Skip to content
TernTern home

Why your phone photos turn sideways in a PDF

Your phone did not rotate the picture — it wrote down which way up it was held. PDF has nowhere to put that note, so the rotation gets lost.

3 min read · updated

The short answer

Your phone almost never rotates the picture it takes. It saves the image exactly as the camera sensor saw it and attaches a small note — the orientation tag — recording which way up the phone was being held.

Photo apps read that note and turn the picture for you, which is why the photo looks right everywhere you normally see it. PDF has no equivalent note. A converter that copies the image straight into a document copies the sensor’s version and discards the instruction, and you get a page lying on its side.

How to tell this is what happened

  • The photo looks correct in your gallery, on your phone, and in a preview — but wrong in the PDF.
  • It is rotated by exactly 90 or 180 degrees, never by a random angle.
  • Portrait photos are affected and landscape ones often are not, or the other way round, depending on how you held the phone.
  • Screenshots and images downloaded from the web are usually fine, because they were saved upright with no note attached.

If the rotation is inconsistent across a batch — some pages right, some wrong — that is a strong confirmation. The ones that are wrong are the ones you took holding the phone differently.

Fixing it

  1. Use a converter that reads the tag

    The image to PDF tool here checks each photo’s orientation and applies it before the image goes into the document. Photos that need no rotation are embedded untouched, so nothing is re-compressed unnecessarily.

  2. Or rotate and re-save first

    Open the photo in any editor, rotate it, and save. Editors write out the corrected pixels with the note cleared, so anything downstream sees an upright image.

  3. Or rotate the finished PDF

    Workable for one or two pages, tedious beyond that, and it does not help if different photos are rotated different ways.

Why a converter should not just rotate everything

The obvious fix — redraw every image so the rotation is baked in — works, and costs something on every photo that never needed it. Redrawing means decoding and re-encoding, which loses a little quality and usually makes the file bigger.

Most images have no rotation to apply. Reading the tag first and only redrawing the ones that are genuinely rotated keeps the rest byte-for-byte as they were shot. It is a small distinction that shows up as a noticeably smaller PDF when you convert a folder of screenshots or scans.

A related iPhone problem: HEIC

If your iPhone photos will not load into a browser-based tool at all, they are probably HEIC rather than JPG. It is a newer format that saves space, and most browsers other than Safari cannot read it.

To change it: Settings → Camera → Formats → Most Compatible, which saves new photos as JPG. For photos you already have, sharing or exporting them usually converts to JPG on the way out.

Frequently asked questions

Why does the photo look fine on my phone but not in the PDF?

Because your phone reads the orientation tag and turns the picture for display. The picture itself was never turned. Anything that ignores the tag shows you the untouched version.

Will fixing the rotation reduce the quality?

Slightly, for the photos that actually need rotating, because the image has to be decoded and re-encoded. Photos that are already upright are not touched at all.

Why are only some of my photos affected?

Because the tag records how you were holding the phone, and that varied. Photos taken in the sensor’s natural orientation carry no rotation and come out correctly wherever they go.

Do scans have this problem?

Rarely. Scanners usually write the page the right way up with no orientation tag. Phone scanning apps are a different matter — they are cameras, and behave like one.

Related tools

Guides worth reading