I've Installed Control4, Crestron, and Savant for 22 Years. Here's Why I Wouldn't Put One in My Own House.

1. The one place these systems belong

I want to be fair to the engineering. Crestron, Control4, Savant, and AMX aren't junk. They're sophisticated platforms with capabilities that genuinely don't exist anywhere else in the consumer market. But the question isn't "is this good engineering?" It's "is this good engineering for a house?"

The honest answer is: these systems were designed for mission-critical commercial environments. Crestron in particular came up through corporate boardrooms, broadcast studios, university classrooms, courtrooms, and hospital integration projects. AMX has a similar pedigree. When the cost of downtime is "the meeting can't happen" or "the surgery has to be delayed," 99.99% uptime is worth what it costs to maintain. The IT team that supports a Crestron-equipped boardroom is full-time staff. The annual maintenance contract is line-itemed in the facility budget.

In that world, Crestron wins. It's overkill for almost everything else, and I'd recommend it without hesitation for a 50-room conference center or a teaching hospital's surgical wing. Even there, the industry standard is to plan for full hardware refresh every 5 years and pay for regular maintenance. That's how the platform is designed to age.

Now look at what that means for a residential customer.

2. Why residential is different

Your house doesn't need 99.99% uptime. It needs to work in 10 years without a service contract and without a four-figure call to add a button.

That's not the bar these systems were built to clear. They were built for environments with dedicated IT staff, recurring maintenance budgets, and a willingness to do scheduled hardware refresh on a corporate cycle. None of that maps to how people live in their homes.

I've watched the same story play out in 100+ residential installs:

• Year 1 — the system is incredible. Everyone is happy. The dealer is responsive. The scenes work. The touchpanels are clean and fast.
• Year 2 — a few small things need adjusting. Most dealers absorb the labor or charge modestly. The relationship is still good.
• Year 3 — the homeowner has change requests stacked up (new TV needs to be added to the URC remote, new Sonos zone, kitchen scene needs to fade slower, schedule changes). The dealer's labor bill starts to feel high. The original programmer may have already left the dealership and the new one is rebuilding context from scratch.
• Year 4 — an OS update breaks a third-party driver. A touchpanel battery dies. The wireless lighting hub needs a firmware update that requires the dealer's password. Service calls go to $1,000-$2,000 each.
• Year 5 — something major fails. Often the central processor. The original dealer is gone, sold, or refusing to service legacy installs. New dealer walks in, looks at the project file, and says: "I'm not touching this. We need to start over." That "start over" is a $40-80K rip-and-replace.

I've never seen one happy residential customer after three years. Usually by five, when it starts falling apart, the original company is long gone — and the next company is afraid to touch the ticking time bomb. They want to do a full upgrade.— Rick Baron, 22 years residential AV install

3. The 3-year wall

The pattern is so consistent I can predict it now. The honeymoon is real. The fatigue is real. Between them is what I call the 3-year wall — the point at which the system has accumulated enough small annoyances, change requests, and minor failures that the homeowner stops thinking of it as "magic" and starts thinking of it as "the thing I'm constantly paying to fix."

That shift is psychological more than mechanical. Nothing catastrophic has broken. It's a thousand small things — a button that does the wrong thing, a scene that no longer matches the season, a TV that drops off the matrix occasionally, a touchpanel that's getting slow. Each is fixable. Cumulatively they sap the goodwill that the install honeymoon created.

At year 3, the homeowner is no longer evangelizing the system. By year 4, they're warning friends not to do it. By year 5, they're shopping for the rip-and-replace.

4. The 5-year ticking time bomb

Around year 5 is when the structural failures start. The big ones:

• Processor / hub end-of-life. Manufacturers stop updating drivers, support a current product line, and quietly orphan the previous one. The hardware still works, but you can't add modern devices to it, you can't get fresh drivers, and you can't get manufacturer support if something goes wrong.
• Touchpanel obsolescence. The in-wall panels are bespoke hardware that no longer get software updates. They start running slow, the battery wears out, the screen flakes. Replacement panels cost $1,500-$4,500 installed each.
• The dealer disappears. Custom installation is a tough business. Dealers get acquired, close, or pivot away from residential. Their proprietary project files don't always transfer cleanly. Even when they do, the next dealer rarely has the context.
• The new dealer wants to rip and replace. This is the most predictable part. The new dealer doesn't want to inherit liability for 7-year-old programming they didn't write. They quote a "full system refresh" — usually $40-80K, sometimes more — rather than service the existing system. From their business standpoint, this is rational. From the homeowner's standpoint, it feels like extortion.

