Smart home automation · Mission Hills.

What a smart home automation project in Mission Hills involves

The walls fight back, and that is the whole design constraint. Lath and plaster over solid framing does not take a fish tape the way drywall does, and the exterior masonry on many of these houses rules out the easy path around a corner. So we survey the attic, the basement joist bays and the old chase runs before a blade touches a wall, and we plan every drop on paper. Where a run genuinely cannot be made, we say so and design around it with wireless control rather than cutting an opening we cannot close cleanly.

The finishes are the reason people bought the house, so every visible device is a design decision rather than a detail. Original crown molding, leaded glass and paneled libraries do not forgive a keypad dropped in the wrong place. We size keypads to the existing switch boxes wherever the wiring allows, choose finishes alongside the homeowner’s designer, and put architectural speakers into ceilings a renovation has already opened. In rooms still carrying original plaster we would rather add a discreet audio zone than cut a period ceiling.

Climate deserves particular attention in these houses, and it is the subsystem homeowners most often underestimate. A 1920s house that has been renovated in stages usually has more than one heating and cooling zone, added at different times by different contractors, with thermostats in rooms that do not represent the floor they control. Bringing those onto one system — so the schedule, the shades and the lighting cooperate rather than fighting — often does more for daily comfort than anything else in the project.

And the lot matters. Mature oaks and deep setbacks are the point of a Mission Hills property, and they are also why one router at one end of the house never reaches the third floor, the yard and the structure behind the garage. The network is not a footnote to the automation here; it is the thing the automation stands on, and we plan the two together rather than discovering the gap after the controllers are installed.

• Control4. Where most Mission Hills houses land. Deep driver support means lighting, shades, climate, audio and cameras integrate without custom work, and the keypads suit period switch boxes.
• Crestron. For bespoke logic or an unusual subsystem count. Programmed in house, so the system is not orphaned by whoever wrote it — which in this village is a real risk with real history.
• Two-stack. Lutron for lighting and shades, Control4 or Crestron for the rest. The pattern that carries most large retrofits here, because Lutron solves the no-neutral problem these houses have.

What is worth automating in a house like this

The systems that get used are judged on four or five daily actions, not a feature matrix. In these houses that list is consistent: a lighting scene at the door people actually come in through, shades on the west and south glass that protect finishes as much as they manage glare, climate that behaves the same in the third-floor bedroom as it does on the main floor, audio that starts from a keypad rather than a phone, and a single goodnight press. Everything past that should be added after you have lived with the system.

What gets abandoned is a keypad with ten scenes on it where the family uses two. A keypad you have to read is a keypad you stop using, and in a house where the plate is part of the millwork, a smaller keypad in the right place beats a larger one in the wrong place every time. We start at about four scenes per main room and add only on request, after the house has been lived in with the system running.

Voice is worth being honest about. Josh.ai runs locally in the home rather than through a cloud service, which is why it is the one we install where voice matters — it keeps working when the internet does not, and it is not listening to a data centre. It is a good third way in and a poor only way in. So is the app. In a house with staff, guests and family moving through it, the physical control on the wall is the interface that has to work, and it should be designed like it.

Adopting a system somebody else left behind

A steady share of the Mission Hills work we take is not a new system at all. It is a house whose original integrator retired, moved on or stopped answering, and whose owner cannot get a driver updated, a thermostat talking again, or a password out of anyone. We adopt Control4, Crestron and Lutron systems in that position, and we service and maintain existing Savant systems as well, even though we are not a Savant dealer.

The first visit documents rather than quotes. We open the rack, photograph and label what is genuinely connected, trace what has been abandoned in place — in a house renovated three times there is usually a lot of that — and check firmware and licensing. Then you get a plain recommendation: what stays, what is at end of life, what can be brought current with programming rather than hardware, and what would honestly be cheaper to replace than to keep alive.

The other thing that visit produces is documentation, which these houses almost never have. The panel gets labelled, the runs get recorded, the configuration gets written down. That is worth as much as any hardware, because the next person who touches the house — us in five years, or someone else — is then not reverse-engineering a hundred-year-old building through three renovations.