. ### Why did my images not appear in the Word file? The images pointed at a remote address, or they were embedded in a format the converter does not place. Both cases are counted in the conversion notes. Embed the pictures as PNG or JPEG data URLs and convert again. ### Does the Word file look like the web page? No, and it is not meant to. The file carries the structure and the text of the page in Word's own styles. Fonts, colours, spacing, and layout come from Word's defaults. ### Can I go the other way, from Word to HTML? Yes. [DOCX to HTML](/convert/docx-to-html/) reads a `.docx` and writes semantic HTML, so the two pages make the round trip. ## Where to Go Next Convert one page first and read its notes before you convert a folder of them. When the result looks flat, the source markup was probably styled divs rather than headings. - **[HTML to DOCX](/convert/html-to-docx/)** turns pasted HTML or an `.html` file into an editable Word document. - **[DOCX to HTML](/convert/docx-to-html/)** makes the reverse trip, from a Word file to semantic HTML. - **[Markdown to Word](/convert/markdown-to-word/)** builds a `.docx` when the source is Markdown rather than HTML. - **[HTML to Markdown](/convert/html-to-markdown/)** converts a pasted fragment to Markdown for a README or a wiki. - **[DOCX to Markdown](/convert/docx-to-markdown/)** takes a Word document the other way, into Markdown. - **[HTML Formatter](/developers/html-formatter/)** indents and inspects the markup before you convert it.

--- url: https://findutils.com/guides/html-to-docx title: "Convert HTML to a Word DOCX in Your Browser" description: "Convert HTML into an editable Word .docx without an upload. Learn what the converter keeps, how links and images are treated, and why CSS has no effect." category: converters content_type: guide guide_type: subtopic cluster: data-conversion locale: en read_time: 7 status: published author: "olgunozoktas" published_at: 2026-09-11T12:00:00Z excerpt: "Turn a pasted article, an exported page, or an .html file into a real Word document. Review which elements become which Word parts, and how links and images are handled." tag_ids: ["converters", "document-management", "data-conversion", "productivity", "developer-tools"] tags: ["Converters", "Document Management", "Data Conversion", "Productivity", "Developer Tools"] primary_keyword: "convert html to docx" secondary_keywords: ["html to word", "html to docx converter", "html to word document", "html file to docx", "web page to word document"] tool_tag: "html-to-docx" related_tool: "html-to-docx" related_tools: ["html-to-docx", "docx-to-html", "markdown-to-word", "html-to-markdown", "docx-to-markdown"] og_image: "/images/content/guides/html-to-docx-cover-20260911.webp" image_alt: "Loose blank paper slips and plain ribbons on a workbench are gathered into one bound volume with a stiff cover beside a bookbinder's press." updated_at: "2026-09-11T19:45:46Z" --- Convert HTML to DOCX by reading the structure of the markup, mapping every element to the matching Word construct, and writing a real `.docx` file. FindUtils [HTML to DOCX](/convert/html-to-docx/) does this in your browser. The page does not upload the markup, and it fetches nothing while it converts. This guide explains what the converter reads, which HTML elements become which parts of a Word document, how links and images are treated, and why a stylesheet changes nothing about the result. The page is in beta and converts structure only, never a page design. ## What Ends Up in the Word File? The Word file holds the document structure of the HTML: headings, paragraphs, lists, tables, links, inline formatting, code blocks, quotes, and rules. Everything a stylesheet controls stays behind. The converter accepts a full HTML document or a bare fragment. When the markup contains a ``, only the body is read, so the `` and everything a browser keeps there never reaches the document. - **Headings:** `h1` to `h6` become Word's Heading 1 to Heading 6 styles, not bold text that looks like a heading. - **Lists:** `ul` and `ol` become bulleted and numbered Word lists, and nesting is kept to eight levels. - **Tables:** a `table` becomes a Word table at full width, with a header row taken from `` or from a row of all `` cells. - **Links:** `http`, `https`, and `mailto` targets become real Word hyperlinks. - **Inline formatting:** bold, italic, underline, strikethrough, and inline `code` in Courier New. - **Blocks:** `pre` becomes a shaded monospace block, `blockquote` an indented italic paragraph with a left border, `hr` a horizontal rule. - **Dropped:** fonts, colours, sizes, alignment, floats, columns, and everything else a stylesheet sets. FindUtils parses the markup with its own HTML tree parser, the same one its