How to combine photos into one PDF

Someone asked for “one PDF, not ten attachments.” Here is how to build it, get the pages the right way up, and keep it under the limit that will bounce it back.

Updated 28 August 2026 · 11 min read

The request usually arrives in the same flat sentence: please send this as one PDF, not ten attachments. It sounds like an administrative preference. It is not. A pile of photos and a PDF containing the same photos behave completely differently at the other end, and the difference is the whole reason anyone asks.

What follows covers the parts that actually go wrong — page order, page shape, file size, and photos that arrive lying on their side — rather than a list of buttons to press.

Why the PDF is asked for at all

Ten image attachments are ten separate objects. They arrive in whatever order the mail client lists them, they download into a folder where they mix with everything else, and the person receiving them has to work out which page is the front of the tenancy agreement and which is the back.

A PDF fixes four things at once.

The order is part of the file. Page three is page three on every machine that opens it. Nothing re-sorts it by date, name or size.

It prints predictably. A PDF page carries its own physical dimensions — this page is A4, this one is Letter — so a printer is not left guessing how big a 4032 × 3024 photograph is supposed to be on paper.

It opens everywhere. Every phone, every desktop, every browser, and every ageing terminal at a claims department opens a PDF without installing anything. That is not true of HEIC photos from an iPhone, which still confuse a surprising amount of office software.

It is one thing. One file to forward, one file to file, one file to lose or not lose. For anyone processing a hundred of these a week, that alone is the argument.

Page size, orientation and the white space

A photograph has no size. It has a pixel count and an aspect ratio, and nothing else. A PDF page, by contrast, is a fixed rectangle measured in millimetres or inches. Putting the first into the second is a decision, and it has to be made for every page.

Two settings do all the work. Page size chooses the rectangle: A4 (210 × 297 mm) almost everywhere outside North America, US Letter (8.5 × 11 in) in the United States and Canada. If the document is going to be printed by someone else, match their paper — a Letter page printed on A4 loses a strip from one edge, and an A4 page printed on Letter comes out shrunken with an odd border. Orientation chooses whether that rectangle stands up or lies down.

The interesting case is a landscape photograph meeting a portrait page. There are only two honest options, and it is worth knowing which one you are getting.

Fitting scales the photo until it fits entirely inside the page, then centres it. Nothing is lost. The leftover space above and below shows as white bands. Filling scales the photo until it covers the page completely, which means the sides are cropped off and disappear. Filling looks tidier and is quietly destructive — on a photograph of a receipt, the parts that vanish are the parts at the edges, which is where the total tends to live.

The tool on this site always fits. It scales each photo down to sit inside the page, centres it, and leaves white space rather than cutting anything away. It never crops.

Margins are the last piece, and people often set them to zero on the assumption that white space is waste. It is not. Consumer printers cannot print to the very edge of a sheet; they reserve a few millimetres of unprintable border, and content pushed into that band is either clipped or forces the whole page to be scaled down. A margin of roughly 24 pt — about 8 mm — keeps everything safely inside that zone. For a PDF that will only ever be read on a screen, zero margin is fine.

Getting the pages into the right order

This is the failure that costs the most time, because it is silent. The pages are all there, they are just not in the sequence you had in your head.

When you select twelve files in a picker, the order you clicked them in is not preserved. What gets handed over is the order the folder was listed in — which is almost always alphabetical by filename. And alphabetical sorting of numbers is not numerical sorting.

A plain alphabetical sort compares filenames character by character, left to right, exactly as it would compare words. Comparing 10.jpg with 2.jpg, it looks at the first character of each: 1 against 2. One comes before two, so 10.jpg wins and lands ahead of 2.jpg. The result is the sequence every one of us has met at least once:

1, 10, 11, 12, 2, 3, 4, 5, 6, 7, 8, 9

The fix is to make every number the same width, so that character-by-character comparison and numerical comparison give the same answer. Name the files 01, 02, 03 through 12, and the leading zero does the work: 02 now sorts before 10 because 0 precedes 1. If there is any chance of exceeding ninety-nine pages, pad to three digits and use 001.

Some file managers are cleverer than this and sort numbers the way a human would, which is why the problem seems to come and go. Do not rely on it — padding costs nothing and works everywhere, including on the computer at the other end.

Whatever the picker hands over, check it before you build. The tool here shows a numbered page list with arrows to move any page up or down, so the order shown is the order you get — but it is quicker to name six files correctly than to shuffle sixty.

Size, and the 25 MB ceiling

A PDF made of photographs is, near enough, the sum of the photographs inside it. There is no compression magic in the container. The PDF wrapper itself is trivial; the pages are the pictures, and the pictures are the size.

This matters because of where these documents are going. Gmail personal accounts accept attachments up to 25 MB when sending — that figure comes from Google's own support documentation. Google Workspace administrators can set different limits for work and school accounts, so an address ending in a company or university domain may be more or less generous. Other providers vary, and their ceilings are commonly somewhere in the tens of megabytes, but the only limit worth trusting is the one your provider actually documents.

