The CV you sent through a recruitment portal last week probably arrived with a small block of CV metadata attached: your home computer's user account name, the salary figures from an earlier draft saved as a comment, the original job title from the template you copied, the agency that built that template, and the precise minute you finished editing. None of it appears on the printed page. All of it is visible to any recruiter who right-clicks and selects Properties, and to most applicant tracking systems automatically.
Most candidates assume their CV is a clean document. It rarely is. Recruitment is a place where small metadata signals shift hiring decisions in measurable ways.
What is actually in a Word or PDF CV
The same fields that ship with any Office document apply to CVs. The ones that matter for hiring are:
- Author name: the user account on the machine where the file was first saved. If the candidate built the CV on a friend's laptop, the friend's name appears here.
- Last modified by: the most recent editor. If a recruiter or career coach ran a pass over it before submission, their name shows up.
- Company: the Office activation's company field. Often the candidate's current employer — problematic when the search is meant to be confidential.
- Total editing time: minutes spent in the document since creation. A 1,200-minute editing time on a one-page CV signals a lot of revision; a 3-minute editing time signals a copy-paste from somewhere else.
- Created date: when the file was first saved. If the document was created the morning of submission, the recruiter sees that.
- Template: the .dotx the document was based on. Often an agency template name, or the original company-branded template if the candidate copied an existing CV from work.
- Comments and tracked changes: residue from earlier drafts. The most damaging are notes like "ask for £75k" or "drop this if applying through an agency".
For PDFs exported from Word, all of the above carry through into the /Info dictionary and the XMP metadata block unless explicitly stripped during export.
What recruiters and ATS systems actually see
Applicant Tracking Systems (Greenhouse, Workday, Lever, SmartRecruiters, BambooHR) parse CVs into structured fields automatically. Most of them read the document properties as part of that parse:
- The author name is used as a tiebreaker when name parsing fails on the visible content.
- The created and modified dates are surfaced to recruiters as "CV freshness" indicators.
- The application name (Microsoft Word, Pages, Google Docs, OpenOffice, ResumeBuilder.com) is logged as a candidate signal — some teams use this to detect template-driven applications.
- The company field, if populated, can be cross-referenced against the candidate's stated current employer for inconsistency flags.
Recruiters acting outside an ATS — opening a CV in Word or Acrobat directly — see all of the above through File → Info or File → Properties.
The leaks that catch candidates out
The earlier-draft salary number
A candidate writes "Expected salary: £85,000" into a draft, then deletes it before sending. Word holds that text in its tracked-changes log if Track Changes was on, and in the autorecover stream more often than you would think. The recruiter does not see it on the page; they see it in the version history.
The agency template fingerprint
A candidate uses a template downloaded from a recruitment agency. The template's name persists in the document properties. Submitting that CV to a competing agency, or directly to an employer who mistrusts agency-supplied candidates, signals where the CV came from before any human reads a word of it.
The rejected-job residue
A candidate adapts a CV for a new application by editing a copy of the previous one. The previous job title sometimes survives in the keywords field, the title field, or as a comment that was deleted but not removed. The new recruiter sees what the old application targeted.
The current employer signal in confidential searches
A candidate looking discreetly while still employed sends a CV with the company name in the Office activation field. The recruiter knows where the candidate works before opening the file.
Cleaning a CV before you submit
A reasonable hygiene pass:
- In Word, run File → Info → Check for Issues → Inspect Document. Remove document properties, comments, revisions and personal information.
- Empty the comments pane manually if you have used it.
- Check the keywords and subject fields for residue from earlier drafts.
- If exporting to PDF, untick Document properties in the export options.
- Run the final file through a metadata cleaner to catch what Word's inspector misses (template paths, app version IDs, residual XMP).
The cleaner on the homepage handles step five for both .docx and .pdf, in a browser, without uploading the file. It will also show you exactly what was inside before stripping it — which, the first time most people try it on a CV, is a surprisingly detailed read on the workflow that produced it.
The point is not that CVs need to be sterile. It is that the document you submit should be the document the recruiter sees, not a layered archive of the eight earlier versions of itself.