Site to Markdown converter uses, so no browser DOM is involved. The block model that comes out is written into a `.docx` by the `docx` package, an open-source library under the MIT licence. The result is an ordinary Word document that Word and LibreOffice Writer open and edit. ## How to Convert HTML to DOCX Online Open the page, give it markup, name the document, read the notes, and download. The conversion runs when you click Download. ### Step 1: Open the Converter Open FindUtils [HTML to DOCX](/convert/html-to-docx/). The page pairs an editor for the markup with a sandboxed preview frame that shows the structure the Word file will hold, with no CSS, so you can see what will be converted before you download. ### Step 2: Paste the HTML or Drop a File Paste markup into the editor, or drop an `.html` or `.htm` file onto the page and pick it from the file dialog. A complete document and a copied fragment both work. Empty input shows a plain tip rather than an error. ### Step 3: Set the Document Title The **Document title** field is prefilled from the `` element, then from the first `h1`, then from the file name, and finally with "Document". The value is written into the Word file's properties, so the document carries a name of its own and not only a file name. ### Step 4: Read the Conversion Notes The notes name what the converter left out, and images are the usual entry. A note counts images with remote addresses separately from embedded images in a format the converter cannot place. Read the notes before you send the file on. ### Step 5: Download the .docx Click Download. The file is named after the source file's stem, so `page.html` becomes `page.docx`, and pasted markup with no file behind it becomes `document.docx`. The body text is Calibri at 11 points, which is the default a reader expects to see. ## Which HTML Elements Become Which Word Parts? Each supported element has one Word result. The table below is the whole mapping, and anything absent from it is dropped or treated as plain text. | HTML | Word result | |---|---| | `<h1>` to `<h6>` | Heading 1 to Heading 6 styles | | `<p>` | Paragraph | | `<ul>`, `<ol>` | Bulleted and numbered lists, nesting kept to eight levels | | `<table>` | Full-width table, header row shaded and bold | | `<a href="https://...">`, `mailto:` | Word hyperlink | | `<b>`, `<strong>` | Bold | | `<i>`, `<em>`, `<cite>` | Italic | | `<u>`, `<ins>` | Underline | | `<s>`, `<del>` | Strikethrough | | `<code>`, `<kbd>`, `<samp>` | Courier New text | | `<pre>` | Shaded monospace block, one paragraph per line | | `<blockquote>` | Indented italic paragraph with a left border | | `<hr>` | Horizontal rule | | `<br>` | Line break inside the paragraph | | `<figcaption>` | Italic paragraph | | `<dt>` | Bold paragraph | | `<img src="data:image/png;base64,...">` | Picture, capped at 550 px wide | | `<img src="https://...">` | Left out and counted in the notes | | `<script>`, `<style>`, `<iframe>`, `<svg>`, form controls | Dropped | | `class`, inline `style` | Ignored | Elements hidden with the `hidden` attribute or with `aria-hidden="true"` are dropped along with their text. Content inside `nav` and `footer` is kept, because a pasted fragment's footer is usually part of what the reader wants. Tables get two extra touches. A row with fewer cells than the widest row is padded with empty cells, so the table stays rectangular in Word, and the header row is marked as a header so Word repeats it when the table runs past a page break. ## What Happens to Links? A link survives when Word can follow it. An `http` or `https` href and a `mailto:` href become real Word hyperlinks, clickable in the document and recorded in the file's relationships. A relative link such as `/pricing` becomes plain text, because a Word file has no base address to resolve it against. A `javascript:` href becomes plain text for the same reason a document should not carry executable fragments around. Several pieces of text that share one href become a single hyperlink rather than several, so a link with bold words inside stays one clickable link. Make the links absolute in the source markup when the page you copied uses relative paths, or add them in Word afterwards. ## Why Are Remote Images Left Out? Only images already embedded in the markup are placed in the document. An `<img>` whose `src` is a `data:` URL in PNG, JPEG, GIF, or BMP form becomes a picture. Every other image is left out and counted. An image with a remote address stays out because the page fetches nothing during a conversion. Downloading images would mean sending requests on your behalf while you convert a private document, and the page does not do that. You can watch the browser's network panel during a conversion and see that it stays quiet. A