Dynamic or static? The invisible difference between QR codes

They are stuck to billboards, they stand on restaurant tables as table talkers and they appear on almost every package: QR codes have long been part of everyday life. From the outside the black-and-white squares all look the same — technically, though, a fundamental decision separates them, one that governs usefulness, lifespan and follow-on costs: the choice between a static and a dynamic code. Anyone who knows the difference avoids the most expensive mistake you can make with QR codes — the error printed permanently into the artwork.

A brief look at the fundamentals helps. The QR code — the abbreviation stands for “quick response” — was developed in 1994 at the Japanese automotive supplier Denso Wave, in order to record components in logistics more quickly. Unlike the classic barcode it stores information two-dimensionally: the arrangement of the small squares, called modules, encodes the actual data. Three prominent corner squares help the camera to recognise the code in any position and from any angle. The more data is meant to go in, the finer the grid becomes — and the more demanding the scan.

Built in on top of that is error correction using the Reed-Solomon method. It ensures that a code stays readable even when parts are missing or covered — depending on the correction level chosen, up to around 30 per cent of the area may be damaged. That is why scratched codes on the parking ticket machine still work, and why a logo can be placed in the middle of many codes without destroying legibility. This reserve is finite, however: anyone who combines a logo, unusual colours and delicate patterns uses it up on all three counts at once. Before printing, therefore, every styled code belongs on several telephones, older devices included.

Which brings us to the actual fork in the road. A static QR code translates the information directly into the pattern. A web address, a short text, a telephone number, an email address, the credentials of a Wi-Fi network or a digital business card in vCard format — all of it is written straight into the code. When scanning, the telephone reads the data out of the image; no server is needed for that.

From this follow the strengths of the static code: it is completely independent. No service has to be running, no account has to exist, no provider has to be in business for it to work. A static Wi-Fi code on the noticeboard of a holiday flat connects guests to the network even when the mobile network is on strike — the data is in the image itself, after all. Once generated, a static code works in principle for ever, and it costs nothing on an ongoing basis.

One technical side effect of this construction is often overlooked in everyday use: with a static code the quantity of data determines the density of the pattern. A short address yields a coarse, good-natured grid; a long address with campaign parameters appended, or a complete business card with postal address, company name and several telephone numbers, drives up the number of modules. The finer the grid, the larger the print has to be and the better the light and the camera have to be for the scan to succeed reliably. Anyone working statically therefore fares better with frugal content — in case of doubt it is worth asking whether a short address on your own domain can reduce the amount of data in the code. A dynamic code sidesteps the problem by construction: it always contains only the short redirect address, however long the destination address behind it may be.

Not every kind of content, incidentally, suits both variants. Wi-Fi credentials, a free text or a telephone number are static by their very nature — the whole point is that they sit in the code without a detour via the internet. The decision between static and dynamic arises above all with web addresses, by far the most common case.

The weakness of the static code is the flip side of its independence: what is printed is printed. If the code points to a web address and the page moves, if the domain changes or a typo creeps in, every single copy becomes worthless. Five thousand flyers with a dead link cannot be corrected after the fact — they can only be reprinted. For long-lived print material that is a real risk, because experience shows that web addresses are less durable than one assumes when placing the print order.

This is exactly where the dynamic QR code comes in. It does not encode the actual destination but a short redirect address on a provider's servers. On scanning, the telephone first calls up that address; the server then forwards to the stored destination. The decisive point: the destination can be changed at any time without the printed pattern changing. The poster campaign can point to a new promotion page while it is running, the menu to the winter card, the programme booklet to the updated schedule — the print stays, the destination moves.

In everyday work this switch is unspectacular: the new destination is entered in the provider's account, and from the next scan onwards the code leads there — the printed material remains untouched. That opens up room for manoeuvre beyond mere error correction. A campaign can be run in stages: before the launch the already printed code points to an announcement page, on the appointed day it is switched to the promotion page, and after the end to an archive or the home page — instead of leftover flyers leading nowhere. The reverse case is covered too: if a typo in the stored destination address comes to light after printing, it is corrected in a few minutes. With a static code the same error would be final.

What this looks like in practice is shown by the German web service QR2GO: codes are created there directly in the browser, either as a static code or dynamically with an editable destination. No installation is needed, and the data is hosted on servers in Germany.

