Exporting
Export in the toolbar (it says how many pages, or files, it will make) opens the export options: everything below. At the foot, always in view, are the estimated file size and Export PDF, which builds the file and saves it to your downloads (or Export JPG, Export 3 PNGs and so on, when saving as images). When a password is set, it says Export locked PDF, and with PDF/A on, Export PDF/A. While a file is being built, the toolbar button shows its progress.
Save as
PDF is the usual choice. JPG or PNG saves each page as an image instead, for an upload form that won't take a PDF. The image is the page exactly as the PDF would show it: crop, filters, blacked-out areas, signatures and text you added, watermark and page numbers included. It's drawn at your image quality's resolution: 300 dpi for Original and High, 200 for Balanced, 150 for Small. JPG makes smaller files and suits photos; PNG keeps text perfectly sharp but is larger.
Images take All pages or Selected pages, one image each. One page downloads as an image (Lease.jpg); several come in one ZIP (Lease.zip, holding Lease-page-1.jpg, Lease-page-2.jpg and so on, numbered 01, 02 once there are ten or more). On a phone, when they're ready, Share… opens the share sheet, so you can save them to your photos or send them straight to an app; Download (or Download ZIP) saves them as a file instead, and Not now drops them. An image can't carry searchable text, a password, bookmarks or a size limit, so those options are hidden while JPG or PNG is chosen.
Title and filename
The title is stored in the PDF (readers show it in their title bar) and names the file. Leave it empty for a dated name.
Pages
- All pages: one PDF of the whole document.
- Selected pages: one PDF of just the pages you've selected (see Arranging pages), in document order.
- Split by page ranges: one PDF per range. Type them like
1-3, 4-10, 11-:11-runs to the last page and-3starts at the first. Ranges can overlap and come in any order; each file follows the order you wrote. A single range, such as3-5, gives just those pages as one PDF. - Split every few pages: a new file every N pages, the last one holding what's left.
Under the choices, a line says what you'll get, or what's wrong with the ranges. Several files download together as one ZIP, named from the title (Bills.zip, holding Bills-part-1.pdf, Bills-part-2.pdf and so on). Every file gets the same options: quality, size limit (each file on its own), searchable text, page numbers, watermark and password. With page numbers on, In each file chooses whether each file starts again from the first number or carries on from the file before.
Print as
How the pages go onto paper. PDF only.
- Pages: each page on its own sheet, the usual choice.
- Booklet: two pages on each side of a sheet turned sideways, in the order that folds into a booklet. Print on both sides, flipping on the short edge, then stack the sheets, fold them in half and staple along the fold. Blank pages are added at the end to make a multiple of four. Each page is shrunk to fit half the sheet, against the fold: A4 pages on A4 paper come out A5.
- Poster: each page enlarged across a grid of sheets, up to 8 across and 8 down. Neighbouring sheets share a strip (Overlap, 10 mm to start), so the white edge printers leave is trimmed away. With Cut marks and sheet labels, each sheet with a neighbour to its left or above has short marks at each end of a line: cut that strip off along it and lay the sheet over its neighbour so the picture lines up. Labels (
A1,A2, … by row and column) say where each sheet goes, printed where they end up cut off or covered. A sheet that would come out blank, such as the ends of a wide grid under a tall page, is left out.
Paper is the sheet you print on: Same as the pages takes the page size from Page setup (your usual paper when pages match their pictures). A line under the choices says how many sheets it takes. A booklet or poster has no bookmarks, since its pages are no longer the document's; page numbers, the watermark, searchable text and a password all stay. With a split, each file is its own booklet or poster.
Image quality
- Original: images go in as they are, with no recompression. JPEG photos keep their exact image data. A page is re-encoded once only when it must be: a filter, a crop, a blacked-out area, or Fill / a scale above 100% hiding part of the image (so the hidden part isn't stored).
- High, Balanced, Small: cap each image at 300, 200 or 150 dpi as printed, and compress it. Small is for email and uploads.
Pages from a PDF follow the quality too: the pictures inside them, such as a scanned page or a photo on a page, are capped and compressed like your photos, so a heavy scanned PDF gets smaller. Everything else on the page stays as it was, so its text stays sharp and searchable (an invisible text layer from a scanner app included). A picture is only replaced when that makes it smaller. Black-and-white scans are usually small already and stay as they are, as do pictures in CMYK, JPEG 2000 or with a limited palette, which recompressing could change. At Original, PDF pages go in untouched.
