QR code landing pages: the half of the campaign that happens after the scan

Most advice about QR campaigns stops at the moment the camera recognises the code. Size, contrast, the call to action, the placement: all of it exists to win a scan. But a scan is not a result. The result happens on the page that loads next, on a phone, usually on a cellular connection, held by someone standing in a shop or at a bus stop. That page is the QR code landing page, and it is where most campaigns quietly die.

The numbers suggest agencies know this and still look away. In Bitly's 2025 survey of marketers, only 31 per cent said they monitor what happens after the scan, and just 16 per cent tie QR campaigns to revenue. The scan count goes in the client report because it is easy to get, not because it is the number the client paid for. This guide covers the other half of the campaign: what the page after the scan needs to do, the specs that make it work on a phone, and how to prove it converted.

Homepage or landing page?

The first decision is the one most often fumbled: where should the code point? The honest answer is almost never the homepage.

A homepage is built to serve everyone: existing customers, job applicants, investors, people who typed the brand into Google. It offers navigation, not an answer. Someone who scanned a code next to "Get 20% off your first order" does not want a menu of departments. They want the 20 per cent, and they want it on the first screen.

A QR code landing page is a single page built for a single campaign, with one job. It repeats the promise that was printed next to the code, it asks for one action, and it removes everything else. If the visitor cannot complete the action the print material promised, the page has failed, no matter how good it looks.

There is a practical exception. If the print promise genuinely is "visit our website" (a brand awareness play with no specific offer), the homepage can be the right destination. But then measure it as awareness, not response, and say so in the report. Our guide to which scan metrics matter covers that distinction in depth.

The placement and the page are one funnel

Information foraging research from Xerox PARC in the 1990s gave this idea its academic backbone: people follow "information scent", and they abandon a trail the moment the scent breaks. In landing page circles the applied version is called message match, a term popularised by Unbounce. The principle is simple: the page must visibly continue the promise that triggered the click. No controlled study puts a reliable percentage on it, so treat it as a strong rule of thumb rather than a law, but the logic is hard to argue with.

For QR campaigns the "ad" is a physical object, which makes message match more literal than in paid search. The scanner was just looking at a poster, a menu, a package or a flyer. The page that loads should look like it belongs to that object: same offer, same wording, same visual world. If the poster says "Win a weekend in Vienna" and the page headline says "Welcome to our prize portal", the scent is broken and the back button is one thumb-flick away.

Context goes beyond wording. A code on a bus shelter is scanned by someone standing up, probably one-handed, possibly about to board. A code on packaging is scanned at home with time to read. The bus shelter page should ask for one tap; the packaging page can afford a paragraph. If the same campaign runs across placements, that is a reason to use one trackable code per placement, not just for attribution but because the pages themselves may need to differ.

Speed: the physics of a cellular connection

Every QR scan happens on a phone, by definition, and most happen away from good wifi. Speed is therefore not one optimisation among many; it is the gatekeeper for everything else on this page.

The most quoted number in mobile performance is Google's 2016 finding that 53 per cent of mobile visits are abandoned when a page takes longer than three seconds to load. It is a real study, but it is a decade old, so treat it as evidence that patience is short rather than as a current benchmark. The current bar is Google's Core Web Vitals: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint within 200 milliseconds, and Cumulative Layout Shift under 0.1, each measured at the 75th percentile of real page loads. That 75th percentile detail matters for QR work: it is field data, so slow cellular connections are in the sample, not averaged away. Akamai's 2017 retail study found a 100 millisecond delay could cut conversions by up to 7 per cent; dated as well, but directionally right.

Redirect chains are the QR-specific speed trap. A trackable QR code already spends one redirect hop by design: the code resolves a short URL, the platform logs the scan, then forwards to the destination. That hop is what buys you analytics and the ability to change the destination later, and it is worth it. What is not worth it is stacking more hops behind it: a vanity domain that bounces to www, an http URL that upgrades to https, a link shortener pointed at another link shortener. Google's Lighthouse documentation is blunt about it: each redirect adds a full request-response round trip and can delay the page by hundreds of milliseconds, which on cellular is often worse. Audit the chain once: scan the real printed code on a phone with wifi off and count the address bar changes. One hop is the price of tracking; three is negligence.