Now do the arithmetic. A photograph straight out of a recent phone camera is frequently between 3 and 6 MB. Ten of them is well past any personal mailbox limit before you have opened anything, and the sequence that follows is familiar: build the PDF, attach it, wait, get told it is too large. If you are wondering why the individual files are that big, a photo's size comes down to pixel count and format more than anything else.

Aim well under the number rather than at it. Attachments are encoded into text for transport, which inflates them on the wire, so a file sitting exactly at a stated limit can still be refused.

Compress before you build, not after

The instinct on seeing an oversized PDF is to search for something that will shrink the PDF. It is the wrong direction, for a mechanical reason.

Every photo placed into a PDF is stored inside it as a compressed image, usually JPEG. Shrinking a finished PDF means opening each of those images back up, discarding more detail, and writing them again — a second lossy pass layered on top of the first, applied uniformly to every page whether that page needed it or not. The compression artefacts from round one are still in the picture, and round two spends bytes faithfully describing them. You get a worse-looking document for a smaller saving than doing the job once, properly, at the source.

Compressing the photographs first gives you control the PDF route cannot. You can shrink the four enormous camera photos and leave the two screenshots alone, check each result before committing, and aim at a specific number per file — six photos at 800 KB each lands a six-page PDF inside Gmail's limit — though some corporate mail gateways cap attachments at 5 MB or less, so check if it matters.

There is a second lever built into the conversion itself. Phone photos are around 4000 px on the long side, which is far more detail than a printed page can reproduce, so this tool caps the longest side at 2200 px as each page is drawn and re-encodes it as JPEG at the quality you choose. That happens before the PDF is written, not afterwards, which is exactly the right order. If the result is still too heavy for email, lowering the quality slider to around 70% will often halve it — on photographs of text and paper the difference is close to invisible.

Pages that come out sideways

You photograph a document the right way up, it looks correct in your phone's gallery, and in the PDF it is lying on its side. Nothing rotated it. The photo was always stored that way.

Phone cameras have a fixed sensor. Rather than physically re-arranging the pixels when you hold the phone upright, the camera records the pixels in sensor order and writes a small note in the file's metadata — the EXIF orientation flag — saying which way up the picture should be displayed. Software that reads the flag shows it correctly. Software that ignores the flag shows the raw sensor data, rotated by ninety degrees, and that is where sideways pages come from. It is also why the same file can look right in one application and wrong in the next.

The tool on this site reads the orientation flag when it decodes each photo, so pages should come out the way you took them. If you are meeting the problem in other software, or want the full explanation and the permanent fix, see why a photo appears sideways and how to correct it for good. The durable solution is to bake the rotation into the pixels themselves, so no downstream program has the chance to get it wrong.

Photographing paper so it reads as a document

Most of these PDFs are photographs of paper, and a few habits separate a page that is accepted from one that gets sent back.

Light the page flatly. A window to the side, or general room light, beats a single bright lamp or a direct flash. Flash from directly above produces a hot white blob in the middle and dark corners, and on glossy receipt paper it wipes out whole lines.

Keep your own shadow off the paper. This is the most common failure, and it is invisible while you are doing it — you lean over the document, your head and phone block the light, and a grey band lands across the text. Stand to one side of the light source rather than between it and the page.

Shoot square on. Hold the phone parallel to the paper, centred over it, rather than at an angle from the edge of the table. Angled shots produce a trapezoid page that looks careless and reads badly when printed.

Crop to the paper edges. Trim away the tablecloth, your thumb, and the mug. It looks deliberate, and it removes pixels you are otherwise paying for on every page of the finished file.

Check it is legible at full size before you build. Small print that is soft on a phone screen will not improve inside a PDF.

Common questions

Can I put several photos on one page? Not with this tool — it places one image per page. If you need four receipts to a sheet, arrange them in a single image first and add that image as one page.

Why is my PDF bigger than the photos I put in? It normally is not, but it can be if the source files were unusually small — a heavily compressed 100 KB image re-encoded at a higher quality setting will grow. Lower the quality slider and the total comes back down.

How many pages can one PDF hold? There is no fixed limit here; the practical constraint is your device's memory, since every photo is decoded as it is added. A few hundred pages is comfortable on a desktop, while an older phone is happier with smaller batches.

Will scanning with a phone be accepted instead of a real scanner? For receipts, ID documents, homework and most claims paperwork, yes — provided the whole document is in frame, in focus and legible. Where a certified or witnessed copy is specified, a photograph is not a substitute.

Does converting to PDF strip the location data from my photos? The pages are re-encoded as images, so the camera's EXIF block — GPS coordinates included — does not travel into the finished PDF. Treat that as a side effect rather than a guarantee; if privacy is the point, strip the metadata from the originals deliberately.

The recipient says the pages are in the wrong order. What now? Rebuild rather than asking them to cope. Rename the source files with padded numbers, check the page list matches what you intend, and send a fresh PDF — a corrected file is less trouble than a note explaining that page seven is really page two.