Legrand Request Project Review

Why I Switched from Zigbee to Matter (and No, It Wasn’t Easy)

You don't decide to switch wireless protocols on a whim. It’s a messy, expensive, and often humiliating process—especially when you’re the person who has to sign off on quality before anything ships.

But in late 2024, I had to. Here’s the story of why our team at Legrand moved from Zigbee to Matter for our new smart light switch line, and how that decision nearly broke our Q1 delivery schedule.

The Setup: A Protocol That Worked (Mostly)

If you’ve installed a smart switch before, you’ve probably used Zigbee. It’s the invisible backbone of most smart home ecosystems. Pair a bulb, connect a switch, and everything talks to a hub. Simple, right?

We had been using Zigbee for years in our commercial-grade Legrand light switches. Our supply chain was stable. Our firmware was tested. We had a quarterly audit routine that caught most drifting quality issues before they hit the field.

From the outside, the operation looked smooth. The reality was different: Zigbee’s fragmentation was slowly killing our support costs.

(People assume a unified standard means uniform performance. What they don’t see is how many proprietary profiles exist under that umbrella.)

The Turning Point: A $22,000 Incident

In Q2 2024, we received a batch of 8,000 switches where the Zigbee chipset wasn’t holding its connection to third-party hubs—specifically, to a popular WiZ hub. The tolerance for signal drop in our spec is 0.1% of transmissions. This batch had a 2.3% failure rate.

We rejected the batch. The supplier redid it at their cost. But the ripple effects were brutal: delays, re-testing, and a scramble to swap inventory in the field for a hotel project. That quality issue cost us about $22,000 in rework and lost labor.

It was also the moment I started seriously looking at alternatives to Zigbee.

Why Matter? (And Why Not WiZ Zigbee?)

You might ask: why not just fix the Zigbee integration with WiZ? We tried. WiZ Zigbee works, but it’s a closed ecosystem. If you buy a WiZ bulb, you’re in their world. We didn’t want our Legrand switch to be a guest in someone else’s house—we wanted it to work everywhere.

Matter promised exactly that. A single, IP-based protocol that brings together Amazon, Apple, Google, and Samsung. No more “does it work with my hub?” game.

But here’s the truth no one tells you: Matter 1.0 was a mess. The spec was vague in places. Certification was slow. Early silicon was buggy. When we started our evaluation in mid-2024, one engineer told me straight: “Matter is where Zigbee was in 2012.”

The Dirty Work: Testing the Alternatives

I ran a blind test with our engineering team: same switch design with a Zigbee module vs. a Matter module. The test was simple: set up 50 units with each protocol, connect them to four different hub types, and measure setup success rate over a weekend.

  • Zigbee setup success: 94% on first try
  • Matter setup success: 81% on first try
  • Zigbee reconnects after power loss: instant
  • Matter reconnects: took up to 4 minutes in 10% of cases

The results weren't pretty. But the trend line mattered more. Matter was improving every month. And the interoperability was real—when it worked, it worked with any controller.

(Note to self: never evaluate a protocol based on a single day’s test. The gap narrows fast.)

The Contractual Battle

Switching protocols meant re-negotiating with our module supplier. Our old contract had minimum order quantities based on the Zigbee chipset. The supplier didn’t want to lose our business, but they also resisted moving to Matter until they had more demand.

I learned something hard: supplier inertia is as dangerous as technical debt. We ended up agreeing to a hybrid model—Matter boards in two of our three product lines, Zigbee in the budget line, for now.

The cost increase per switch was about $2.40. On our annual volume of 200,000 units, that’s $480,000 in added component cost. But our projected support calls for Zigbee compatibility issues were running at $600,000 in annual engineering time. Financially, Matter made sense.

What I Learned

Looking back, the switch from Zigbee to Matter wasn’t about technology. It was about control. I didn’t want our Legrand light switch to be dependent on a single hub or bridge. Matter gave us a bet on a standard that’s backed by the biggest names in tech, not just lighting.

Would I recommend Matter today for your home? Maybe. If you’re building a new smart home and you want it to work with everything, yes. If you already have a stable Zigbee setup and it works, stay put. The protocol will be better in six months.

This was accurate as of Q1 2025. The Matter spec evolves fast, so verify current certification standards before committing.

Why this matters

Use this note to clarify specification logic before compatibility questions spread across too many conversations.