Why some QR codes survive a scuff and others don't — and how to size one so it scans the first time. · 5 min read
The four error-correction levels, and what they really buy you
Every QR code carries redundancy. The encoder splits the payload into data blocks and then adds Reed–Solomon parity blocks over them, so a scanner can reconstruct the message even when part of the symbol is unreadable. You choose how much of the symbol is allowed to be sacrifice material: level L recovers up to about 7% damage, M about 15%, Q about 25% and H about 30%.
The practical trade-off is density. The same URL at level H needs a physically larger symbol or a finer module size than at level L, because extra parity needs extra modules. A code that only has to survive a smooth, flat surface (a website footer) is fine at M. A code that will be printed on a bumpy surface, embossed, worn by weather, or placed behind a logo overlay wants Q or H — remember that a logo overlay counts as damaged area, which is exactly why stylised codes are always built on Q or H.
A common mistake is assuming higher error correction also fixes 'blurry camera' problems. It does not: those are usually sizing or quiet-zone problems, not parity problems. Parity handles local damage like a scuff, hole or smudge, not a globally soft image.
Sizing: the 10:1 rule and the quiet zone
The rule of thumb scanners and retail signage are built around is simple: the shortest scan distance should be about ten times the width of the symbol. A 4 cm code is comfortable from arm's length; a 20 cm poster code is readable from two metres. Below that ratio, module contrast falls off and phones start to hesitate — especially with dense payloads.
The quiet zone is the empty margin around the symbol. The specification asks for four modules of clear space on every side, and many print shops shave it off to make a layout tighter. That is the single most common reason a printed code stops scanning: the scanner can no longer find the three finder patterns cleanly against a busy background. If the background has graphics, put the code in a white or very light rounded card rather than shrinking the quiet zone.
Contrast matters as much as size. Dark modules on a light background is the specification's happy path; inverted codes (light on dark) scan poorly on many readers, and colour pairs work only when the dark colour stays genuinely dark and the light one genuinely light — think luminance contrast, not hue contrast.
PNG, SVG or PDF for the final artwork
For screens, PNG at two to three times the expected display size is plenty and prints fast. For anything that will be scaled — a poster, packaging die-cut, or a vinyl cutter — SVG is the better master, because the modules stay razor-sharp at any size and the file is a few kilobytes. PDF is the right choice when the artwork goes straight to a print shop, since it can embed the code at exact physical dimensions along with crop marks.
One more distinction worth keeping in mind: a static QR code points at a fixed URL forever, so the printed symbol only survives as long as that URL does. Dynamic codes (short URLs that redirect) let you change the destination after printing, but they add a third party to the chain — the code now stops working if that redirect service dies. For anything long-lived and under your control, print a URL you own.