HospitalityRoom keys, staff credentials and the pieces around them.
Retail and loyaltyGift, loyalty and membership cards with variable data.
Access controlCards and fobs matched to the reader you installed.
Clubs and associationsOne chip for the turnstile, the locker and the desk.
Events and festivalsWristbands, badges and lanyards for each edition.Why the lock brand never specifies the card, the four variables that decide whether a credential authenticates, and what to establish before ordering.
Every hotel credential order starts with the same question, and it is almost always asked the wrong way round. The question that gets asked is "do these cards work with Onity?". The question that decides whether the order succeeds is "what exactly is the reader expecting to see when the card is presented?".
Those are not the same question. A lock brand is not a specification. Two properties running the same brand, commissioned three years apart, can need different cards.
This guide is written for the person who has to specify the credential: the integrator, the purchasing team, or the facilities lead at a group. It covers what actually determines compatibility, what to establish before a purchase order goes out, and what a manufacturer needs in order to get the first sample right instead of the third.
A hotel lock system has several layers, and the card only has to satisfy some of them. The brand name on the escutcheon tells you almost nothing about the layer that matters.
It narrows the likely radio technology and the likely chip family. If someone says Salto, you can be confident you are in 13.56 MHz territory. If someone says Onity and the card has a black stripe on the back, you are looking at magnetic stripe, not RFID at all.
It does not tell you the chip, the memory layout, the security mode, or the keys. All four are set when the property's system is configured and encoded, not when the lock is manufactured. The same brand can be deployed with a low-security legacy chip in one hotel and an AES-encrypted credential in another.
The card has to match the configuration, not the logo
This is why "compatible with Onity" is a marketing claim rather than a technical one. A card is compatible with a specific system, in a specific encoding mode, holding specific keys. Anything less precise is a guess that someone pays for later.
Whatever the brand, compatibility comes down to four things. Get all four right and the card opens the door. Get any one wrong and it does not, regardless of how good the card is.
Low frequency at 125 kHz and high frequency at 13.56 MHz are different radio technologies. They are not interchangeable and no amount of encoding fixes a frequency mismatch. A 125 kHz card presented to a 13.56 MHz reader is invisible to it. Properties that run both, usually because they upgraded part of the estate, need a dual frequency card with two independent transponders.
Within 13.56 MHz there are several families that are not compatible with each other at the application level even though they share the radio layer. MIFARE Classic, MIFARE Ultralight, MIFARE Plus and MIFARE DESFire all answer an ISO/IEC 14443 Type A reader, but they organise memory differently and authenticate differently. A reader configured to read a DESFire application will not accept a Classic card. What each family actually does is set out in the guide to choosing the chip.
Two cards with the same chip can still behave differently if the data is written in a different place or a different format. Sector layouts, file identifiers and block structures are set by the lock system's software. This is the layer where most surprises happen.
On any system with real security, the reader will only talk to a card that holds the correct keys. Keys belong to the property, not to the card manufacturer. In practice this means a factory delivers cards that are either blank and encodable on site, or pre encoded from a key file the property supplies through the proper channel.
The following is what we see in practice across the systems we supply for. Encoding depends on the chip and on the system's technical requirements, keys and authorisation. Treat it as the starting point of a conversation with the property's integrator, not as a substitute for it.
Almost always 13.56 MHz on current installations. Older Signature estates are frequently magnetic stripe. VingCard deployments have been moving to AES encrypted credentials, and that migration is the single most common reason a hotel's existing card stock stops working after a software upgrade. If the property mentions a recent lock software upgrade, assume the encryption question is live and confirm it before ordering. What changes and what to test is covered in the AES key card guide.
Both magnetic stripe and RFID estates are still in service, often in the same group. RFID deployments sit at 13.56 MHz. The practical risk here is a property that describes itself as "Saflok" without knowing whether its floor is on the legacy or the RFID generation. A photograph of the reader face usually settles it in one message.
A large installed base of magnetic stripe alongside newer 13.56 MHz systems. Onity is the system where the magnetic stripe question comes up most often, and where getting the track layout and coercivity right matters as much as the chip does.
13.56 MHz throughout. Salto estates are frequently mixed use, with the same credential opening guest rooms, staff doors and back of house. That matters for the card body as much as the chip: a credential carried on a lanyard all day is not the same product as one that lives in a wallet for two nights.
If nobody at the property knows the answer
That is common and it is not a problem. The property's lock integrator knows, and the encoder software itself reports the credential type. A card that is already working can also be read and identified. Any of these three routes is faster than guessing.
A significant share of hotel doors worldwide still open with a magnetic stripe, and those properties are often the ones least well served, because most manufacturers have moved on. If that is your system, three things matter.