The page itself should be built like it expects a bad connection: compressed images sized for a phone screen, no autoplaying video above the fold, no third-party scripts that block rendering. Test it with the throttling tools in Chrome DevTools or PageSpeed Insights, not on the office network.

The first screen does the work

When the page loads, the scanner sees one phone screen, roughly 360 by 700 CSS pixels of usable space. Whatever you need them to understand and do has to be obvious within that rectangle.

1. The headline repeats the printed promise, near verbatim. This is the message match check: someone who scanned "20% off your first order" should read those words again within a second of the page loading.
2. One supporting line or visual carries the detail: what the offer is, what happens when they tap. Keep it to a sentence. The scanner gave you their curiosity, not their afternoon.
3. A single call to action sits in the lower half of the screen, inside the natural thumb arc. Observational research by Steven Hoober found about half of phone users operate one-handed with the thumb, so the primary button belongs where a thumb rests, not in a top corner. Make it comfortably tappable: Apple's Human Interface Guidelines say at least 44 by 44 points, Google's Material Design says 48 by 48 density-independent pixels, and WCAG 2.2 sets 24 by 24 CSS pixels as the accessibility floor. Build to the Apple and Google numbers; the WCAG figure is a minimum, not a target.
4. If the action is a form, every field is a toll booth. Baymard Institute's checkout research found the average ecommerce checkout shows 11.3 form fields when most sites could manage with 8. For a QR landing page, the defensible set is usually one or two fields. Ask for what the campaign needs to deliver its promise, and collect the rest later.
5. Anything below the first screen is supporting material: details, terms, trust signals, secondary links. People who scroll were already convinced; the first screen is for everyone else.

One deliberate omission from that list: a navigation bar. A campaign page does not need one. Every exit you add is a path away from the one action the client is paying for.

Consent banners and popups: legally required, badly built

If the page sets non-essential cookies or tracks visitors, EU rules (the ePrivacy Directive and the GDPR) require consent before it happens, and a QR landing page is not exempt. What you can control is the cost. The consent banner is literally the first tap a scanner makes on your client's brand, and field research backs up how much its design matters: a 2019 study of more than 80,000 users found that banner position and button design significantly change how people interact with it. Make it dismissible in one thumb-reach tap, do not let it cover the headline, and remember that the less the page tracks, the less it has to ask. A scan counted server-side by your QR platform does not need a cookie banner at all; it is the marketing pixels on the landing page that summon one.

Marketing popups are a different matter: newsletter overlays, app install interstitials, "before you go" dialogs. Since January 2017 Google has applied a ranking penalty to pages where intrusive interstitials obscure content on mobile, with explicit exemptions for legally required notices like cookie consent and age verification. The SEO penalty is one reason to drop them. The better reason is that a person three seconds into a scanned page who gets an overlay shoved at them will leave, and your analytics will record a scan, a visit, and nothing else.

Proving the page worked

A scan that bounces is a vanity metric. The campaign's real funnel has four stages, and each one is measurable: scans (your QR platform logs these at the redirect, before the page even loads), page loads, the conversion action, and, if the client sells anything, revenue.

The wiring is the same as for any campaign destination: put UTM parameters on the destination URL so the session is attributed in GA4, then register the page's action (form submit, order, click-through) as a key event. We cover the exact setup, including why untagged QR traffic lands in "direct", in our guide to QR codes in Google Analytics 4. With a dynamic QR platform the UTM-tagged URL lives behind the short link, so the printed code never changes even when the tagging does.

Two honest caveats for the client report. First, scans and sessions will not match: some people abandon during the load (that gap is itself a speed metric worth watching), and some sessions get blocked by consent choices. Report both numbers and say why they differ. Second, benchmarks deserve scepticism. Unbounce's 2024 conversion benchmark, drawn from 41,000 landing pages, puts the median conversion rate at 6.6 per cent, ranging by industry from roughly 4 to 12 per cent. That is a useful sanity check for a landing page as a page. A QR-specific "scan-to-action benchmark" is a different thing, and no platform with real data publishes one, so any confident industry figure you read was invented. Set the client's baseline from their own first campaign instead.

