Open DOCX Files Online

1. 1

   Choose a Word document

   Drag a .docx or .dotx file onto the drop area, or press Browse. The file is read on your device — it is never uploaded. Anything that is not a Word document is refused straight away, with a note saying what it is and where to convert it.

2. 2

   Read it with its formatting

   The document appears with its own page size, headings, tables, lists and pictures. Long documents open at the first page and the rest are laid out only as you scroll to them, so a three-hundred-page file is readable immediately.

3. 3

   Navigate, search and zoom

   Use the outline to jump to any heading, type in the find box to highlight every occurrence, step through matches, set the zoom or fit the page to the window, and go straight to any page number.

4. 4

   Take it away

   Read the document's properties and counts, print it with its layout intact, or save its text as TXT, HTML or Markdown. Nothing is stored — close the tab and the file is gone.

Drop a .docx or .dotx file and it opens where you are: its own page layout, an outline you can jump around with, a find box that highlights every match, zoom, page navigation, printing, and three ways to take the text with you. Nothing is uploaded and nothing is stored.

• No Word required: Open DOCX and DOTX files on any modern device without installing Microsoft Word, signing in, or waiting for a file to upload.
• Real page layout: The document is laid out as pages, at the page size the author set, with headings, tables, lists, pictures and running headers where they belong.
• Outline navigation: Every heading in the document becomes an entry you can click to jump straight to that section, however long the file is.
• Find in document: Type a word and every occurrence is highlighted at once, with a running count and next and previous buttons.
• Zoom and page controls: Set a zoom level, fit the page to the window width or height, or type a page number and go straight there.
• Document properties: Author, title, created and modified dates, page count, and word, character and paragraph counts measured from the document itself.
• Take it away: Print with the layout intact, or save the text as TXT, HTML or Markdown — all free, on every plan.

Most of the time you are handed a Word document you only need to read it. Installing a word processor, signing into an account, or uploading somebody else's contract to a stranger's server is a great deal of ceremony for that. A viewer that runs on your own device removes all three at once — and because it all runs on your device, not our servers, it is the only arrangement that is honestly private.

Benefits of an On-Device Viewer

• Nothing is uploaded: The file is read where it already is. There is no server copy to be retained, indexed, breached or subpoenaed.
• No quota: No daily document limit, and no watermark on anything you print or save.
• Works offline: Once the page has loaded, it keeps working with no connection at all — useful on a train, and proof that nothing is being sent.
• Opens immediately: There is no upload to wait through, so a 16 MB document starts rendering as fast as your disk can hand it over.
• Reads what your device reads: The size ceiling is your own machine's memory rather than a number someone chose to price a plan.
• Nothing is kept: Close the tab and the document is gone from memory. Reopening it means picking the file again, which is the correct trade for a tool that stores nothing.

A long document is mostly a navigation problem. Scrolling through two hundred pages to find the clause you remember is the reason people open files in a word processor at all, so the viewer gives you three routes to the same place and makes all of them instant.

Three Ways to Move Around

• The outline: Every heading the document defines becomes an entry, indented by level. Clicking one scrolls to that heading. It is built from the document's own styles, so it matches the navigation pane Word would show.
• The page box: Type any page number and the view jumps to the top of that page. The counter beside it updates as you scroll, so you always know where you are.
• Find: Search for a phrase and step through the matches. This is the fastest route when you remember the wording but not the section.
• Zoom and fit: Fit to width fills the window with the page; fit to page shows a whole page at once, which is how you check where a page break actually falls.

When a Document Has No Outline

• A document whose author used bold text instead of heading styles has no outline to show — the viewer says so rather than showing an empty panel.
• Headings declared by outline level rather than by style name are recognised too, which is how documents exported from other word processors keep their structure here.
• A custom style based on Heading 1 is treated as a level 1 heading, because that is what basing a style on another one means.
• Levels are reported as written. A document that jumps from Heading 1 to Heading 4 shows exactly that, rather than inventing the levels in between.

Search inside a Word file is harder than it looks, and the two ways it usually goes wrong are both invisible. Word splits a word across separate runs whenever formatting changes inside it, so a word that turns bold half way through is stored as several pieces — a search that looks at one piece at a time simply never finds it. And a running header repeated on every page will report one match per page for a phrase that appears in the document once.

What the Find Box Does

• Finds split words: Matching runs across formatting boundaries, so a partly-bold or partly-italic word is one match, not none.
• Ignores running headers: Headers and footers repeat on every page. Counting them would turn one phrase into three hundred matches you cannot navigate.
• Counts the way you would: Overlapping occurrences count once, matching what every browser's own find bar reports.
• Case and whole word: Both are switches. Whole-word search finds "cat" without finding "concatenate".
• Highlights all at once: Every match is marked in place, so you can see how they are distributed before stepping through them.

Every Word file carries information about itself that a reader almost never sees: who wrote it, when it was created and last changed, which application wrote it, and how long it was edited for. The viewer shows all of it, alongside counts measured from the document in front of you rather than copied from what the authoring application recorded at its last save.

