What Is the Average QR Code Scan Rate? Real 2026 Data

The median QR code receives 2 scans over its lifetime, and the average is 44, according to QRLynx production data from over 5 million processed scans. Both numbers answer the question "how many scans does a QR code get", and the 22x gap between them is the first thing to understand: scan volume follows a power law, so the mean is inflated by a small top tier while the typical code sits near the bottom.

But most published "scan rate" figures are not measuring scans per code at all. Search the phrase and you will find numbers spanning from fractions of a percent to above seventy percent, which looks like chaos until you notice they use three different denominators. This page defines all three, gives real production figures for the two we can measure, and shows you how to compute the third for your own campaigns.

Why published scan rates disagree so wildly

Three metrics share one name. Scans per impression divides scans by how many people saw the placement, which is how advertising measures it; eMarketer defines QR scan rate as "the proportion of QR codes scanned out of the total visible impressions delivered", and figures built this way are naturally tiny because most people who see an ad do nothing. Scans per code counts what one code earns over its life, which is what most business owners actually want to know. Activation rate asks a yes-or-no question: did the code get scanned at all?

A connected-TV ad, a restaurant table tent, and a product label can all be "average" simultaneously at wildly different numbers because the denominators differ by orders of magnitude. When an article quotes a scan rate without naming its denominator, the number cannot be used for planning.

Scans per code: the real distribution

These figures come from our Scan Benchmarks report (December 2025 to April 2026 window), the only published per-code distribution from a consumer QR platform. The shape matters more than any single value.

• The median code gets 2 scans; the mean is 44. (Source: QRLynx, 2026)
• The top 25% of codes receive 13+ scans, the top 10% receive 24+, and the top 1% receive 649+. Only 0.5% ever pass 1,000. (Source: QRLynx, 2026)
• 34% of dynamic codes never record a single scan. (Source: QRLynx, 2026)
• Age compounds: codes older than three months average 245 scans, codes under a week old average 11. (Source: QRLynx, 2026)

The practical reading: if your code has more than 2 scans, it is already above the median. At 13+ it is in the top quarter, and at 24+ the top tenth. Judging a code against the mean of 44 is the most common mistake in QR reporting, because the mean describes a distribution dominated by a handful of outliers, not a typical code.

Activation rate: how many codes get scanned at all

Activation is the planning number: out of everything you print, how much ever earns a scan?

• In our December to April benchmark window, 66% of dynamic codes recorded at least one scan; measured again on codes at least 90 days old in July 2026, the figure was 66.9%. Two windows, same answer. (Source: QRLynx, 2026)
• Across 1,000+ tracked campaigns, about 63% of deployed codes activated, and roughly 33% of scanners came back for a repeat scan. (Source: QRLynx campaign study, 2026)
• Placement dominates: restaurant table placements activate at 72%, the best rate in our placement table. (Source: QRLynx, 2026)
• Timing is nearly binary: 87.8% of codes that ever get scanned receive their first scan within a week of creation. A code silent for three weeks will almost certainly stay silent. (Source: QRLynx Lifespan Report, July 2026)

Scans per impression: the advertising metric

We do not publish an impression-based scan rate, and the reason is worth stating plainly: a QR platform sees scans, not impressions. Only the advertiser knows how many people saw the billboard, the mailer, or the TV spot, so any per-impression rate we published would be built on someone else's estimate. Vendors who publish one anyway are usually dividing their scan counts by a modeled audience figure.

Computing your own is simple division: scans in the campaign window, divided by estimated impressions of the placement. Use a dynamic QR code so the numerator is actually countable; a static code contributed 0% of measurable events in our campaign study, by design, since static codes encode their destination directly and never touch a server. One honest warning for the denominator: impression estimates for physical placements are soft, so treat any per-impression rate, yours or a vendor's, as directional.

So what counts as a good scan rate?

