KNX: The Brand Matters, But Rarely Where You’d Expect

Who it’s for: Integrators, installers and design offices.

"Which brand should I choose?"

It is almost always one of the first questions asked in a KNX project.

It is also often the wrong question at the wrong time. Not because the manufacturer doesn't matter (we'll see that it does), but because we're asking it to settle trade-offs that haven't even been defined yet.

The Real Cost Lies in the Structure, Not the Catalogue

A concrete example.

A commercial building is delivered with a working KNX installation: lighting, blinds, heating, supervision, everything responds as expected.

Three years later, the client wants to split an open-plan floor into four independent offices.

The problem isn't the hardware. It's the addressing.

The group addresses were built by floor and by equipment type, with no intermediate level corresponding to functional zones. The floor's lighting commands point to objects that are too tightly coupled.

Re-splitting the space then means reworking the addressing, reconfiguring the switches, reloading the actuators, adapting the supervision and retesting everything.

What should have been a minor change ends up requiring several days of engineering and commissioning work.

The hardware was good.

The architecture wasn't.

And no choice of manufacturer would have changed that outcome.

This is where a large part of the gap between two installations plays out: network topology and segmentation, the structure of group addresses, the interfaces with HVAC, DALI, Modbus or BACnet, and the defined behaviour in the event of a fault or loss of communication.

This work is invisible once the project is delivered.

Yet it determines a significant share of the cost of future changes.

Where Manufacturers Really Differ

Saying "it all depends on the project" is true, but convenient.

KNX manufacturers are not interchangeable. So it’s worth knowing on which criteria they actually differ.

ETS Application Quality

This is one of the most underestimated criteria, and one of the most costly if overlooked.

Two KNX-certified shutter actuators may offer the same function on paper.

One exposes clear logic, well-structured objects and useful parameter granularity. The other forces workarounds to achieve behaviour that should be simple.

KNX certification guarantees interoperability at the protocol level. It guarantees neither the quality of the ETS application nor the ease of integrating and operating the product.

The difference doesn't show in the catalogue.

It is paid for in integration hours, and again with every change.

Application Longevity

A building is operated for twenty years. A product range can evolve far more quickly.

Before validating a product range, three questions are worth asking:

• Does the manufacturer maintain its applications across successive ETS versions?
• Can a new generation of a product take over the function of an older device without a full reconfiguration?
• Can a faulty module be replaced several years later without calling a significant part of the project into question?

Longevity, then, is not measured by a product's availability today, but by the manufacturer's ability to maintain, evolve and replace its ranges over time.

Support and Local Availability

An actuator replaced within 72 hours and one replaced after six weeks do not have the same impact on operations.

In an occupied commercial building, this difference can matter more than a few percentage points on the purchase price.

Technical support, local stock, documentation and the availability of a competent contact are therefore an integral part of choosing a manufacturer.

Functional Density

The right metric isn't always a product's price, but the cost per point or per function.

A high channel-density actuator may have a higher unit price while reducing cabinet space, cabling, the number of references to maintain and commissioning time.

The cheapest product to buy is therefore not necessarily the least costly solution over its life cycle.

Fault Behaviour

Not all devices allow the same level of fine-grained control over their behaviour in the event of bus loss, power restoration or sensor failure.

On secondary lighting, the impact can remain limited.

On wind-exposed blinds, HVAC regulation or a highly integrated installation, this point can become decisive.

The question, then, is not only:

"Does the product work?"

But also:

"What does it do when something stops working?"

Real Interoperability

Mixing manufacturers is not a problem.

It's actually one of the fundamental benefits of KNX.

The problem arises when choices are not documented, when responsibilities become unclear during maintenance, or when proprietary logics overlap without overall consistency.

A multi-brand environment, chosen deliberately and documented, can be perfectly robust.

A multi-brand environment that just happened can become difficult to maintain.

These criteria carry even more weight the larger the project, the longer its operating life, and the richer its interfaces.

On a villa, the choice of manufacturer can remain secondary.

On a commercial building with several thousand points, supervision, HVAC and multiple gateways, it becomes a genuine engineering decision.

Commissioning Is Part of the Engineering

A well-designed installation can be undermined by rushed commissioning.

Testing is not about checking that the lights turn on, but that the functions genuinely meet the defined needs: scenarios, lighting, blinds, heating and cooling, regulation, detection, supervision, alarms and interactions between systems.

And above all, degraded scenarios must be tested:

• loss of communication;
• power outage and restoration;
• sensor fault;
• loss of a gateway;
• equipment behaviour in degraded mode.

This is where default settings that were never actually reviewed come to light.

A Smart Building system is not finished when it is installed. It is finished when it is tested, documented and validated.

Total Cost, Not Purchase Price

The price of the hardware represents only a fraction of a system's real cost.

Integration, commissioning, operation, maintenance, upgrades and replacement must be added to it.

A product that's 20% cheaper but costs two extra days of engineering at commissioning, then half a day of rework with every change, hasn't saved anyone anything.

This is why interoperability, scalability, maintainability and documentation are not just brochure words.

They are budget line items.

So, Which Brand Should You Choose?

The one whose products meet the project's real constraints: ETS application quality, application longevity, support and availability, functional density, fault behaviour, interoperability and total cost.

Choosing a brand is a serious technical decision.

It’s simply a conclusion. Not a starting point.

Expertise is not about defending a manufacturer.

It's about knowing why it was chosen, in what context it makes sense, what its limitations are, and how it fits into an architecture that will still need to work and evolve in ten years' time.

This is the approach we apply at ExaBuilding: start from the building, its uses and its constraints, design the architecture, then select the technologies capable of serving it durably.

The products come afterwards.

And the reasoning that led to their selection must be explainable, documented and defensible.