The flexibility has a price, and it should be named honestly: a dynamic code depends on its redirect service. If the provider ceases operations, if the subscription ends or the account is deleted, the printed code leads nowhere — even though the destination offering itself still exists. Anyone putting dynamic codes into long-lived print material should therefore clarify in advance what happens to existing codes when a tariff runs out, and whether the free range of functions applies permanently.

The second big advantage of the redirect: it creates a measuring point. Because every scan first arrives at the provider's server, it can be counted how often, when and roughly where a code was scanned. QR2GO, for instance, evaluates scans in real time by device type, region and time of day and, on request, adds UTM parameters for web analytics, so that visits can be attributed to the respective campaign in tools such as Google Analytics 4. For advertisers that answers an old question: which poster, which flyer, which advertisement actually triggered a response?

For such figures to bear weight, the set-up needs a little discipline. It has proved sound to create a separate code for every advertising surface — one for the poster, one for the flyer, one for the advertisement in the free local paper — and to assign the UTM parameters according to a fixed scheme, for instance with the surface as the source and the promotion as the campaign name. Only this scheme makes the evaluation comparable: anyone improvising the parameters case by case ends up with figures that can barely be attributed cleanly afterwards. The convention is therefore best settled before the first print order, not after the third.

A qualification belongs here as well. Scan statistics are reach data, not personal profiles — reputable providers collect them in aggregated form and in line with data protection law. Anyone linking the figures with their own website analytics such as GA4, however, is moving within the scope of the GDPR: the destination page then needs a clean consent solution and a privacy policy that mentions the tracking. The QR code does not replace a data protection concept; it is part of one.

There are differences in the kind of redirect as well. QR2GO offers three modes: a visible redirect, in which users can see the intermediate step, a covert variant that counts and forwards immediately, and a direct mode without the detour via counting. The choice is a trade-off between transparency and a seamless user experience — for codes displayed in public there is much to be said for the visible route, because it builds trust.

An often overlooked practical point is the file format. PNG files are raster graphics: for screens and small prints they are sufficient, but when enlarged substantially the edges become blurred — and blurred edges worsen legibility. For print shops, posters and signage, vector formats such as SVG or PDF are the better choice, because they scale without loss. At QR2GO the PNG export is included in the free tariff; high-resolution exports are part of the premium range.

A second, inconspicuous requirement belongs to printing: the quiet zone. What is meant is the white margin all around the code, which the camera needs in order to separate the pattern from its surroundings. In layout programs it is readily cropped away, because it looks like wasted space — with the result that a technically faultless code scans unreliably on the finished poster. Anyone wanting to be sure orders a proof at final size and tests it under real conditions: from a distance, in poor light, with several telephones. Which design and export options are available is listed in the provider's feature overview — none of them relieves you of the duty to test.

So when is which type the right one? Static is the right choice when the destination is permanently settled and no measurement is needed: the Wi-Fi credentials in the guest room, the business card with contact details that have been stable for years, a short text that is meant to be readable without an internet connection. Anyone who on principle does not want to enter into a dependency on a service provider also fares better statically — and deliberately accepts the reprint risk in exchange.

Dynamic pays off as soon as print material outlives its content, or as soon as success is meant to be measurable: campaigns with changing promotion pages, menus, catalogues, packaging, signage. The rule of thumb: the more expensive the print and the more likely a later change of destination, the more strongly the arguments speak for the dynamic code.

That leaves the cost question. Static codes can be generated free of charge with many tools, permanently so. Dynamic codes presuppose a running service with server operation behind it — which is why subscription models are usual. QR2GO offers a permanently free entry level with 20 code slots and basic styling; the premium tariff at 5.99 euros a month comprises 200 slots, full scan analysis, high-resolution exports and logo embedding. The details are listed in the pricing overview; a comparison with other providers is worth making in any case, because the models on the market differ considerably.

In the end the decision between static and dynamic is not a technical nicety but a setting of the points that should be made before the first print order. A static code costs nothing and belongs to nobody — in exchange it stays as it is. A dynamic code stays changeable and measurable — in exchange it hangs on a service. Anyone who asks themselves this one question in good time is spared the fate of countless printed items whose little squares lead nowhere.