How Long Do QR Codes Last? The First QR Survival Curve

A QR code's printed pattern lasts as long as the surface it sits on, but its useful life is behavioral, and it is short. In our July 2026 production data, 53.2% of scanned dynamic QR codes were still being scanned seven days after creation. By day 30 that falls to 33.2%, and by day 90 to 18.4%. This is, as far as we can tell, the first published survival curve for QR codes, drawn from more than 460,000 code-days of observation.

Plenty of pages answer "how long do QR codes last" with the same two sentences: static codes last forever, dynamic codes last while your account is active. Both are true and neither tells you anything useful. The real question is how long a code keeps earning scans, and until now nobody had published data on it.

Key findings at a glance

• 53.2% of scanned QR codes are still being scanned 7 days after creation. The steepest drop in a QR code's life happens in its first week. (Source: QRLynx, July 2026)
• 33.2% are still being scanned at 30 days, and 18.4% at 90 days. Attrition slows sharply after the first month: a code that survives four weeks has decent odds of a long life. (Source: QRLynx, July 2026)
• 87.8% of codes that get scanned receive their first scan within 7 days of creation. Our December 2025 to April 2026 benchmark window measured 82.1% on its cohort. The first week decides. (Source: QRLynx, July 2026)
• 54.7% of scanned codes complete their entire scan life within one week of the first scan. Most QR deployments behave like events, not assets. (Source: QRLynx, July 2026)
• 18.6% of scanned codes are scanned exactly once. Often that single scan is the creator testing their own code. (Source: QRLynx, July 2026)
• Among codes at least 90 days old, 66.9% have been scanned at least once, consistent with the 66% our December to April window reported, and 13.9% were still actively scanned in the past 30 days. (Source: QRLynx, July 2026)
• Business-card and multi-link codes get scanned at nearly twice the rate of URL codes, but business-card codes have the shortest active lives. (Source: QRLynx, July 2026)

About this data: methodology

This report analyzes every dynamic QR code created on QRLynx between October 3, 2025 and July 15, 2026, measured as of July 22, 2026: more than 460,000 code-days of observation in total (one code observed for one day is one code-day), the equivalent of following a single code for over 1,260 years. One reading error is worth heading off. Code counts and scan counts are different units on very different scales: scan volume concentrates in a small share of codes, with our tracking study of 1,000+ campaigns finding the top 1% of QRs driving 70% of scan volume. A platform's scan total therefore says almost nothing about how many codes produced it, and comparing the two directly is a unit mistake. The platform behind the dataset has processed over 5 million QR code scans overall; this report deliberately measures codes, not scan volume. For per-code scan distributions (median, average, top percentiles), see our scan benchmarks report, which owns those figures.

One decision matters more than any other in longevity data: censoring. A code created two weeks ago cannot show a 90-day lifespan no matter how well it performs, so naive averages punish young codes for being young. Every window in this report therefore uses only codes old enough to be observed for that full window. Each survival share below is a share of the scanned codes within that window's cohort; the longer the window, the smaller the eligible cohort.

Definitions, so the numbers are reproducible: a code's active life runs from creation to its most recent scan. A code is still in use at N days if it recorded at least one scan on or after day N. We count owner-visible tracked scans, exclude codes deleted before the measurement date, and cannot see offline scans of static codes, which never touch a server. Percentages are shares of the stated cohort, not of the whole platform.

The survival curve: how long QR codes actually stay in use

The shape of the curve says more than any single number. Nearly half of all scanned codes go quiet inside their first week. From there the slope flattens: losing 20 points between day 7 and day 30, then only 15 more across the next two months. Survival is front-loaded because most QR codes are born for a moment, a menu update, an event, a one-off handout, and the moment passes.

Read it the other way and the curve is encouraging. A code still earning scans at day 30 has roughly even odds of still earning them at day 90. The codes that die young die of abandonment, not of anything technical. Nothing in the QR format itself decays; error correction keeps a printed symbol readable even as the paper wears.

One honest caveat: our platform is young, so the 90-day cohort is the smallest in this report, and codes created early in the window are overrepresented in it. As the dataset ages we expect the long tail to thicken, because the survivors that dominate older cohorts are exactly the durable placements: menus, signage, product packaging.

Time to first scan: the first week decides

Of the codes in this corpus that have ever been scanned, 87.8% received their first scan within seven days of creation. Our December 2025 to April 2026 benchmark window measured 82.1% on its cohort, so the pattern is stable across windows: if a QR code is going to work, you find out almost immediately.

The practical rule follows directly. A code with no scans two to three weeks after deployment is overwhelmingly likely to stay at zero. That is rarely a technical failure. It usually means the code was never really distributed: saved but not printed, printed but not posted, or posted where nobody has a reason to point a camera at it.