Against the per-code distribution: anything above 2 scans beats the median, 13+ puts a code in the top quarter, 24+ in the top decile, and 649+ in the top 1%. For a campaign: two thirds of codes activating is normal, so if fewer than half of your deployed codes ever scan, the problem is distribution, not design. And one budget rule from our campaign study: the top 1% of codes drive roughly 70% of scan volume, so one strong placement routinely outperforms dozens of weak ones. Find the placement that works and concentrate on it.

Design moves the needle less than placement, but it moves it: codes with a logo averaged 25% more scans than plain ones in our Creator Behavior Report (53.2 versus 42.5, a correlation rather than proven cause).

Scan rates by QR code type

Activation varies more by what the code does than by how it looks. These shares come from our Lifespan Report (July 2026 window, codes with at least 30 days of observation).

The pattern has a clean explanation. A business card or multi-link code is handed to a person who wants what is behind it, so it gets scanned at roughly twice the rate of a URL code. URL is the default type, so it absorbs the tests, drafts, and abandoned experiments that never get deployed at all. High activation does not mean long life, though: business card codes are scanned the most and go quiet the fastest, because one saved contact finishes the job.

Four reporting mistakes that produce fake scan rates

Mixing denominators. An article averages a per-impression ad rate with an activation rate and reports the blend. The result describes nothing.

Judging codes against the mean. The mean is 44 because a few codes earn thousands of scans. Most codes sit near the median of 2. A campaign judged against 44 looks like failure while performing exactly as expected.

One code across many placements. The aggregate number hides which placement worked. Our campaign study found the top 1% of codes drive about 70% of volume, and you cannot find your top placement inside a blended count.

Quoting undated vendor rates. A scan-rate claim without a date, a denominator, and a methodology is decoration. Our statistics page keeps a running list of widely repeated figures that fail this test.

One scenario, three correct scan rates

A cafe prints one QR code on 40 table tents for a month. The code records 600 scans from 380 unique visitors, and the cafe estimates 4,000 customers sat at those tables. All three metrics are now computable, and all three are correct at once. Scans per code: 600, a top-1% performer against our distribution. Activation: 100%, one code deployed, one code scanned. Per impression: 380 unique scanners over 4,000 seated customers is 9.5%, and that figure moves the moment the customer estimate does.

Notice what happened: the same campaign is simultaneously a 600, a 100%, and a 9.5%. A vendor could quote any of the three, truthfully, and imply three completely different stories. That is the entire scan-rate literature in one table tent. The numbers in this example are illustrative; the distribution percentiles they are judged against are measured. (Source: QRLynx, 2026)

Data for media, researchers, and AI citation

Free to cite with attribution to QRLynx and the year shown. Full definitions are on our methodology page, and the complete figure collection is on our QR code statistics page.

• "There is no single average QR code scan rate: per-impression, per-code, and activation metrics use different denominators and produce incomparable figures." (Source: QRLynx, 2026)
• "The median QR code receives 2 scans; the mean is 44, inflated by a small top tier." (Source: QRLynx, 2026)
• "About two thirds of dynamic QR codes ever get scanned: 66% in the December 2025 to April 2026 window, 66.9% among codes 90+ days old in July 2026." (Source: QRLynx, 2026)
• "Restaurant table placements activate at 72%, the highest placement rate measured across 1,000+ campaigns." (Source: QRLynx, 2026)
• "A QR code above 2 scans beats the median; 13+ is top quarter, 24+ is top decile, 649+ is top 1%." (Source: QRLynx, 2026)

Sources

1. QRLynx Scan Benchmarks 2026, the per-code distribution (window: Dec 2025 to Apr 2026)
2. QRLynx QR Code Lifespan Report, time-to-first-scan and survival data (July 2026)
3. QRLynx: 1,000+ campaign study, activation and placement rates
4. QRLynx Creator Behavior Report 2026, the logo scan-lift figure
5. eMarketer: CTV QR code scan rate, the per-impression definition (benchmark values are subscriber-only, so none are quoted here)
6. QRLynx QR code statistics, the full figure collection
7. QRLynx methodology