Keep the file under is optional: give a size in MB and Collated tries lower quality, step by step, until the file fits. If it can't, it says so and saves the smallest it made.
Searchable text (OCR)
Make text searchable recognises text on image pages, and on PDF pages that go into the file as an image (ones with a hidden layer), and adds it to the PDF invisibly, under the picture, so you can search the PDF and copy from it. It reads each page as you see it, crop and rotation included. Choose the Language of the text under the checkbox: English, Bulgarian, Croatian, Czech, Danish, Dutch, Finnish, French, German, Greek, Hungarian, Italian, Norwegian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, Turkish or Ukrainian. Right-to-left scripts (Arabic, Hebrew) and Chinese, Japanese and Korean aren't offered yet. It starts at your browser's language when that's one of them, and is remembered. Recognition runs on your device. The first time you use a language, it downloads that language's recognition data, about 2–5 MB in all (from Collated's own site, not a third party); after that, it takes a few seconds per page. Pick the language the page is written in: accented letters and words in another language may come out wrong. Anyone with the file can search and copy the text, and the recognised text may contain mistakes.
Page numbers
Adds a number to every page: bottom centre, bottom right or top right, starting from any number.
Bookmarks
PDF readers list bookmarks in a side panel, to jump around a long file. To start a section, select its first page and type a name under Section starts here in the page's settings; the page shows a bookmark tag with the name. A section belongs to its page, so it moves when you move the page, and undo takes it back like any other change. Clear the name to remove it.
Bookmark each imported file adds a bookmark named after each file you added (without its extension), at the first page that shows it. With sections, the files are listed under the section they're in. When you export several files, or only some pages, each file starts with the bookmark of the section its first page is in, so nothing loses its heading. The file names you bookmark are visible to anyone with the PDF, unless it has a password.
Watermark
Stamps text across your pages, such as "Copy for Acme Bank only", so a copy of an ID or a bill can't be reused for something else. Add date puts today's date after the text, so the copy also says when you sent it.
- Position: Across the page runs it corner to corner through the middle; Along the foot puts it in the bottom margin (above the page numbers, if they're there too). Text too long for the page is made smaller to fit.
- Size, Colour, Opacity: how large, in what colour and how solid it is. The pages show it as it will print.
- On: every page, or only the pages you had selected when you chose Selected pages. If you select other pages later, Use the N selected moves it to those.
- Burn into image pages: normally the watermark is text laid over the page, which a PDF editor could delete. With this on, each page made from photos or scans is saved as one picture with the watermark drawn into it, so there's no clean copy underneath. Those pages are re-encoded (at your image quality, at least 150 dpi), so the file can be larger. Pages from a PDF keep the watermark as text, so their own text stays sharp and searchable.
The PDF's built-in font covers Latin letters, including accents (é, ß, å). Other letters, such as Ł or Cyrillic, can only be burned into image pages; Collated says so under the text.
Like the title, the watermark's text and the pages it's on aren't remembered for next time; how it looks is.
PDF/A for archiving
PDF/A, for archiving saves the kind of PDF that courts, governments and records offices ask for (PDF/A-2b, which they accept wherever they ask for PDF/A). Its fonts and colours are kept inside the file, so it opens and prints the same in any reader, for decades. Photos and scans look as they do in any export; text you typed on a page, the watermark and page numbers are set in Arimo, a font with Helvetica's shapes and spacing, kept in the file. Pages from PDFs you added become pictures of themselves (drawn at about 300 dpi, then stored at the quality you chose), because their own fonts and colours may not meet the standard: turn on searchable text to keep their words searchable. A PDF/A file can't have a password (an archive must open without one), so the password field is set aside while it's on; what you typed comes back if you turn PDF/A off.
Password
Password to open encrypts the PDF (AES-256); it opens only with the password. Once open, nothing is restricted. Not hidden: the file's name, which comes from the title, and how many pages it has. The password isn't saved anywhere: if you lose it, the file can't be opened. Show reveals what you've typed. Passwords can be up to 127 bytes (roughly 60–120 characters, depending on the alphabet).
After you export
The file is saved wherever your browser saves downloads. Now and then, after an export that went as asked, a small note in the corner says Collated is free and links to Ko-fi and GitHub Sponsors, in case you'd like to support it. It shows at most once a month, never over an error, and never again once you choose Don’t ask again or follow one of its links. Every feature stays free either way.