Paste HTML and download a real Word document — headings, lists, tables and links intact, as things Word can actually edit.
Get a .docx with real Word heading styles, lists and tables. Runs in your browser, nothing uploaded.
There are two ways to do this conversion and they disagree about what "worked" means. One tries to carry the page’s appearance across: it reads the CSS and paints the result into the document. What you get looks roughly right and has no structure — every heading is bold text at a size, so Word’s navigation pane is empty, a table of contents generates nothing, and editing it means fighting the formatting.
This one throws the presentation away on purpose and keeps the skeleton. An <h2> arrives as a paragraph styled Heading 2, a <table> as a real Word table with rows and cells, a list as a list, a link with its URL. That is the half Word can genuinely represent, and it is the half that makes the file editable rather than merely openable. The page tells you what was dropped instead of leaving you to find out after you send it.
Paste your HTML above, or upload an .html file. A saved page, an email export, or a fragment all work.
The preview shows the structure that survived — headings, lists, tables, quotes and links — before you commit to a download.
Download the .docx. It opens in Word, Google Docs, Pages and LibreOffice, with real heading styles so the navigation pane and table-of-contents work.
A team keeps its runbook as an HTML page, and once a quarter a compliance reviewer asks for it as a Word file with tracked changes. The first attempt was print-to-PDF then a PDF-to-Word conversion, which produced a document made of text boxes: the reviewer could not put the cursor in a paragraph and type.
The runbook is eight headings, four procedure lists, a table of escalation contacts and a handful of links. Converted here, all four shapes arrive as themselves — the headings as Heading 1 and 2 so the reviewer can jump between sections, the contacts as a real table they can add a column to, the links still clickable. What does not arrive is the page’s own styling: the brand font, the coloured callout boxes, the two-column layout. That is the trade, and for a document someone has to comment on, structure beats appearance every time. If appearance is what matters, print to PDF instead — a PDF is meant to look identical and not meant to be edited, which is the opposite choice, made honestly.
No, and no converter can do it properly. A standalone page styles itself with flexbox, grid, custom properties and media queries, and Word has no model for any of them. A converter that tries produces a document that looks approximately right and has no usable structure. This one keeps the structure and tells you the styling was dropped.
Yes — that is the point of the conversion. <h1> through <h3> map onto Word’s built-in Heading styles rather than onto bold text at a larger size. That is what makes the navigation pane populate and a generated table of contents find anything, and it is what most HTML-to-Word output gets wrong.
Images are not embedded. A .docx carries its images inside the file, so anything referenced by URL would have to be fetched and packaged, and a relative path in a pasted fragment has nothing to fetch. Their alt text is kept where it exists, so the document says what was there rather than silently losing the reference.
The Word to Markdown converter takes a .docx and gives you Markdown, which the Markdown to HTML converter turns into a page. Two steps, and the round trip keeps headings, lists and tables — the same structure this page preserves in the other direction.
No. The parse and the .docx packaging both happen locally, so the HTML is never uploaded and the document is generated on your machine. The download comes from memory, not from a server.
免費帳號的連結保留 7 天。升級付費方案後，你建立的每個頁面都會永久上線。
查看方案與價格