Scan a document to PDF in the browser.

You have a form that needs signing and sending back, or a receipt for an expense claim, or a letter someone wants as a PDF. The paper is in front of you. What you need is a PDF, and what stands between you and one is usually an app store.

Scanner apps are mostly three simple things — a camera, a crop, and a PDF writer — wrapped around an account and, very often, a cloud sync nobody asked for. The three useful parts do not need any of that, and a browser can do all of them.

What it does

Open Scan to PDF and tap Use camera. The browser asks permission, the camera opens inside the page, and each photo you take becomes a page. When you are done, the pages are assembled into a PDF using the same image-to-PDF pipeline as the rest of the site — fitted to A4 or Letter with a margin, in the order you took them.

The camera stream is a local MediaStream inside the tab, the same kind a video-call preview uses before you join. Frames become images in memory. Nothing about the stream is transmitted, and the tracks are stopped when you close the camera.

Multi-page is the normal case, so the camera stays open between shots: snap, snap, snap, then Done. A running count tells you how many pages you have. If one comes out blurred, delete that row from the list before building.

What it does not do

This is worth being plain about. Adobe Scan and the better dedicated apps do automatic edge detection, perspective correction, and contrast normalisation — they find the page in the frame, straighten it, and make it look like it came off a flatbed. This does not do any of that. A photograph taken at an angle stays at an angle. The desk stays in the frame unless you crop it.

What that means practically is that the quality of the result is entirely down to how you hold the phone. Flat above the page, parallel to it, page filling the frame, light coming from the side rather than over your shoulder. Two minutes of care gets you something that looks like a scan; a careless shot looks like a photo of paper, because it is one.

The steps

1. Tap Use camera — or drop photos you have already taken. Both work, and you can mix them.
2. One photo per page — take them in order, and retake anything blurred before moving on.
3. Build the PDF — pages fit to A4 or Letter, saved straight to your device.

On a phone the rear camera is selected by default, which is the one you want for paper. There is a switch control if you need the front one.

On iPhone and Safari

It works, with one requirement: the page must be served over HTTPS, which the live site is. Safari will ask for camera permission the first time and remember the answer for the site. If you refuse and change your mind, the permission is in Settings under Safari, or in the address-bar controls on newer versions.

If the camera button is missing entirely, that means the browser is not offering camera access to the page — on a secure connection with a current browser it is always there. You can still pick photos from your camera roll, which produces the same result.

Page sizes, and why your PDF might be enormous

A modern phone camera produces 12-megapixel images. At full size that is roughly 5 MB per page, so a ten-page scan lands around 50 MB — far more detail than a PDF page can display, and more than most portals accept.

There is a resolution control for exactly this. Fit for a document is the default and caps the long edge at 2600 pixels, which is more than an A4 page shows at print quality. Smaller, for email goes to 1600. Keep original resolution preserves every pixel and produces the large file, which is right only if you are archiving. An estimate of the finished size appears before you build.

After you have the PDF

• To search it — OCR a PDF, then download the searchable version. A scan has no text in it until you do this. The longer explanation is here.
• To fit an upload cap — compress it. Do this before OCR, not after, or you will destroy the text layer.
• If you scanned in two sittings — Merge PDF combines the parts, and photos can go straight in alongside PDFs.

It is not uploaded

The camera stream and the resulting images stay in the tab. That claim is checkable in the same way as any other on this site: open developer tools, switch to the Network panel, scan a page, and watch nothing leave. The method is written out here.

It matters more for scanning than for most jobs, because of what gets scanned. Nobody photographs a novel page by page. People scan the form going back signed, the hospital letter, the utility bill for a landlord, the passport page a visa portal wants. A scanner app that quietly syncs those to a server is doing the one thing you would decline if it asked.

When something else is better

Honestly: at fifty pages or more, use a flatbed scanner or a document feeder. Photographing fifty pages one at a time is miserable regardless of the software, and a feeder does it unattended.

If you need automatic deskew and edge detection — a bound book, a page you cannot flatten, poor lighting you cannot fix — a dedicated app will produce a better image than a plain camera frame. The iPhone's own Notes app has a document scanner built in that does perspective correction and never uploads anything, and for a quick one-page scan on an iPhone it is genuinely a good option.

This is the right tool when you are at a computer, the paper is in front of you, you need a PDF now, and you would rather not install anything or hand the photograph to a company you have not heard of.

Common questions

Tools on this page

• Scan to PDF — Photograph pages with your camera, get one PDF.
• OCR PDF — Make a scanned PDF searchable, or pull the text out.
• Merge PDF — Combine PDFs in any order, in this tab.