Where Office documents keep this information
A .docx, .xlsx or .pptx is a ZIP archive of XML files, and two of those parts describe the document
rather than its contents. docProps/core.xml holds the Dublin Core fields — title,
subject, author, last modified by, created and modified dates, revision number.
docProps/app.xml holds the Office-specific ones — the application and version that
wrote the file, the company on the licence, the template, the total editing time in minutes, and the word,
character, page or slide counts.
OpenDocument files store the same idea in a single meta.xml. Either way the information
is written automatically, updated on every save, and copied with the file wherever it goes.
How to see the metadata in an Office document
- Open the document. Drop the .docx, .xlsx or .pptx onto the box above, or click to browse. OpenDocument files (.odt, .ods, .odp) work too. The file is read from your disk by the browser.
- Read who and when. The author, the name of whoever last saved the file, and the created, modified and last-printed timestamps appear first. Timestamps are shown in UTC exactly as stored.
- Check the application and statistics. Which program and version wrote the file, the company on the licence, the template it came from, total editing time, and the word, page or slide counts.
- Export the findings. Copy all properties puts everything on your clipboard; Save as CSV writes group, property and value to a spreadsheet-ready file.
What comes out
- Author and last modified by — two separate names. The second is whoever saved the file most recently, which is frequently not the person who sent it to you.
- Company and manager, taken from the Office installation that created the file.
- Created, modified and last printed timestamps, in UTC.
- Revision number — how many times the document has been saved.
- Total editing time in minutes, shown in hours as well.
- Application and version — which program wrote it, which reveals whether a file claiming to be from Word came out of LibreOffice, Google Docs or a script.
- The template the document was based on, often a company’s internal one.
- Counts — words, characters, pages, paragraphs, lines, slides, hidden slides and notes.
- Document security flags, including whether the file is password protected or read-only.
Anything present but not on that list is still shown under Other properties, so nothing is silently dropped.
What is supported, and what is not
Supported: .docx, .docm, .dotx, .xlsx, .xlsm, .xltx, .pptx, .pptm and .ppsx, plus OpenDocument .odt, .ods, .odp and .odg.
XML namespace prefixes are matched on local name rather than on the literal dc:creator,
because the prefix is not fixed by the specification — files written by LibreOffice and Google Docs
use different prefixes for the same fields, and matching the written form misses them.
Not supported: the pre-2007 binary formats .doc, .xls and .ppt. Those are Compound File containers rather than ZIPs and store their properties in a SummaryInformation stream. A file that is a ZIP but not an Office document is identified as such and pointed at the ZIP extractor.
Why the document is never uploaded
The file is read from your disk and unzipped inside your browser. No server takes part.
That is not a detail here — it is the entire point. The reason to check a document’s metadata is that it may be disclosing something: a client’s name, an internal author, a company on a licence. Uploading the file to a website to find out what it reveals would hand over the very thing you are trying to control.
What this tool does not do
This is a reader, not an editor. Four boundaries:
- It does not remove or change metadata. Use Word, Excel or PowerPoint: File → Info → Check for Issues → Inspect Document, then remove what it finds.
- It does not read the document’s content. For that use the DOCX text extractor or the Excel data extractor.
- It does not show tracked changes or comments. Those live in the document body, not in the properties, and are a separate disclosure risk worth checking in Word itself.
- It does not open .doc, .xls or .ppt. The pre-2007 binary formats store properties somewhere else entirely.
Saving a document as PDF does not strip any of this. PDF carries its own author and producer fields, and Office fills them in from exactly these properties — which is what the PDF metadata extractor reads.
Why people check Office metadata
- Before sending a document out — checking whether a proposal still names the client it was adapted from, which is the classic and expensive mistake.
- Anonymous submissions — academic peer review and blind tender processes are routinely de-anonymised by the author field alone.
- Verifying a document’s history — creation date, revision count and editing time together say a great deal about whether a document is what it claims to be.
- Legal disclosure and forensics — establishing who touched a file and when.
- Journalism — a leaked or published document’s properties have exposed its origin more than once.
- Auditing a template — confirming a company template is not carrying a former employee’s name into every new document.
Office metadata compared with PDF and image metadata
Every format keeps a version of this, and each needs its own reader. Office keeps it as XML in
docProps/. PDF keeps author, producer and creation date in its document
information dictionary and XMP packet — see the
PDF metadata extractor. Photographs keep camera,
lens, timestamp and often GPS coordinates in EXIF — see the
EXIF data extractor. MP3 files keep artist and album in
ID3 tags — see the MP3 tag extractor.
The pattern is worth internalising: converting between formats moves this information rather than removing it. A Word file exported to PDF arrives with its author intact, in a different field.
For a broader look at the types of metadata that travel with every file you share, read what is file metadata.
OOXML metadata storage and edge cases
OOXML metadata lives in two XML parts inside the package ZIP:
docProps/core.xml (Dublin Core fields: title, subject, creator,
lastModifiedBy, created, modified) and docProps/app.xml (application-specific
fields: Application, AppVersion, Pages, Words, Characters, Slides, Company, HyperlinkBase).
A third optional part, docProps/custom.xml, holds arbitrary key-value pairs
added by the organisation or the document template.
Three edge cases: (a) the creator field is set when the document is first
saved and does not update on subsequent saves — a document circulated and re-saved under a
different account still shows the original creator; (b) the revision count in
app.xml increments on each save but is not reset when "Accept All Changes" is
applied — a 300-revision document may have had all tracked changes accepted years ago;
(c) the Company field comes from the Office installation's registration, not from the
document itself — every file saved on the same machine carries the same Company value,
regardless of who owns the content.
Frequently asked questions
How do I see who created a Word document?
Drop it onto this page. The Author and Last modified by fields are read straight out of docProps/core.xml, along with the created and modified dates.
What is the difference between Author and Last modified by?
Author is whoever originally created the file. Last modified by is whoever saved it most recently. On a document that has been passed around, they are usually different people, and the second one is the one that surprises people.
Is my document uploaded to a server?
No. It is unzipped and read inside your browser. That matters here more than anywhere else on this site, since the whole point is to find out what the file discloses.
How do I remove metadata from a Word file?
In Word: File, then Info, then Check for Issues, then Inspect Document. Tick Document Properties and Personal Information and remove them. This tool shows you what is there; Office is what takes it out.
Does saving as PDF remove the metadata?
No. PDF has its own author and producer fields, and Office fills them in from the Word properties. The information moves rather than disappearing.
Can I see the total time someone spent on a document?
Yes, where the application recorded it. Total editing time is stored in minutes in app.xml and counts how long the document has been open, not how long was spent typing.
Does it work on .doc and .xls files?
No. Those are pre-2007 binary Compound File containers, not ZIPs, and store their properties in a completely different structure.
Why does my document show no metadata?
It has no docProps parts, which normally means it was produced by a converter, an export or a script rather than by Office. There is genuinely nothing identifying inside it.
Can metadata prove when a document was really made?
It is evidence, not proof. The dates come from the clock of the machine that saved the file and can be changed by anyone with a text editor and a ZIP tool. Revision count and editing time are harder to fake convincingly, which is why they are shown alongside.