Lighting
I put a six button keypad by the back door of a kitchen once. I was proud of it. Every load in the room had its own button, plus a scene, plus an all off.
Six months later I was back on a service call. The bottom four buttons still had the plastic shine on them. Only the top two were ever touched.
The homeowner told me the truth without meaning to. She said, "I just push the top one until it looks right."
That is not a happy customer. That is a customer who gave up.
Three buttons is comfortable. Four is fine. Five is the top end for a room somebody walks through every day.
Six is a special case. A theater, a rack room, a master bath where the client asked for it by name. Not a hallway.
Anything past six and you are asking a person to read. People do not read a wall. They reach for it in the dark, half asleep, with a laundry basket on their hip.
A light switch works because there is only one thing to do. Your hand knows where it is without your eyes.
Add buttons and you take that away. Now the hand has to find the right one. That means looking. That means being awake. That means engraving, and engraving means reading tiny letters in a dim room.
There is a second cost too. Every button you add is a thing that can be wrong. Wrong load, wrong level, wrong ramp. More buttons is more programming, more testing, and more chances the client finds one that does not behave.
I ask one question about every button: will somebody press this at least once a week?
If the answer is no, it does not get a button. It goes in the app, or in a scene on another keypad, or it goes away entirely.
Then I sort what is left into three jobs.
Arrive, change, leave. Most rooms in most houses need nothing else.
If a room has one light, put in a switch or a dimmer. Not a keypad.
I have watched dealers put a four button keypad on a pantry because the system was already there. The client pushes the top button and the pantry light comes on. The other three buttons do nothing they care about.
That is not a smart house. That is a switch with clutter attached.
Save the keypads for places where they solve something. A room with several loads. A room that starts a path through the house. A room somebody leaves from.
Clients still want the extra scenes. That is fine. They just do not belong on the wall.
Put them in the app. There is room to label them properly there, and nobody hunts by feel. Put them on a schedule if the client keeps asking for the same thing at the same time. Put them on a button as a second tap, but only if the client will remember that.
Most people will not remember a double tap. Assume they will not, and be pleasantly surprised.
Here is the test I use before I ever build the project. I stand in the room with the client and describe the keypad out loud, in their words.
"Coming in from the garage, top button is kitchen on. Middle is dinner. Bottom is everything downstairs off."
If I can say the whole keypad in one breath and they nod, it is right. If I run out of breath, there are too many buttons.
If they say "what does the third one do again" before I have even left the room, we cut a button. Right there, standing in the room, before anything is engraved.
I hated repeating that conversation on every job. So TrimOut asks about scenes and button labels up front. Nobody has to invent them at midnight.
The best keypads I have installed all look boring. Two or three buttons, plain words, in a spot your hand finds without help.
Nobody has ever called me to complain that a keypad was too simple. They call when they cannot remember what button four does.
Start at three. Add one if you have a real reason. Stop there.