How to scan a document without installing an app
Open a scanning tool in your phone's browser, photograph the page, and let it find the edges and straighten it. No install, no account, and on a tool that works client-side the photograph never leaves the phone. The whole job takes about thirty seconds.
Scan to PDF does exactly this in the browser you are reading in. PDF Scanner
Scanning a page in a browser, step by step
- Put the page on a surface that is a different colour from it, and stand over it rather than leaning across it.
- Open the scanning tool and choose the camera when the file picker offers it.
- Check the four corners it found. Drag any that are wrong, which takes about two seconds and is exact.
- Add another photograph for each further page, and drag the thumbnails into the right order.
- Export the PDF. It downloads from your own device, because that is where it was made.
Why you might not want the app
Document scanners are among the most aggressive apps on either store: several of the popular ones ask for contacts, location and full storage access, show ads between every screen, and put a watermark on the free export. Some of the best known have moved to subscriptions that cost more each year than a basic flatbed scanner costs once. For scanning one form, once, that is a great deal to hand over.
There are also the situations where installing anything is not an option: a work phone with a managed profile, a borrowed device, a library computer, a phone that is simply out of storage. A tool that runs in the browser you already have removes the whole question, and leaves nothing behind afterwards.
What the browser can genuinely do
Take a photograph, find the four corners of the page, correct the perspective so it looks scanned rather than photographed, lift the contrast so paper reads white and ink reads black, and write a multi-page PDF. All of that is ordinary work for a modern browser and needs no server.
It is worth being precise about why that is possible now. The pieces a scanner app is made of, camera access, image processing, PDF writing, have all been standard browser capabilities for years. What was missing was anyone bothering to assemble them without a server in the middle, because a server in the middle is how this kind of site usually makes its money.
How the processing works, in detail
What it cannot do
Read the text. Recognising words on the page is a different job, and on the web version it is not there: the PDF holds pictures of your pages. Find-in-page will find nothing, and copy and paste will select nothing, because to the file there are no words, only ink-shaped pixels. If you need the text searchable, that is text recognition, and it is in the Android app.
How a searchable PDF actually works
When a browser tool is the wrong choice
If you scan every day, install a real scanner app. Its live viewfinder shows the crop before you take the picture instead of after, which is faster page after page, and it works with no connection from the first use. The browser version earns its keep for the occasional scan, not the daily pile.
If you already have Google Drive on Android, or use the Notes app on an iPhone, you may not need anything new at all: both have a competent document scanner built in. Where your page goes when you use them is governed by those companies' terms rather than by anything on this page, which either matters for the document in your hand or does not. On pure capability they are good, and a page like this would be worth less if it pretended otherwise.
Bound books are the other honest failure. Perspective correction straightens a flat page photographed at an angle; it cannot un-curve the arc a page makes as it dives into a book's spine. For a few pages, photograph them one at a time and hold the page flat. For a whole book, a copier with a book edge does what no phone tool can.
And for old photographs, use a flatbed scanner. It controls the light, sits the print flat, and captures far more tonal range than a phone pointed at a curling photo. A document is information on paper and survives rough handling; a photograph is the thing itself.
Getting a result that does not look like a phone photo
Put the page on something that is not the same colour as it, stand over it rather than leaning across it, and keep your own shadow off it. Those three things matter more than any setting in any tool, because processing can only work with the light the camera captured.
The shadow one is the ruin people do not notice until afterwards. Light from behind you puts your own outline across the middle of the page; light from a window to the side puts it harmlessly off the edge. Move yourself, not the page, and look at the page on the screen for a second before you shoot rather than after.
Where the photograph goes, and where it does not
On this site, nowhere. The photograph is opened by the page in your browser, processed in your device's memory, and the finished PDF is written by your browser into your downloads folder. There is no upload step, and no server on this site that could receive one; the claim is enforced by the build and the server configuration, not by a policy paragraph. On a shared or public computer the caution is local instead: the photo and the finished PDF sit in that machine's downloads folder until somebody deletes them, so for anything sensitive, use your own device.
Questions people ask next
Do I need to sign in?
No. There is no account on the web tools, no email step and no daily cap. There is also nothing to sign in to, because nothing about your document is stored anywhere.
Does the photograph get uploaded?
Not on a tool that works client-side. On this site the file is opened inside the page and processed in your device's memory, and there is no endpoint here that could receive one. You can watch the network tab while you scan.
Will it work on a laptop with no camera?
Yes. Choose a file instead of the camera. A photograph taken on a phone and copied over, or a picture already saved, goes through exactly the same steps.
Do I need to stay online while I scan?
Loading the page needs a connection, and the first export fetches the PDF writer, about two hundred kilobytes over the wire. The scanning work itself uses no network at all, which you can prove by switching to aeroplane mode after that first export and doing the next scan entirely offline.
How many pages can one PDF hold?
No number is imposed, and the honest answer is that your device's memory decides. Every page is held at working resolution while you build the document. Dozens of pages are comfortable on any recent phone; hundreds belong on a copier with a feeder rather than in any phone tool, ours included.
Can I scan a book page?
A single page, yes: hold it as flat as you can, accept a shadow near the spine, and drag the corners yourself, because the automatic detector looks for a straight-edged sheet. What no phone tool does well is the curve of a tightly bound page, and pretending otherwise sells you a bad scan.