Your QR updates live as you type. The logo uses high error-correction so it still scans.
Free QR Code Generator with Logo — No Watermark, No Sign-Up
Create a custom QR code in seconds. Type any URL or text, pick your brand colours, add a centre logo, and download a high-resolution PNG up to 1024×1024. No watermark, no account, no expiry, no limits. Your logo and links never leave your browser.
Why a logo in the middle still scans
A branded QR code earns more scans than a plain one, but covering the centre risks destroying the data. Every code here is generated at error-correction level H, the highest of the four QR levels, which allows roughly 30% of the code to be obscured and still decode. The logo is capped at 22% of the area and sits on a rounded background plate, keeping you comfortably inside that margin.
Static codes never expire
These are static QR codes — the destination is encoded directly into the pattern. No subscription, no tracking redirect, no risk of the code dying because a service shut down. The trade-off is that you cannot change the destination later, so for print, point the code at a URL you control.
Printing tips that prevent dead codes
- Minimum size: print at least 2×2 cm. For a poster read from 3 metres, scale to roughly 25×25 cm.
- Keep the quiet zone: leave clear space on all four sides. Crowding is the most common cause of scan failure.
- Keep contrast high: dark foreground on a light background. Inverted codes fail on many scanners.
- Test before printing: scan the downloaded PNG with two different phones first.
Also useful: our Password Generator and Image Compressor, or browse our software reviews.
Error correction, and how much logo you can actually get away with
QR codes use Reed-Solomon error correction, which builds redundancy into the code so it still decodes when part of it is damaged, dirty or covered. There are four levels, and the difference between them is substantial:
| Level | Recoverable | Use when |
|---|---|---|
| L (Low) | ~7% | Clean digital display, no logo, long URL |
| M (Medium) | ~15% | General print, no logo — the usual default |
| Q (Quartile) | ~25% | Small logo, or print likely to get scuffed |
| H (High) | ~30% | Centred logo, outdoor signage, packaging |
This is why a logo in the middle still scans: the covered modules fall within what the error correction can reconstruct. The practical rule is to keep the logo under roughly 20 to 25% of the code area even at level H, and always centred. The corners must stay clear — the three large squares are the finder patterns a scanner uses to locate and orient the code, and covering any of them breaks it outright regardless of error correction level.
Note the trade-off: higher error correction means more redundant data, which means more modules, which means a denser code that needs to be printed larger to remain scannable. Level H with a long URL produces a dense grid that will fail at business-card size.
Sizing for print: the 10-to-1 rule
The working industry guideline is that maximum scanning distance is roughly ten times the width of the code. A 2 cm code scans reliably from about 20 cm — fine for a business card or a menu. A poster meant to be scanned from three metres away needs a code around 30 cm wide.
Two things that break codes in print more often than size:
- The quiet zone. QR codes need a clear margin of at least four modules on every side. Designers routinely crop it away or run artwork right up to the edge, and the code stops working. The blank border is part of the code, not padding.
- Inverted colours. Dark code on a light background is the specification. Light-on-dark works on many modern phone cameras and fails on plenty of older scanners and apps. If the code is going on something expensive to reprint, do not invert it.
Keep the encoded URL as short as you reasonably can. Fewer characters means fewer modules, a less dense code, and a more forgiving scan — a long tracking URL loaded with UTM parameters can double the density for no visible benefit.
A security note: QR phishing is now a real attack vector
Because a QR code is unreadable to a human, it moves a link past the instinct people have learned to apply to suspicious URLs in email. Attackers exploit this directly — commonly called quishing — by embedding phishing links in codes on emailed PDFs, or by printing sticker overlays on legitimate codes in car parks, restaurants and on payment signage.
Practical defences worth knowing, and worth telling your staff: always check the URL preview your camera shows before opening it; be suspicious of any code that leads to a login or payment page; and physically check whether a code in a public place is a sticker sitting on top of printed artwork. If you are deploying codes for a business, a short branded domain helps customers verify the destination is really yours. Our email header analyser covers the same problem on the email side.
Frequently asked questions
Why does a QR code still scan with a logo covering the middle?
Because of Reed-Solomon error correction, which builds redundancy into the code. At level H roughly 30% of the code can be obscured and still reconstruct. Keep the logo under about 20 to 25% of the area and centred, and never cover the three large corner squares — those are the finder patterns and covering one breaks the code entirely.
How big should I print a QR code?
Maximum reliable scanning distance is roughly ten times the code width. A 2 cm code works from about 20 cm, which suits cards and menus. For a poster scanned from three metres, print it around 30 cm wide. Also preserve the quiet zone — the clear margin of at least four modules on every side is part of the code, not decoration.
Do the QR codes made here expire?
No. These are static codes, meaning the destination is encoded directly into the pattern itself. There is no redirect service in between, so nothing can expire, be disabled, or start charging you. The trade-off is that the destination cannot be changed after printing — if you need to change it later, you need a dynamic code from a service that maintains the redirect.
Are these codes tracked or watermarked?
Neither. Generation happens entirely in your browser, no code or destination URL is transmitted or logged, and there is no watermark, attribution mark or usage limit. The output is yours to use commercially.
Can a QR code be malicious?
The code itself is just encoded text, but it can encode a link to anything, and since humans cannot read it directly it bypasses the caution people apply to visible URLs. QR phishing is now common, both in emailed documents and as sticker overlays on genuine codes in public places. Always read the URL preview before opening, and treat any code leading to a login or payment page with suspicion.