How to Make a Flyer That Converts, With a QR Code That Scans

What does “a flyer that converts” actually mean?

It means one measurable action from a small share of the people who see the flyer. Industry-tracked UK figures put that share below one per cent for unaddressed leaflets, so “converts” should mean “produces a response you can count and compare between batches”, not “most readers act”. Plan the print run and the offer around that.

According to JICMAIL’s Q1 2026 report, door drops measured by its Response Rate Tracker averaged a 0.3% response over 2023–2025. The UK’s Data & Marketing Association, summarising its Door Drop Report 2025, gives 0.5% for 2021–2024. Both are leaflets through letterboxes, tracked by the advertisers who sent them. We found no published benchmark for flyers handed out at a counter or pinned to a noticeboard, so treat these as the best available reference point, not a forecast.

Our own arithmetic on those rates: 1,000 flyers at 0.3–0.5% is three to five responses. That is why the rest of this guide cares about small gains, and about giving each batch its own code so you can tell where those few responses came from.

What makes people notice a flyer in the first place?

The size of the text, more than the size of the picture. In a large eye-tracking study of print adverts (1,363 adverts, 3,600+ consumers), pictures caught the eye at any size, but text earned attention in proportion to the space it was given. Spend the area on the offer headline and keep the photo supporting, not dominant.

That study is Pieters and Wedel (2004, Journal of Marketing): 1,363 print adverts tested with infrared eye tracking on more than 3,600 consumers. We have read its abstract, which reports that “only increments in the text element’s surface size produce a net gain in attention to the advertisement as a whole.” It was run on magazine adverts, not flyers, but it is the closest large-sample print evidence there is.

How should a flyer be laid out?

In three levels of importance, set by size and contrast. The offer comes first, the QR code and its label second, and everything else (who, where, terms) third. Someone glancing at it for a moment should be able to name the first two. Do not design to an “F” or “Z” reading pattern; those describe pages that have no hierarchy.

Nielsen Norman Group, whose web eye tracking produced the F-pattern, says in a 2017 update that it appears on text with “little or no formatting” and that “the F-pattern is negative for users and businesses.” The Z-pattern, or Gutenberg diagram, fares worse: the design reference Universal Principles of Design says “there is little empirical evidence” that it improves reading.

So the common advice to put the QR code bottom-right “because that is where the eye ends up” has no study behind it. We could not find any study of QR placement on a flyer. Our recommendation is the lower half of the page, where the reader arrives after the offer, inside its own panel with the label attached. That is a judgement call, not a finding.

What should the offer on a flyer say?

One specific thing, with a real number, a real date and, where it is true, a real previous price or a local rating. “Your first coffee is on us until the end of the month” does more work than “Great coffee deals”. Concrete claims are judged more believable, and a genuine limit makes an offer feel worth acting on now.

The studies in this section were read as their published abstracts, and none of them was run on flyers. Each point below is the finding, then how we apply it.

• Be concrete. Hansen and Wänke (2010) found that “statements of the very same content were judged as more probably true when they were written in concrete language than when they were written in abstract language.” Use exact figures, and only figures that are true.
• Frame the saving for the price. González and colleagues (2016) found that for a higher-priced product people preferred a discount shown as an amount off rather than a percentage. Under $100 the reverse pattern was not statistically significant, so the popular “percent off under $100, dollars off over $100” rule is only half supported.
• Show a genuine regular price. In the classic study by Urbany, Bearden and Weilbaker (1988), a plausible reference price raised the perceived value of the offer. It has to be the real price; an inflated “was” price is a consumer-protection problem, not a tactic.
• Add one true limit. Inman, Peter and Raghubir (1997) found that purchase limits, time limits and conditions “serve to accentuate deal value”, mainly for readers who are not analysing the offer closely. Put one line directly under the offer: an end date or a number of places.
• Make social proof local. In the hotel field experiments of Goldstein, Cialdini and Griskevicius (2008), printed signs describing what other guests did beat a standard environmental appeal, and worked best when the group described matched the reader’s own situation. On a flyer, a true rating from customers in your area beats a generic testimonial.
• Lead with what they get. Loss framing (“Don’t miss out!”) is not a proven lift: a meta-analysis by O’Keefe and Jensen (2006), covering 165 effect sizes and 50,780 people, found loss-framed appeals “are not generally more persuasive than gain-framed appeals.” Most of those were health messages.

How many calls to action should a flyer have?

One. Give the scan the visual weight, and let the phone number and address sit in a small footer line. A short web address printed under the code is not a second call to action; it is a second way into the same page, and in the mail experiments covered below, printing both helped a little or made no significant difference.

The closest direct evidence comes from mail surveys, which are printed pieces asking for a real response. Medway and Fulton (2012) pooled nineteen experimental comparisons and found that mail surveys offering a web option alongside paper had “significantly lower response rates” than those that did not. Their abstract suggests the extra option “may increase the complexity of responding”.

