If you're about to buy a smart lighting system, start with the protocol, not the switch. After reviewing 200+ lighting-control configurations a year as a quality and compliance manager, I'm convinced of one thing: for most projects in 2025, a Zigbee-based setup with a Matter-capable controller such as the Legrand Netatmo Matter light switch is a better investment than another Wi-Fi switch—and the money you save on proprietary bridges should go into a Legrand wall occupancy sensor placed where it actually sees people.
I know the word Matter has reached buzzword status. I'll use it anyway. I've watched what happens when a manufacturer closes its app API or retires a cloud service. Your switches don't fail. They just become furniture. Zigbee plus a Matter path is not forever, but it's the best hedge lighting buyers have right now.
Here's my background: I'm a quality and brand compliance manager at a lighting company. In our Q1 2024 review cycle, I checked roughly 200 product configurations and rejected about 4% of first production samples for things like finish mismatch or sensor alignment problems. For me, quality is consistency—and most returned 'smart' products are not hardware failures. They are installation and expectation failures.
Why Zigbee? Because one network beats a stack of bridges
Let's talk honestly about why I recommend Zigbee over Wi-Fi switches. The first reason is interoperability. If a building already has a Danfoss Zigbee thermostat, that device can live on the same Zigbee network as your Legrand lighting controls. Different brands, same language. That means fewer gateways, less convoluted wiring, and a single place to check when something isn't responding.
I recently reviewed a retrofit plan where the facility manager assumed we needed a second smart-home bridge for lighting because the thermostats were from a different brand. We didn't. Both endpoints were Zigbee. The installed solution ended up simpler on paper and more reliable in practice.
The second reason is load behavior. A Wi-Fi router can handle plenty of devices, but its main job is not low-power lighting commands. Zigbee is designed for dense, low-power meshes. A good Zigbee network with mains-powered devices as repeaters tends to stay stable as you add rooms. Wi-Fi switches work, but when you're doing a whole floor, small network delays become user complaints.
When I evaluate the Legrand Netatmo Matter light switch, I don't start with the app or the Matter logo. I start with the mechanical switch. Will it feel solid after years of use? Does the paddle move without sticking? Does the LED bleed through the faceplate and annoy a client at 2am? The Matter capability and Zigbee radio matter for the long-term, but tradespeople still judge it as a physical switch. That's the quality standard I test against.
What about Zigbee vulnerabilities?
Let's address the phrase 'Zigbee vulnerabilities' directly. It is not fiction. Researchers have found real attack paths, mostly in older implementations or poorly configured networks. When I review security for a lighting project, I group those risks into three buckets:
- Network key extraction when a network is left in join mode or uses default install codes.
- Replay attacks on devices that don't handle frame counters properly.
- Physical-layer jamming in the 2.4GHz band, which no software feature can fully prevent.
Those risks do not make Zigbee unusable. They make commissioning discipline mandatory. Zigbee 3.0 uses AES-128 encryption and a trust-center model for device joins. Practically, that means updates and install codes matter more than a vague fear of the protocol. I've seen projects with Zigbee vulnerabilities that were introduced by the installer leaving the network open for a week after commissioning. That is a process problem, not a radio problem.
To be fair, no wireless protocol is immune. Wi-Fi devices have their own update problems, and proprietary systems create a worse long-term risk: the manufacturer can simply disappear. The real question is whether you are willing to use out-of-band commissioning, unique install codes, and firmware updates before handing the system to the client.
Legrand wall occupancy sensor: location is a spec, not a suggestion
The switch gets the attention, but the wall occupancy sensor often decides whether the project feels smart or annoying. A sensor that leaves people in the dark is worse than a dumb switch.
The counterintuitive thing I've learned from testing is that more sensor coverage is not always better. A PIR wall occupancy sensor with a huge field of view can see hallway traffic through an open door and keep the lights on in a room where nobody is sitting. Meanwhile, a sensor with a tighter, correctly oriented pattern is easier to live with. The Legrand wall occupancy sensor is typically chosen by square footage, but the coverage chart and mounting height matter more.
In one project, lights in an empty room kept turning on at night. The sensor wasn't defective. It was installed above a supply air diffuser, so moving air was making the PIR think something was in the room. Someone had chosen a dual-technology model because it was 'better,' but the ultrasonic side was being triggered by air currents. The fix wasn't a firmware update. It was moving the sensor away from the duct.
So here's my blunt advice: use PIR-only for simple spaces without much stationary work. Use dual technology for offices with partitions where small movement needs to be caught. But never buy a sensor without understanding where it will be mounted and what else is in the room.
Moving a light switch? The real work is before the drywall
If you search for 'how to move light switch,' you'll see photos of two wires and a screwdriver. In reality, moving a light switch is mostly about code compliance, wire routing, and box volume. It's not about moving the plastic paddle.
A classic switch loop often has no neutral wire in the box. Many smart switches—including the Legrand Netatmo Matter light switch—need a neutral to power the radio when the light is off. If there's no neutral, you have two choices: run new cable or pick a device designed for no-neutral installations. No amount of Zigbee magic fixes an absent neutral.
The quality manager version of the process looks like this:
- Confirm what's actually in the existing switch box: line, load, neutral, ground, and how many cables enter it.
- Choose a box that has enough volume for the smart switch body and the additional wires. Smart switches are deeper than standard toggles.
- Route the cable to the new switch location without creating hidden junctions. If the old switch box becomes a junction box, it needs an accessible cover and proper splices.
- Label every conductor before touching anything. If the cable is feeding a switch loop and not a neutral, that matters even more.
Before you cut drywall, ask yourself whether you actually need to move the line-voltage switch or just add another control point. A wireless companion switch can often give you control from a new location without running a new cable. But a wireless switch is not the same as relocating the physical switching point. If the project needs a hardwired switch in a new place, hire an electrician and follow local code. I can review a spec sheet or a returned product, but nobody should use an article as a substitute for a licensed installer.
When I'd tell you to skip this setup
Not every lighting project needs Zigbee, and not every project needs Legrand. If a building already has a working wired control system, I would not rip it out for the sake of a fresh wireless installation.
I also have a hard boundary around life-safety and critical lighting. If the purpose is emergency lighting, exit paths, or any area where wireless failure creates real danger, stick with wired controls. Zigbee is a great layer for convenience and energy management, but I don't pretend it replaces a dedicated hardwired system.
If your security team says no 2.4GHz wireless devices in the building, listen to them. There are cases where jamming, policy, or just an extremely noisy RF environment makes Zigbee the wrong answer. In those situations, I still recommend a direct-wired sensor and control solution. Wired lighting is not obsolete. It has a different job.
And if this is a one-room project with no plans to expand, honestly, you may not need a smart system at all. A good traditional switch will last for decades. Smart controls earn their keep when you want scheduling, vacancy detection, remote awareness, or integration across multiple spaces. If those needs don't exist, skip the complexity.
So my final advice is simple: choose Zigbee 3.0 devices where possible, look for Matter-capable products like the Legrand Netatmo Matter light switch, put the wall occupancy sensor where it can actually see a person, and spend real time on the electrical details if you're moving a switch. That's the boring, reliable path—and for lighting, boring is a compliment.