5. The economics nobody warns you about

The deepest problem with whole-home control in residential isn't engineering. It's that the relationship between dealer and homeowner is mathematically incapable of working.

Here's why:

The client's reality. You spent more on this system than most people spend on a car. In your mind, that buys "perfect" — and perfect means infinite tweaks. The thermostat schedule needs adjusting after the kids' soccer schedule changes. The kitchen scene needs to fade slower. The new TV needs to be added to the remote. You added a Sonos to the patio that needs to join the multi-room group. Each ask feels like five minutes. You assume it's covered under what you already paid.

The dealer's reality. None of those changes are five minutes. The programmer has to drive to your house (1 hour each way), open Composer Pro or the equivalent toolchain (15 min), make the change, push it to your processor and reboot if needed (15-30 min), test that nothing else broke (15-30 min), drive back. That's $400-$600 of actual labor cost on the dealer's side for "moving a button." A $1,500 service call for a few small changes isn't a markup — it's the real cost of having a certified programmer in your home.

The mismatch. Nobody can sustain free change requests at that cost. So one of two things happens:

• The dealer eats the labor until they can't anymore, then stops returning calls. The homeowner is stuck. The dealer eventually goes out of business or pivots away from residential.
• The dealer starts charging realistic rates. The homeowner feels nickel-and-dimed on a system they thought was finished. The relationship wears thin. By year 3-5, they're looking for a way out.

This isn't anyone's fault. Both sides are reasonable in isolation. The model is just broken in residential. A car needs oil changes and nobody expects them free. A control system needs ongoing programming, but the original sale framed the install as "we'll handle everything," which created an obligation no business can support.

6. The modular best-in-class stack

So if not Control4, Crestron, or Savant — then what? For 95% of the experience, with none of the handcuffs, here's the stack I'd actually install in my own home and what I recommend to clients who are willing to step away from the proprietary path.

7. Why modular wins long-term

The modular approach beats the unified-processor approach for one reason that nobody talks about until year 5: nothing is load-bearing.

When your Lutron hub needs a firmware update, your Sonos still works. When Sonos pushes a software change, your locks still unlock. When you replace your Apple TV with the next-gen model, your shades don't need to be reprogrammed. When Ring releases new hardware, your lighting doesn't care. Every product runs on its own update cycle on its own protocol, with its own app, with its own customer support.

Compare that to a Control4 or Crestron install. The central processor IS the load-bearing wall. When it fails or goes end-of-life, the whole house's automation goes with it. Touchpanels become inert. Drivers stop loading. The audio matrix loses control. The lighting scenes break because the scene logic lived in the processor, not the lighting hub. The shades stop responding because their integration was processor-mediated.

The modular stack also gives you something the proprietary path never can: component-by-component upgradability. The Lutron lighting installed today still works perfectly in 2036, even if you've replaced the Apple TV three times, swapped the Sonos lineup twice, and added a new doorbell. There's no system-wide refresh cycle. There's no all-or-nothing.

8. The honest scorecard — Control4 vs Crestron vs Savant vs AMX

Each of these platforms has its supporters and its strengths. Here's where each one actually wins, and where each one fails — based on what I've seen in 22 years of residential installs.

9. The exception list

I'm not going to claim there's never a case for a Control4 or Crestron install in residential. Just that the exception list is narrow, specific, and probably narrower than the industry would have you believe. Here's where I might still recommend a unified control system:

• Mansions over 15,000 square feet with full-time staff. Multiple theaters, integrated commercial-grade security, 12+ HVAC zones, motorized everything, and a dedicated facilities employee whose job includes managing the system. The maintenance budget is realistic and ongoing.
• Multi-property setups where one app needs to manage three+ homes. Even here, modern alternatives (Apple Home with shared invitations, multi-location Sonos) cover most of the use case.
• Specific commercial-grade requirements inside a home. A home recording studio that needs broadcast-grade integration. A medical-grade home health setup. Edge cases that almost no real client actually has.

That's the list. Maybe 1% of luxury homes. If you're outside that list, the modular stack is genuinely the better long-term play — and almost no installer will tell you that, because almost no installer makes money saying it.

10. Ten questions to ask BEFORE you sign a $50K+ contract

Any dealer worth signing with should answer all ten clearly and in writing. The contract should reflect those answers. If a dealer hesitates on the password question, the response-time question, or the "talk to my 5-year-old clients" question — walk away.

11. When to call us

Verified 2026-05-21 by Bear & Rick Baron. About Rick (22 years residential AV install) · About Bear (38 years cable industry) · Affiliate disclosure.