The famous jam study by Iyengar and Lepper (2000) found more purchases from 6 choices than from 24 or 30, but a meta-analysis of 50 experiments (N = 5,036) by Scheibehenne, Greifeneder and Todd found “a mean effect size of virtually zero”. Later work shows choice overload appears only under certain conditions, and a reader glancing at a flyer with no commitment to it plausibly fits them; that is our reading, not a tested result.

How big should the text on a flyer be?

Big enough for the distance it will be read from. By our arithmetic on the reading research, a handout read in the hand can use 10–12 pt body text, while a flyer on a noticeboard read from a metre or more needs key details around 20 pt and a headline of 60 pt or more. Keep text dark on a light background.

The reading research is Legge and Bigelow (2011, Journal of Vision). They report that fluent reading spans “approximately 0.2° to 2°” of letter height, which at a 40 cm reading distance means x-heights of 1.4 mm (4 points) to 14 mm (40 points). Scaling their lower bound to other distances is our own arithmetic, not their table. Treat each row as a floor for normally sighted adults in good light, and go larger.

A handout and a noticeboard flyer are therefore different designs, not the same file printed twice. For a window or noticeboard, enlarge the headline and the QR panel and cut the copy rather than shrinking it.

Keep body text dark on a light background and off the photo: Piepenbrock and colleagues (2013) found that “dark characters on light background lead to better legibility” regardless of the reader’s age (tested on screens; abstract read). And capitals are fine for a short headline, despite the common claim that all caps is harder to read: Arditi and Cho (2007) found upper-case text read faster than mixed case at small sizes near the limit of legibility, with the advantage disappearing at larger sizes (abstract read).

Do people actually scan QR codes on flyers?

Some do, and adding a code helps a little rather than a lot. In randomised experiments with printed invitations, adding a QR code raised response by between 1.3 and 3 percentage points, or made no significant difference. What a code reliably does is move people onto their phones, which makes the page behind it matter.

Nearly everyone has a phone that can scan: according to Pew Research Center, 91% of US adults owned a smartphone in its 2025 survey. Whether they will is a smaller number. Three experiments with printed survey invitations:

• Endres, Heiden and Park (2024, SMIF), a US statewide experiment: adding a QR code “resulted in a small (+1.31 percentage point) increase in overall participation”.
• Schmiedeberg, Schröder and Bozoyan (2025, Field Methods), a German panel: “The response rate was three percentage points higher with QR codes than without” (abstract read).
• Lewis and colleagues (2025, SMIF), Chicago: 28.0% with a QR code against 29.3% without, a difference that was “not significant”.

Those are survey invitations, often with incentives, not adverts. The fair reading is that a code is worth adding but is not a multiplier.

Scale matters too. In JICMAIL’s Q1 2026 data, 0.6% of all mail items led someone to use a QR code, against 9.4% that prompted a website visit. The base includes mail with no code on it, so the QR figure is a floor, but most mail-driven visits still arrive some other way.

How big should a QR code be on a flyer, and where should it go?

As large as the layout allows, with a clear margin of four modules on every side, and off any fold or glossy patch. We recommend at least 25 mm on a handout and 50–60 mm on a noticeboard version, a judgement based on the arithmetic below. The widely repeated “2 cm minimum” and “distance divided by 10” are rules of thumb, not standards.

Here is what the published standards actually say, next to the conventions that get quoted as if they were standards:

Sources: Denso Wave on the quiet zone, print size and error correction; the USPS guidebook; the GS1 guideline.

Why the length of the link changes the size you need

A QR code is a grid of square modules, and a longer link needs more of them. In our own calculation, a 63-character dynamic redirect link fits a 33-module code at level M, while an 85-character static link carrying tracking parameters needs 41 modules at the same level, and 53 at the highest. Printed 25 mm wide, the 33-module code has modules of about 0.76 mm, comfortably above GS1’s 0.495 mm target. The 53-module code squeezed into 20 mm drops to 0.38 mm, below GS1’s retail minimum. That is our arithmetic on a packaging standard, not a flyer standard, but the direction is clear: short link, standard error correction, and print it bigger than feels necessary.

Keep the code itself plain

The GS1 guideline describes the ideal as “highly contrasting light and dark modules, such as black and white” on “a perfect grid using even, square modules”, and says logos “are technically considered as damage, since they generally sit on top of the data”. It also warns that “surfaces with a sheen can reflect light and affect what the scanner can decode”, and to keep codes away from folds. Light-on-dark codes are allowed by the QR standard, but some decoding libraries only try an inverted image when asked to. Put the brand colour on the panel and the label, not on the code, and keep the panel at least 10 mm from the trim edge.

For export resolution, DPI and size by format in more detail, see our guide to printing a QR code at the right size.

What should the label next to a QR code say, and where should the scan lead?

The label should say what the scan gets: “Scan for your free coffee”, not “Scan me”. Print a short web address under the code for people who would rather type. Then send the scan to one fast mobile page that repeats the flyer’s offer, not to your homepage.