High coercivity at 2750 Oe resists accidental erasure far better than low coercivity at 300 Oe. Hotel environments are full of phone magnets and bag clasps. For guest cards, high coercivity is almost always the right answer.
Which of the three tracks the system writes to, and in what format, is set by the lock software. Ordering a stripe on the wrong track produces a card that encodes cleanly and opens nothing.
Distance from the card edge is standardised, but artwork that runs to the edge can interfere with the encoder's read if the stripe is applied over printing rather than under the overlay. This is a manufacturing decision, not a design one, and it should be raised before artwork is approved. The stripe itself is a standard option on PVC and recycled PVC cards, by itself or alongside a chip.
Every question below exists because the answer prevents a reprint. When we quote, this is what we ask for.
Brand, product line and software version if you have it. "Onity" narrows it a little. "Onity Advance, RFID readers installed 2019" narrows it enough to be useful.
A sample of a card that currently opens the doors is the single most useful thing you can send. It can be read, identified and matched. If you cannot send one, the reader or lock model and the chip type, if known, are the next best thing.
Most properties encode on site with their system, which means the factory ships blank cards of the correct type. Some want cards pre encoded, which is one of the personalisation and encoding steps we run in house. If it is the second, keys have to be transmitted through a channel the property is comfortable with, and that is a conversation to have before quoting, not after.
Quantity drives the price break. Material drives the feel and the sustainability story, and the trade offs between PVC, recycled PVC and wood are worth settling before artwork starts. Finish drives the lead time. Wooden cards, for example, run twelve calendar days at the plant against twelve for a standard PVC run, but they carry different artwork constraints because the grain shows through.
A rough is enough to quote. Print ready files can follow. What we need to know at quoting stage is whether the design runs to the edge, whether there is foil or embossing, and whether anything has to be variable per card. The full list of what a factory needs, and why, is in the guide to specifying a card order.
No serious order should skip this, and the cost of doing it is a few days.
Colour on screen is not colour on PVC, and it is certainly not colour on walnut. Ask for a free sample to test it first, and on long runs the first pieces are made and sent to you to check before the rest goes on the line.
Not one door. Test a guest room, a staff door, a lift reader and any back of house reader that the credential has to work with. Firmware can differ between reader generations inside the same property, and a card that works on floor three and fails on floor eleven is telling you something important.
A blank card that reads correctly proves the chip is right. It does not prove the encoding is right. The meaningful test is a card encoded by the property's system and presented to the property's lock.
One card, three different tests
Reading at the encoder proves the chip. Opening one door proves the configuration. Opening doors across every zone and every reader generation proves the order. Only the third one is worth committing a full run to.
Compatibility is not a feature that gets bought. It is a match that gets established, and establishing it takes one exchange of messages when the right information is on the table, or three rounds of samples when it is not.
The cost of the first route is an email. The cost of the second is a reprint, a delayed opening and a conversation with a client about why the cards do not work. That asymmetry is the whole argument for doing the identification properly before the order is placed.
Kaway manufactures credentials in-house for hotel groups, integrators and programme owners. Material data sheets and certifications go into the delivery dossier with every order.
Encoding at the plant covers hotel lock systems including ASSA ABLOY VingCard, SALTO, dormakaba, Onity and Saflok, across hospitality and general access control work. Encoding depends on the chip and on the system's technical requirements, keys and authorisation. Minimum order is 500 units. Standard production is 12 calendar days for PVC and wooden cards. Quotes come back the same working day, including the ones where the answer is no. The rest of the commercial detail sits on the FAQ.
Sometimes. If both systems read the same chip and the same data, yes. If one is 125 kHz and the other 13.56 MHz, you need a dual frequency card with two transponders, which is a standard product. If both are 13.56 MHz but use different chip families, usually no.
Almost always the system, not the cards. A software or firmware upgrade can change the credential the readers expect, most commonly by introducing or tightening encryption. The cards did not degrade, the requirement moved.
We can, but most estates encode on site with their system and their keys, and that is usually the better arrangement. Keys belong to the property, and the fewer places they exist the better. Where an integrator handles encoding as part of the installation, we ship blank stock of the correct type.
Yes. The chip and antenna are the same components; the body material is different. Read performance is equivalent for practical purposes. What changes is thickness tolerance and artwork, not compatibility. What a wooden card does and does not change is covered separately.
Send an existing card that opens the doors, together with a photograph of the reader or its model number. Between the two, the system is usually identifiable. If it is not, the property's integrator can confirm it in a single email.
About this guide
Published by Kaway Group. The date on this page shows when this guide was last updated. If anything here is out of date, or contradicts what your integrator tells you, tell us and we will correct it.
Ask for a free sample in your material and chip.