Page size is the decision that matters
There are two sensible answers and they suit different jobs, which is why this tool asks rather than choosing for you.
Fit the image makes every page exactly the size of the picture on it. Nothing is cropped, nothing is surrounded by white bars, and a document of mixed-shape screenshots looks deliberate rather than padded. The pages will have different dimensions from each other, which is fine on screen and unpredictable in a printer.
A4 or Letter gives every page the same standard size, with each image scaled to fit inside a margin you choose. This is what you want when the document is going to be printed, emailed to an office, or uploaded somewhere that expects a real document. Portrait and landscape can follow each image automatically, so a mixture of shapes still fills the page sensibly.
One deliberate limit: images are never enlarged to fill the page. A small scan blown up to A4 is a blurry scan on a big page, and the tool will leave it at its natural size within the margins instead.
What happens to your images
PDF supports JPEG natively, so each image is encoded once and embedded — the document carries image data, not a re-render of it. That means one generation of compression for photographs, which is invisible at the quality used here but worth knowing if the pages have already been through several rounds.
If you have exactly one image and want it embedded with no re-encoding at all, the JPG to PDF converter takes the passthrough route and copies the original bytes untouched. This builder cannot, because a mixed batch has to be normalised before it can be assembled.
Ordering pages without dragging
Page order is the entire reason this tool exists, so it uses buttons rather than drag-and-drop. Dragging a list on a touchscreen while the page scrolls underneath is awkward, and it is impossible with a keyboard — which would make the one feature that matters unusable for anyone who does not use a mouse.
Frequently asked questions
How do I change the page order?
Use the up and down buttons next to each image. They are ordinary buttons rather than drag handles, which means the order can be set with a keyboard and works reliably on a phone — dragging a list on a touchscreen while it scrolls is genuinely awkward, and page order is the whole point of this tool.
Should I choose "fit the image" or A4?
Fit the image when the pages are screenshots, photos or scans of different shapes and you simply want them in one file — nothing is cropped and no white bars appear. A4 or Letter when the document will be printed or submitted somewhere that expects a standard page, since a mixture of odd page sizes prints unpredictably.
Does combining reduce quality?
Each image is encoded once at quality 88 before being embedded, so a photograph loses one generation — invisible at normal viewing but real. Images that are already JPEG go through that step too, because a mixed batch has to be normalised before it can be embedded. If you need one image untouched, use the single-image JPG to PDF converter, which embeds the original bytes.
Is there a limit on how many images?
No fixed limit. Everything is assembled on your device, so the constraint is your memory rather than a quota we set. A few dozen photographs is comfortable; several hundred large ones will be slow and may run out of memory, and the tool will say so rather than crashing silently.
Can I add a PDF to the mix?
Not yet — this builds a document from images only. Merging existing PDFs together is a different operation and is not offered here rather than half-implemented.
Is the text in my scanned pages searchable?
No. The PDF contains pictures of pages, so nothing is selectable or searchable even when the words are clearly visible. That needs OCR, which recognises the characters and adds an invisible text layer.