What the Properties Panel Shows

• Authorship: Title, author, last modified by, company and category, exactly as stored in the file.
• Dates: Created and last modified, plus the revision number and total editing time the authoring application recorded.
• Measured counts: Words, characters with and without spaces, paragraphs, headings, tables and pictures — counted from the document, so they are current even if the file was edited elsewhere.
• Word counting that handles CJK: Japanese, Chinese and Korean text has no spaces in it, so each character counts as a word. Splitting on whitespace would report a whole page of Japanese as one word.
• Revisions and comments: A note when the document contains tracked changes or comments, so you know before you rely on what is shown.

Reading is usually not the end of it. Printing has to use the document's own page layout rather than a picture of the window, and saving the text has to produce the whole document rather than the part that happens to be on screen — which is why all three exports are built from the parsed document, not scraped from the view.

Four Ways to Take It With You

• Print: Uses the document's page size and margins, so page breaks fall where the author put them.
• Plain text (.txt): The whole document as text, with table cells separated by tabs. The right choice for pasting somewhere else.
• HTML (.html): One self-contained file that keeps headings, emphasis, lists and tables, with no scripts, no web fonts and nothing that loads from anywhere. It opens in any browser and works offline.
• Markdown (.md): Headings as hashes, lists as dashes, and tables padded to a fixed width so they render rather than collapsing. Good for a wiki, a README or a notes app.

A .docx file is not a document in the way a PDF is. It is a ZIP archive containing a folder of XML files: the text and its structure in one part, the styles in another, the numbering rules in a third, the pictures alongside, and a small file of properties describing the whole thing. Opening one means unpacking the archive, reading the markup and deciding what it should look like — pagination included, because the file does not record where the pages break.

What Happens When You Open a File

• Identify: The first few bytes say what the file really is. A PDF renamed to .docx is caught here and refused with a message naming the format, rather than failing later behind a spinner.
• Unpack: The archive is opened and the document, styles, numbering and properties parts are read out.
• Parse: The markup is walked once to produce the text, the outline, the counts and the properties. This happens on a background thread, so the page stays responsive throughout.
• Lay out: The document is turned into pages at the size the author set, with fonts, tables, pictures and headers in place.
• Show the first page: Pages you are not looking at are left unlaid-out until you scroll to them, which is why the time to the first page barely changes between a two-page memo and a three-hundred-page manuscript.

Some numbers about the format and about this viewer, all of them measured rather than estimated. They explain both what is easy and what is genuinely hard about showing somebody else's Word document faithfully.

Key Statistics

• DOCX has been Word's default format since Word 2007, replacing the binary .doc format used from 1997.
• A .docx file is a ZIP archive: the text lives in word/document.xml, and a 16 MB file can hold more than a million characters of text.
• The format does not store a page count. Pagination is a result of laying the document out, which is why two applications can disagree about how many pages a file has.
• A 300-page, 16.6 MB test document opens and lays out in about 0.7 seconds on a desktop machine, with no main-thread pause long enough to be felt.
• Searching all 1.1 million characters of that document takes under 50 milliseconds, so the highlight count updates as you type.
• Holding that document open costs about 7 MB of memory once the source file has been released, because pages off screen are not laid out.

A few habits make Word documents easier to read here and everywhere else — and a couple of them are about knowing what a viewer can and cannot tell you.

Getting the Best Result

• Use real heading styles rather than bold text. It is what builds the outline here, the navigation pane in Word, and the bookmarks in an exported PDF.
• Expect fonts to substitute. A document set in a font your device does not have will be shown in the closest available one, which can change where lines wrap.
• Check the properties panel before quoting a document. It tells you whether the file contains tracked changes, which are shown accepted.
• Export to HTML rather than text when you need the structure. Plain text loses headings, lists and table shape; the HTML export keeps all three in one self-contained file.
• Convert older formats first. A .doc, .odt or .rtf file is a different structure, not a variation on DOCX, and converting it is a real step rather than a rename.
• For a document you will read repeatedly, keep the file — nothing is stored here, which is the point, so each visit starts by choosing it again.

Common Use Cases

• Reading an attachment: Open a document somebody emailed you without installing anything or handing the file to a third party.
• Reviewing a contract: Search for a clause, jump to the section, and check the properties for tracked changes before you rely on the wording.
• Checking a submission: Confirm a manuscript's page count, word count and structure against what was asked for.
• Working on a phone or a locked-down laptop: Read a Word file on a device with no word processor installed and no permission to install one.
• Pulling text out: Save the document as Markdown or HTML to move its content into notes, a wiki or a static site.
• Reading something confidential: Open a document that must not be uploaded anywhere, and verify that by disconnecting first.

Everything above happens on the machine in front of you. The document is read from your own disk, unpacked and parsed on a background thread so typing and scrolling never stall, and laid out page by page as you reach them. No part of the file is sent anywhere, and there is no server that could keep a copy even if one were sent.

What That Buys You

• Privacy that is structural: Not a promise about what is done with an upload — there is no upload.
• Speed that scales with your machine: A faster device opens documents faster, and a large document is limited by your own memory rather than by a plan.
• It keeps working offline: Load the page once and it opens documents with the network switched off.
• Nothing to clean up: No files to delete from a server later, and no retention policy to read.

Documents are read and laid out on your own device.

• Word to PDF
• Word to HTML
• Word to Markdown
• Word to text
• Word metadata
• Compress Word document