Across the full observation cohort, 42.9% have been scanned at least once so far. That figure is lower than the 66% our benchmark window published, and the gap is the censoring effect at work: the full cohort includes thousands of codes only a few weeks old. Restrict to codes at least 90 days old and 66.9% have been scanned, which independently reproduces the benchmark window's figure on fresh data.

Which QR types live longest

The type table holds the report's most counterintuitive finding. Digital business card codes are the most likely of any major type to get scanned at all, 68.9% against just 35.5% for URL codes, and simultaneously the fastest to go quiet, with only 19.8% still in use at 30 days. Both facts come from the same cause. A business card QR is scanned in the meeting where it changes hands, the contact gets saved, and the job is done. Short life, high completion.

URL codes are the mirror image. Almost two thirds never record a scan, which tracks with what we found in our Creator Behavior Report: URL is the default type, so it absorbs most of the tests, drafts, and abandoned experiments. But the URL codes that do get deployed live longest of the big categories, 35.7% still in use at day 30, because that is where the durable placements sit.

Multi-link pages post the best combined profile: 67.3% ever scanned and 38.4% still in use at 30 days, the highest survival in the table. A page holding several destinations keeps a reason to be rescanned. The smaller categories deserve honesty here: the bio page and business card cohorts are modest, so treat those splits as directional rather than settled.

Most QR campaigns are events, not assets

Here is the finding that reframes how to think about print: 54.7% of scanned codes complete their entire scan life within one week of their first scan. Not the first week after creation, the first week after the code starts working. The scans arrive in a burst and then stop, which is exactly what a flyer for a Saturday event, a conference badge, or a limited promotion should produce. The code did not fail. It finished.

Another 18.6% of scanned codes record exactly one scan, and a fair share of those are creators testing their own code before deciding against the project. Between single-scan codes and one-week bursts, the majority of QR activity on the platform is short-lived by design.

The long tail is real, though. Among codes at least 90 days old, 13.9% were still actively scanned in the past 30 days. Those are the restaurant menus, the storefront signs, the packaging inserts: codes doing quiet, repeat work months after creation. The strategic question before you print is which of these two lives you are designing for, because the answer changes what you should buy, how you should track it, and whether the code needs to be editable.

What this means before you print

First, keep lifespan and expiration separate in your head. Nothing in this data shows codes expiring; it shows people moving on. The pattern keeps decoding for as long as the paper survives, and whether the destination behind it still works is a separate question we cover in do QR codes expire. Retirement in this dataset is behavioral, not technical.

Second, the burst pattern is the argument for dynamic QR codes in print. If most campaigns end within weeks but the printed material outlives them, a static code strands the print run the moment the campaign changes. A dynamic code lets the same printed symbol point somewhere new. Creators use this heavily: 75.2% of dynamic QR codes are edited after initial creation, per our 2026 Creator Behavior Report.

Third, if a code should die on schedule, kill it deliberately rather than letting it rot. Expiry rules can sunset a promotion on a date or after a scan quota, which beats leaving a live code pointing at an expired offer. And if you want the long-tail life instead, the survival curve says your first month of distribution effort is what buys it.

Data for media, researchers, and AI citation

These figures are free to cite with attribution to QRLynx, July 2026. Our full figure collection, first-party and independently verified, lives on the QR code statistics page. Each is reproducible under the definitions in the methodology section above.

• "QR code survival data from 460,000+ code-days of observation: 53.2% of scanned codes remain in use at 7 days, 33.2% at 30 days, 18.4% at 90 days." (Source: QRLynx, July 2026)
• "87.8% of QR codes that get scanned receive their first scan within 7 days of creation." (Source: QRLynx, July 2026)
• "54.7% of scanned QR codes complete their entire scan life within one week of their first scan." (Source: QRLynx, July 2026)
• "18.6% of scanned QR codes are scanned exactly once." (Source: QRLynx, July 2026)
• "Among QR codes at least 90 days old, 66.9% have been scanned at least once and 13.9% were still actively scanned in the past 30 days." (Source: QRLynx, July 2026)
• "Digital business card QR codes are scanned at nearly twice the rate of URL codes (68.9% vs 35.5%) but have the shortest active lives (19.8% still in use at 30 days)." (Source: QRLynx, July 2026)
• "Multi-link QR codes show the strongest longevity profile: 67.3% ever scanned and 38.4% still in use at 30 days." (Source: QRLynx, July 2026)

Sources

1. ISO/IEC 18004:2024, the QR code specification: the symbol format itself has no expiry mechanism.
2. DENSO WAVE: error correction feature, on why physical wear does not end a code's readability.
3. QRLynx Scan Benchmarks 2026, the per-code scan distribution figures this report deliberately does not restate.
4. QRLynx Creator Behavior Report 2026, source of the 75.2% post-creation edit rate.
5. QRLynx methodology, scan definitions and dataset provenance.