A QR code is only useful if it scans. That sounds obvious, but a huge share of QR codes printed on flyers, posters, and packaging every year fail silently — not because QR codes are unreliable, but because a handful of avoidable mistakes get baked in before the file ever reaches a printer. This guide covers the practical decisions that actually determine whether a code works in the real world, not just on a screen.
Start with enough contrast, not a “nice” color scheme
Cameras read a QR code by distinguishing dark modules from light ones. The moment that distinction gets murky — a pastel foreground on a slightly darker background, a gradient behind the code, a semi-transparent overlay — scan reliability drops fast, especially in low light or from an angle. A dark foreground on a light, solid background remains the safest combination. If you want brand color in the code, keep it dark and saturated rather than light and muted, and always test the specific combination you land on rather than assuming it will be fine.
Our QR code generator includes live color pickers so you can preview exactly how a given foreground and background will render before you export anything.
Respect the quiet zone
The blank margin surrounding a QR code — the “quiet zone” — isn’t wasted space. It’s what tells a scanner where the code begins and ends. Cropping a QR code too tightly, or placing it flush against a busy background pattern or another design element, is one of the most common reasons a code that looks fine on a monitor fails to scan once printed. As a rule of thumb, keep a margin of at least four modules (the small squares that make up the code) on every side, and never let text, images, or a page edge intrude into that space.
Size it for the scanning distance, not the page
A QR code that will be scanned from arm’s length — a table tent, a business card, a product label — can be small, often 2–3 cm square. A code on a billboard, a bus stop poster, or a trade show banner needs to be dramatically larger, because the rule of thumb is that a code should be roughly one-tenth the expected scanning distance. A code meant to be read from three meters away needs to be at least 30 cm across. Undersizing for the distance is the single most common design mistake we see, because a code that looks reasonably sized on a laptop screen can be far too small once printed at poster scale.
Choose the right error correction level
QR codes support four levels of error correction — L, M, Q, and H — which determine how much of the code can be damaged, dirty, or obscured and still scan correctly. Level M (about 15% recovery) is a sensible default for most printed material. If you plan to place a logo in the center of the code, or expect the printed piece to get scuffed, folded, or exposed to weather, step up to Q or H. Keep in mind that higher error correction slightly increases the visual density of the code, which matters if you’re working with a very small print size — there’s a genuine trade-off between robustness and legibility at small scale, so for tiny codes it’s often better to increase the physical size than to max out error correction.
Test on real devices before you commit to print
A code that scans perfectly on your own phone doesn’t guarantee it will scan on someone else’s. Camera quality, ambient lighting, screen glare, and the scanning app itself all affect the result. Before finalizing artwork, print a test copy at the actual size and material you plan to use — not just a laptop screen preview — and scan it with at least two or three different phones, ideally including an older or lower-end model. If you’re using the code on packaging, test it under the lighting conditions it will actually be viewed in, such as fluorescent store lighting rather than daylight.
Keep the encoded content short and stable
The more data you encode, the denser and more visually complex the resulting code becomes, which makes it both harder to scan reliably at small sizes and less forgiving of print imperfections. If you’re encoding a URL, use the shortest stable link you have — a clean landing page URL rather than a long parameter-laden tracking link, for example. If you do need analytics, consider a short link service that redirects to your full tracked URL, so the QR code itself stays compact.
Give people a reason to scan — and tell them what happens next
A QR code with no context is just a pattern of squares. The most effective codes are paired with a short, specific call to action: “Scan for the full menu,” “Scan to redeem 15% off,” “Scan for setup instructions” — rather than an unlabeled code sitting alone on a page. Set expectations about what scanning will do, especially if it opens an app, starts a download, or asks for a permission like camera or location access; an unexpected outcome erodes trust and reduces the chance someone will scan a code from you again.
Don’t rely on a single distribution channel
If a QR code is going on something permanent — packaging, a printed sign, a vehicle wrap — make sure the destination page is stable and won’t be taken down or restructured later. A code that leads to a 404 page reflects poorly on the brand and can’t be fixed once thousands of copies are already printed. If there’s any chance the destination will change, route the code through a URL you control and can update, rather than encoding a specific one-off page directly.
A quick pre-launch checklist
- High contrast, dark-on-light color combination
- Quiet zone of at least four modules on all sides
- Sized appropriately for the expected scanning distance
- Error correction level matched to the printing and handling conditions
- Tested at actual print size on multiple real devices
- Destination page is stable, mobile-friendly, and loads quickly
- A short, clear call to action next to the code
Getting a QR code to scan reliably isn’t about luck — it’s a short list of deliberate choices, most of which take no extra effort once you know to check for them. If you’re ready to put these into practice, our free QR code generator lets you adjust contrast, size, and error correction directly, with a live preview so you can see the effect of each change before you export.
Free, unlimited, and it takes about ten seconds.