Your QR Campaign Is Just a Guess: How to A/B Test It [2026]

You printed 5,000 flyers. You put a QR code on them. A month later someone asks, "How did the campaign go?" and you realize you have no idea.

You know some people scanned it, maybe, but you cannot tell if the blue flyer outperformed the red one, if the code on the front got more attention than the one on the back, or if a single person bought anything after scanning. Your campaign is a black box. And if you are using a static QR code, that black box is locked shut.

Why most QR campaigns are impossible to measure

A static QR code bakes the destination URL directly into the black-and-white pattern. Once it is printed, it is permanent. You cannot change where it points. You cannot compare two versions. You cannot even count scans unless you bolt UTM parameters onto the URL and hope your analytics software separates QR traffic from everything else.

This is not a small problem. In 2026, 102.6 million Americans will scan QR codes and 72% of consumers report scanning one in the past month. The audience is there. But only 1 in 8 marketers actually connects QR scans to revenue. The rest are flying blind.

The result is wasted budget. You reprint the same flyer because you do not know what worked. You miss the placement that would have doubled your scans. You keep sending traffic to a landing page that never converted in the first place.

What A/B testing a QR code actually means

A/B testing a QR campaign is simple in concept. You create two versions of the same campaign, change exactly one thing, split your audience evenly between them, and measure which version gets more scans and, more importantly, more conversions.

The variable you test could be:

• Design - branded color QR code vs plain black-and-white
• Placement - front of a flyer vs back, table tent vs receipt
• Destination - product page vs dedicated landing page
• Call to action - "Scan for 20% off" vs "Scan for a free gift"
• Size - small code in a corner vs large centered code

The key is testing one thing at a time. If you change the design, the placement, and the destination all at once, you will not know which change drove the result.

Why you need dynamic QR codes to do this

You cannot A/B test with static codes. If you print two different static codes, you can count scans, but you cannot redirect the loser to the winner after the fact. You cannot fix a broken landing page on the losing variant. You cannot extend the test if one variant is trending toward significance. The code is frozen.

Dynamic QR codes work differently. The pattern stores a short link, not the final URL. When someone scans, the short link checks your dashboard settings and redirects to the current destination. That redirect layer is what makes A/B testing possible.

Here is what dynamic codes give you:

• Edit after printing - change the destination anytime without reprinting
• Scan analytics - count, time, location, and device for every scan
• UTM-friendly URLs - feed clean data into Google Analytics 4
• Multiple codes, one dashboard - compare variants side by side

Dynamic QR codes now hold 65% of the business QR market, and that share is growing for exactly this reason. Marketers are tired of guessing.

If you are not sure whether dynamic or static is right for you, our guide on /blog/dynamic-vs-static-qr-codes breaks down the full comparison.

A simple 5-step QR A/B test you can run this week

1. Pick your variable

Choose one thing to test. If this is your first campaign, start with design. It is the highest-impact change and you can see results quickly.

2. Create two dynamic QR codes

Generate two dynamic codes in your QRhubly dashboard. Point variant A to your control URL and variant B to the same URL (or a different one if you are testing destination). Add UTM parameters so Google Analytics can separate the traffic. Our post on /blog/qr-code-utm-tracking-google-analytics shows exactly how to structure those parameters.

3. Split your audience evenly

Print equal quantities of each variant. If you are testing digital placement, rotate the two codes evenly across time slots or locations. Do not let bias creep in. Same paper stock, same printer, same everything except the one variable you chose.

4. Run for a set period

Let the test run until you hit statistical significance. For most small business campaigns, that means at least 100 scans per variant. If your scan volume is low, run the test for two to four weeks rather than rushing to a conclusion.

5. Compare and roll out the winner

Look at scan count first. Then look at what happened after the scan. Did one variant drive more purchases, more signups, or longer page sessions? The winning code is the one that moves your actual business metric, not just the one that got more scans.

Common mistakes that ruin QR A/B tests

Testing too many things at once. If you change the color, the placement, and the offer all in one go, you will never know what actually moved the needle. Pick one variable. Run it. Learn. Then test the next.

Calling the test too early. Fifteen scans is not a trend. It is noise. Wait for a sample size that means something. If you are not sure how many scans you need, look up your industry's baseline scan rates in our /blog/qr-code-scan-rate-benchmarks-by-industry-2026 post and work backward from there.

Ignoring post-scan behavior. A code that gets 200 scans but zero conversions is not a winner. It is a traffic leak. Always tie your QR test to a downstream metric you care about: sales, signups, bookings, or time on page.

Using static codes. We will say it again: if the code is static, the test is final the moment it leaves the printer. A dynamic code lets you pivot, extend, or fix the test while it is still live.

What changes once you start testing

The first time you run a proper QR A/B test, you will probably be surprised. The version you liked best is not always the version your customers scan most. The small code on the receipt might outperform the giant one on the poster. The boring landing page might convert better than the flashy one.

That is the point. You are no longer designing by instinct. You are designing by data.

Over time, these small wins compound. A 30% lift from a better design, combined with a 20% lift from a better placement, combined with a 15% lift from a better landing page. That is the difference between a campaign that breaks even and one that pays for itself three times over.

If you want to understand what scan analytics can tell you beyond simple counts, read our full guide on /blog/how-to-track-qr-code-scans.

FAQ

Can I A/B test a QR code after it is already printed?

Yes, but only if you used dynamic QR codes. With dynamic codes you can change the destination, pause a variant, or redirect both codes to the winning URL after the test ends. Static codes do not allow any of this.

How long should I run a QR A/B test?

Run it until you have enough data to trust the result. For most small businesses, that means at least 100 scans per variant or two to four weeks, whichever comes first. Low-traffic campaigns may need longer.

What metrics should I track?

Start with total scans and unique scans from your QR platform's dashboard. Then look at downstream metrics in Google Analytics 4: conversion rate, session duration, bounce rate, and revenue per session. The scan is just the beginning.

Do I need two different QR codes?

Yes. Each variant needs its own dynamic code so you can track them separately. In QRhubly you can create, label, and compare multiple codes from one dashboard.

Can I do this with free QR codes?

Free static generators cannot A/B test in any meaningful way because the destination is locked and there are no analytics. You need dynamic QR codes with built-in scan tracking. QRhubly gives you dynamic codes free for 7 days or 50 scans so you can run your first test without paying.