Honesty first: we found no controlled test of labelled against unlabelled QR codes. What exists is institutional practice. The USPS 2026 Integrated Technology promotion requires mailers to include “directional copy with each technology trigger”, with examples such as “Scan to shop our mobile website.” That shows what a large mailer insists on, not how much it helps.

The typed address matters more than people assume. In a 2025 US national survey of older adults reported by Hupp and colleagues, where both were printed, 62% of the people who completed online came in through the QR code and 38% typed the URL. Leave the address off and you are designing for six in ten. Keep it short enough to type on a phone.

The landing page is part of the flyer. A QR scan always arrives on a phone, and phone arrivals drop out more: in the Lewis Chicago experiment the QR condition produced about 77% more partly completed surveys (3.4% against 1.9%). A 2016 Google study found that 53% of mobile visits were abandoned when a page took more than 3 seconds to load. It is ten years old, so read it as the classic finding it is, not a current benchmark. Point the code at one page whose headline repeats the flyer headline, with one button. We could not find a test of a dedicated page against a homepage for print traffic, so that is a recommendation consistent with the evidence rather than a measured result.

This is where a dynamic QR code earns its place on a flyer. A dynamic QR code is a code that points to a short redirect link instead of the final page, so you can change where it lands after the flyers are printed. Swap the offer page for an “offer ended” page after the deadline, fix a broken page without reprinting, and count the scans from each batch separately.

What should you check before printing a flyer?

Print one copy at 100% and scan it with two different phones at arm’s length in dim light. Then check that the typed address works, that the page loads on mobile data, and that each distribution batch has its own code. Ten minutes here is cheaper than a reprint.

1. Export the code as SVG, or as a PNG at the exact size you will print. Denso Wave warns that when a QR code is resized in an image tool “every module becomes distorted”.
2. Print one copy at 100% on the paper and finish you will use, including any gloss.
3. Scan it with an iPhone and an Android phone from arm’s length, in the dimmest light it will meet.
4. Type the short web address on a phone and confirm it reaches the same page.
5. Open the landing page on mobile data, not office wi-fi, and check the headline matches the flyer.
6. Give each batch (door-to-door, counter, noticeboard) its own code, so the scans separate by placement.
7. Write down the end date, and what the code should point to once it has passed.

Our help centre has a fuller checklist for testing a QR code before printing, including what to look for on the printer’s proof.

Which flyer myths can you stop repeating?

Most of the precise-sounding statistics on flyer design pages. The ones below circulate widely and either have no traceable source, or the source says something different. None of them should be on your flyer or in your pitch for one.

• “Visuals are processed 60,000 times faster than text.” Investigations by Alan Levine and PolicyViz traced it to a 1997 3M brochure with no references.
• “Colour increases brand recognition by 80%.” The Loyola study usually cited “doesn’t mention that 80% increase at all”, according to an investigation by Insights4Print; the earliest source its investigator found was probably a Xerox marketing leaflet.
• “Attention spans are 8 seconds, shorter than a goldfish.” The BBC found the figure did not come from the Microsoft research it is credited to. Its cousin, “a flyer has 3 seconds to grab attention”, is one we could not trace to any source.
• “People remember 10% of what they read and 80% of what they see.” According to Work-Learning Research, there is “no body of research” behind the percentages, and Dale’s original cone contained no numbers.
• “A labelled QR code gets 40–70% more scans” and “flyers with QR codes get 30% more engagement.” No source found. The measured effect of adding a code to print is about 1.3 to 3 percentage points, or nothing.
• “Hick’s law proves one call to action converts better” and “single-CTA emails get 371% more clicks.” Hick measured reaction time between known stimuli and responses, not persuasion, and we could not trace the 371% figure to a primary source.
• “Direct mail gets a 4.4% response rate.” It traces to a 2012 trade article, and we could not verify any current US average.

Are there flyer templates with a QR code already in place?

Yes: we made eight flyer templates, each A4 and each built around one offer, one QR code and one scan label. You can preview all eight on our flyer templates page. Using one is part of the Pro plan and above: you open it in your QRCodeStack dashboard, pick one of your QR codes, change the words and print it.

You can also open your own copy in Canva, which adds a second page with a how-to and a “Before you print” checklist. The eight below are the flyers. The same library also has signs, cards and table tents, 53 templates in all, and you can filter it by industry.

The sample QR code on each template is a plain dark-on-white placeholder. In the dashboard you pick one of your own codes and it replaces the sample. If you work in Canva instead, the QRCodeStack app inside Canva inserts a dynamic code without leaving the design. The templates need the Pro plan, which is $19 a month. The 7-day free trial, which needs no credit card, lets you preview every template and make dynamic codes before you pay. One limit to know before a flyer goes on a noticeboard for months: if the subscription lapses, the code stops working rather than carrying on untracked.