What is a QR code?
A QR code is not a link to your data. It is a picture of your data. That one fact explains the three squares, why a damaged code still scans, and why they never expire.

Ama Mundum LLC Team5 min read
The useful thing to understand about a QR code is that it is not a link to your data. It is a picture of your data.
Nothing is stored anywhere. The text, whatever it is, has been encoded directly into that arrangement of black and white squares, and a scanner reads it back the way you read a word. The code is the data, written in a shape a camera can handle.
Almost everything else that confuses people about QR codes follows from that one fact.
It is why a QR code works with no internet connection, at least to read it. It is why a code printed in 2010 still scans today, because there is no server that could have been switched off. It is why putting more text in makes the pattern visibly denser: you are not changing a reference, you are writing more.
The name is short for Quick Response. It was invented in 1994 by a Japanese company called Denso Wave, which needed to track car parts faster than a regular barcode allowed.
What the parts of a QR code are
Look at any QR code and you can pick out the pieces.
The three big squares, at three corners. These are finder patterns, and they are how a scanner knows it is looking at a QR code at all. Three corners marked and the fourth left empty is also how it works out the orientation, which is why you can scan a code upside down or sideways without thinking about it.
A smaller square near the empty corner. This is an alignment pattern. It lets the scanner correct for the code being photographed at an angle, or printed on something curved like a bottle or a mug.
The dotted line running between the big squares. This is the timing pattern, and it works as a ruler. It tells the scanner how large one module, one little square, actually is in the photo it has taken.
Everything else. Your data, plus a substantial amount of error correction, which is the interesting part.
Why a damaged QR code still works
QR codes carry redundancy on purpose, using a technique called Reed-Solomon error correction. The code can lose a chunk of itself and still be read perfectly, because the missing information can be reconstructed from the rest.
There are four levels, and you choose one when you generate the code:
| Level | Roughly how much damage it survives |
|---|---|
| L | 7% |
| M | 15% |
| Q | 25% |
| H | 30% |
This is why a company can put a logo in the middle of its QR code. The logo is damage. At level H the code simply reconstructs what the logo is covering.
The trade is real, though. Error correction is stored inside the same symbol, so a higher level means more modules, which means a denser code that needs to be printed larger to stay readable. Choosing H for a code that will only ever live on a clean screen is paying a cost for nothing.
How much can one hold?
At the largest size, and with the lightest error correction, the limits are about:
- 7,089 digits
- 4,296 letters and digits
- 2,953 bytes of arbitrary data
You should use a small fraction of that. A code holding a long URL has far more modules packed into the same square, so each one is smaller, and small modules are exactly what fails when the code is printed, photographed at a distance, or scanned by an older phone.
Short input, sparse code, reliable scan. If your link is long, shorten it before encoding it rather than after.
What you can put in one
A QR code holds text, and the scanning app decides what to do with that text based on how it starts. That is the whole mechanism behind the different "types" you see offered:
- A web address opens a browser
tel:followed by a number offers to callmailto:offers to write an email- A specially formatted block of text can join a Wi-Fi network or save a contact
None of this requires a special kind of QR code. It is all just text, formatted in a way the phone recognizes.
Scan counting is the one thing a plain QR code cannot do, since there is no server involved to count anything. The practical workaround is to encode a URL that is already tagged, so your analytics tells you where the traffic came from. Our UTM builderOpen tool makes that link.
Are QR codes safe?
The honest answer is that a QR code is a link you cannot read before you follow it, and that is a genuine weakness rather than a theoretical one.
Two habits cover most of it:
Read the preview. Every modern phone shows the address before opening it. Look at the domain, not at the rest.
Be suspicious of stickers. The common attack is physical: a printed sticker placed over the real code on a parking meter, a restaurant table or a public charger. If the code looks stuck on rather than printed as part of the sign, type the address by hand instead.
A code you generated yourself and printed yourself carries none of this risk, because you know what is in it.
Make one
- QR code generatorOpen tool for a link, phone number or plain text, downloaded as a PNG
- UTM builderOpen tool to tag the link first, if you want to know how many people scanned it
- WhatsApp link generatorOpen tool if the code should open a chat rather than a web page
The code is built in your browser, contains your data directly and never expires, which is the whole point of the thing.