Does Converting a Photo to PDF Remove Its Metadata?

HEIC-to-JPG is same-family image conversion — tags often copy forward (what metadata survives HEIC to JPG conversion). Photo-to-PDF is a different mechanism: raster or image XObject embedded in a structured document with its own Info dictionary and optional XMP metadata stream.

PDF-specific stripping is remove metadata from PDF. This page is the photo→PDF path only.

What happens on conversion

Rasterize path: Word/Print to PDF draws pixels; EXIF may drop because the page is new bitmap — but document XMP can still record Author, CreatorTool, CreateDate from the app.

Embed path: ‘Insert image’ stores the JPEG stream inside PDF — EXIF APP1 may survive inside that stream untouched. Inspect with exiftool on the PDF.

XMP promotion: Some tools map EXIF GPS or camera fields into document-level XMP for search — your PDF Properties panel shows coordinates even when the page looks like a flat scan.

Why ‘PDF feels professional’ misleads

Courts and HR accept PDF — that is format preference, not privacy hygiene. A geotagged selfie embedded in a one-page PDF is still identifying if tags copied. Email clients display PDF filenames and metadata previews.

Recommended workflow

1. Strip the photo in MetadataWipe — download cleaned JPEG/PNG.
2. Build PDF from the cleaned still.
3. Clear PDF document properties (Title, Author, Producer) in your PDF editor if needed — see remove metadata from PDF.
4. Inspect final PDF before sending.

Scenarios

Insurance claim: photo pasted into PDF form. EXIF GPS in embedded JPEG under ‘damage photo.’

Legal exhibit: print-to-PDF from Photos. May strip image EXIF but add macOS Quartz Creator metadata.

Portfolio PDF with 20 RAW exports. Each embedded JPEG can carry camera serial in EXIF.

Microsoft Word and Google Docs paths

Pasting a photo into Word then exporting PDF often embeds the image object with whatever metadata Word preserved. ‘Reduce file size’ may recompress without stripping GPS. Google Docs download as PDF similarly embeds linked images — inspect with exiftool on the PDF, not only the original JPEG you pasted from.

Scan-to-PDF versus photo-to-PDF

Scanner apps that save multi-page PDFs often write CreatorTool and CreateDate about the scanner session — different field names than camera EXIF but still identifying. A phone photo of a whiteboard converted to PDF carries the photo’s metadata into the embedded image stream unless you stripped first. Treat the PDF as a composite: document-level Info plus per-image EXIF.

Email clients that ‘automatically convert attachments to PDF’ for security scanning may copy metadata from source images into archived PDFs stored on mail gateways — another reason to strip before the message leaves your outbox.

Legal teams sometimes prefer PDF because counsel can Bates-stamp pages — stamping does not strip embedded EXIF in photos on those pages. Strip photos first, assemble PDF second, stamp third. Order matters more than format prestige.

Real-estate listing PDFs with twenty interior photos are twenty embedded JPEGs worth of metadata — one strip pass on the PDF shell is not twenty strip passes on the images inside. Strip each photo in MetadataWipe before layout, or inspect every embedded stream after assembly.

Tax preparers embedding client ID scans into PDF packets should strip the JPEG before PDF creation — the IRS portal may not strip for you on the way in or out.

Multi-page photo albums exported as PDF from Apple Photos inherit batch metadata behavior from the export engine — test page one and page twenty, not only the cover.

Insurance adjusters photo-report PDFs are a common leak — strip before the PDF merge step in the field app workflow.

When in doubt, strip the JPEG, build the PDF, then inspect the PDF — two inspection points, one habit.

Related guides

See also:

• Remove metadata from PDF
• What metadata survives HEIC to JPG conversion