The page is iterable, the print is not

The asymmetry that makes QR landing pages worth an agency's attention: the printed code is fixed for the life of the material, but with a dynamic QR code the page behind it can change any day of the campaign. Weak headline? Rewrite it. Long form killing conversions? Cut it to one field. Wrong page entirely? Point the code somewhere else without reprinting.

That mutability is also what makes real experiments possible mid-flight: split traffic behind one printed code, or run placement variants against each other, and let scan-stamped data decide. We wrote a full guide to A/B testing a QR campaign without fooling yourself; the short rule is to change one thing at a time and to give variants enough scans to mean something.

This is also the answer to a budget objection agencies hear: "we already paid for the print, the page is extra". The print run is the expensive, frozen part. The page is the cheap, improvable part. Spending the whole budget on the half you cannot fix and nothing on the half you can is exactly backwards. If the scan volume itself is the problem, that is a different fix: the offer, the call to action and the placement decide whether anyone scans at all.

Frequently asked questions

Should a QR code go to my homepage or a landing page?

A dedicated landing page, in almost every case. A homepage serves every possible visitor and therefore answers none of them quickly; a landing page repeats the specific promise printed next to the code and asks for one action. The exception is a pure brand awareness campaign whose printed promise really is just "visit us", and even then a tracked destination tells you more than an untracked homepage.

What should a QR code landing page include?

Five things: a headline that repeats the printed offer, one line of supporting detail, a single call to action sized at least 44 by 44 points and placed in thumb reach, the shortest form the offer allows (one or two fields, or none), and supporting detail below the first screen. Leave out site navigation and marketing popups; both leak visitors from the one action that matters.

How fast should a QR code landing page load?

Aim for Google's Core Web Vitals thresholds: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint within 200 milliseconds, and layout shift under 0.1, at the 75th percentile of real loads. QR traffic is entirely mobile and often on cellular, so test with network throttling on, and keep the redirect chain to the single hop your QR platform needs for tracking.

Do I need a website to use a QR code?

No. The code needs a URL, not a website. Many campaigns point at a hosted single page, a form, a file or a booking tool. For anything with a print run, use a dynamic QR code so the destination stays editable after printing, whatever it points at today.

What is a good conversion rate for a QR code landing page?

There is no trustworthy QR-specific benchmark; no major platform publishes scan-to-action data, so quoted figures are guesses. As a sanity check for the page itself, Unbounce's 2024 benchmark across 41,000 landing pages puts the median conversion rate at 6.6 per cent, varying by industry from about 4 to 12 per cent. Treat your client's first campaign as the baseline and improve against it.

Why do people scan my QR code but not convert?

Work through the funnel in order. If scans are high but page loads lag, the page is too slow for cellular or the redirect chain is too long. If loads are fine but nobody acts, check message match (does the headline repeat the printed promise?), the call to action (one, visible, tappable?) and the form length. Scan data tells you which placements convert worst; fix those pages first.

Does a QR code landing page need a cookie banner?

Only if it sets non-essential cookies or runs tracking scripts, which most marketing pages do. The scan itself can be counted server-side by your QR platform without any banner. If you do need consent, make the banner dismissible in one tap and keep it off the headline; it is the first interaction a scanner has with the page.

The short version

A QR campaign is two halves: the print that wins the scan, and the page that turns the scan into a result. The page is a single-purpose mobile page, not the homepage. It repeats the printed promise word for word, loads inside the Core Web Vitals thresholds on a throttled connection, puts one thumb-sized call to action on the first screen, and asks for at most a field or two. It carries UTM parameters so conversions land in analytics, and it sits behind a dynamic code so you can rewrite, retarget or replace it without touching the print run. Before the next campaign ships, scan the real code on a phone with wifi off and walk the funnel yourself: load, read, tap, convert. If any step makes you hesitate, it is costing the client conversions. Fix that page this week; the print can wait until the next run.