placed image keeps its pixel size, read from the file header, capped at 550 px wide with the height scaled to match. When the header cannot be read, a default size is used and the picture can be resized in Word like any other. The `alt` attribute is carried into the picture's alternative text. An embedded image in a format the converter does not place, such as SVG or WebP, is counted in its own note. Convert those to PNG first when the picture matters, then embed the result and convert again. ## What Does CSS Do During the Conversion? CSS does nothing. Classes, inline `style` attributes, floats, grids, multi-column layouts, and media queries have no effect on the Word file, and a `<style>` block is dropped along with its rules. The reason is a difference in kind. A stylesheet describes a page that reflows at any width, while a Word document is a fixed page built on named styles such as Heading 1, Normal, and List Paragraph. No mapping between the two is faithful, so the converter records what each element means instead of guessing at how it should look. A document built from real Word styles is also the document a colleague can restyle. Changing Heading 2 once updates every Heading 2 in the file, which is impossible when headings arrive as loose bold text with a hardcoded size. When you need the page's appearance rather than its text, take a screenshot or print the page to PDF. A Word file is the wrong container for a visual copy of a web page. ## How Do You Get Cleaner Input? Feed the converter the article, not the website. The page reads tags, so markup that names its structure honestly produces a document that needs no cleanup. - **Paste the body, not the whole page.** A saved page carries its navigation and its footer into the document, because those are content to the converter. Copy the `<article>` or the main column instead. - **Export from the CMS.** A content editor that can show the stored HTML of a post gives you the cleanest possible input, with no site chrome around it. - **Rebuild layout tables.** An email template built from tables for layout converts into Word tables, one per layout box. Lift the message body out first, or accept that the tables need deleting in Word. - **Use real headings and lists.** A `<div class="title">` is a paragraph to the converter. An `<h2>` is a heading. The difference decides whether Word's navigation pane can see the structure. - **Tidy the markup first.** [HTML Formatter](/developers/html-formatter/) indents the source so the structure is visible before you paste it. Broken markup still converts, because unclosed tags are closed the way a browser would close them, but a nesting mistake is easier to spot in formatted source. ## FAQ ### Is HTML to DOCX free to use? Yes. The converter is free, needs no account, and converts one document at a time in the browser. ### Is my HTML uploaded anywhere? No. The markup is parsed and the Word file is built inside the tab. Nothing is uploaded, nothing is fetched, and you can confirm both in the browser's network panel while you convert. ### Can I drop an .html file instead of pasting? Yes. Drop an `.html` or `.htm` file, or pick one from the file dialog. The download is named after that file, so `report.html` becomes `report.docx`. ### Why did my images not appear in the Word file? The images pointed at a remote address, or they were embedded in a format the converter does not place. Both cases are counted in the conversion notes. Embed the pictures as PNG or JPEG data URLs and convert again. ### Does the Word file look like the web page? No, and it is not meant to. The file carries the structure and the text of the page in Word's own styles. Fonts, colours, spacing, and layout come from Word's defaults. ### Can I go the other way, from Word to HTML? Yes. [DOCX to HTML](/convert/docx-to-html/) reads a `.docx` and writes semantic HTML, so the two pages make the round trip. ## Where to Go Next Convert one page first and read its notes before you convert a folder of them. When the result looks flat, the source markup was probably styled divs rather than headings. - **[HTML to DOCX](/convert/html-to-docx/)** turns pasted HTML or an `.html` file into an editable Word document. - **[DOCX to HTML](/convert/docx-to-html/)** makes the reverse trip, from a Word file to semantic HTML. - **[Markdown to Word](/convert/markdown-to-word/)** builds a `.docx` when the source is Markdown rather than HTML. - **[HTML to Markdown](/convert/html-to-markdown/)** converts a pasted fragment to Markdown for a README or a wiki. - **[DOCX to Markdown](/convert/docx-to-markdown/)** takes a Word document the other way, into Markdown. - **[HTML Formatter](/developers/html-formatter/)** indents and inspects the markup before you convert it.