Smart home security connects cameras, door sensors, motion and locks so they report to one app and can act on each other. The design question is which platform holds the devices and what still works when the internet is down. A pile of separate apps is not a system.
A smart home system earns its place by reducing the number of things a household has to remember. One app, one notification habit and a handful of devices that work together beat a drawer full of gadgets that each need their own bridge.
We choose the platform first, then the network, then the devices, because the reverse order is how people end up replacing everything after a year.
Based in Westminster. Serving Denver and clients across Colorado. See all Colorado service areas.
One platform or several apps
There are two workable models. One is a single ecosystem where cameras, sensors, locks and the hub come from one vendor and are configured together. The other is a local controller that speaks to devices from several manufacturers through common standards. The first is simpler. The second survives a vendor decision better.
What does not work is a camera from one cloud, a lock from another and a doorbell from a third, with no shared automation. The household gets three notifications for one delivery and stops reading any of them.
- A platform decision made before any device is bought.
- Devices that report to the platform, not only to their own cloud.
- An automation engine, so devices can trigger each other.
- A written list of accounts, so the household owns them rather than an installer.
Choose the network before the gadgets
Most smart home problems are Wi-Fi problems with a user interface. Doorbells, cameras and locks sit at the edge of coverage, which is exactly where signal is weakest. A mesh node in the same room as the camera is a better purchase than a more expensive camera.
Putting the devices on their own network or guest network separates them from the computers holding personal files, and it makes it easier to see what is connected. For a larger property, a small managed network with a separate wireless network for devices is worth the effort.
- Signal measured at each device position before it is mounted.
- A separate network for devices, with its own password.
- Cameras on a wired connection wherever a cable can be run.
- An inventory of connected devices, checked once a year.
Sensors worth installing first
The first devices should be the ones that report something a household can act on. Door and window contacts on the entry doors, a motion sensor covering the main hallway and a video doorbell on the front approach cover most of what a break-in looks like. Water leak sensors and smoke or carbon monoxide listeners come next, because the losses they prevent are the more common ones.
Glass break sensors and interior cameras matter in specific situations rather than in every house. A camera in a bedroom is a privacy decision, and the household should make it deliberately rather than inherit it from a package.
Notifications people will act on
An alert is useful when it is specific, prompt and rare. Person detection at the front door is worth a notification. Every car passing on the street is not. Motion zones, object classification and schedules reduce the volume, and a short check after the first week usually identifies the handful of rules producing most of the noise.
- Zones that exclude the footpath and the street.
- Schedules that match when the household is actually home.
- Different treatment for a door alert and a driveway alert.
- A weekly check of what was recorded, not only what alerted.
Privacy, cloud and local recording
Where the footage lives is a household decision with consequences. Cloud services are convenient and survive the theft of a recorder. Local recording keeps video in the house and needs a device that comes back after a power cut. Many homes use both, recording locally and sending event clips to a cloud service.
Privacy settings deserve the same attention as camera placement. Two-way audio on a doorbell records conversations on the porch, and a camera that sees a neighboring window creates a problem no firmware update will solve.
- Cameras aimed at the property, not at a neighbor.
- Recording that continues when the internet does not.
- Accounts protected with a unique password and multi-factor authentication.
- A house rule on which cameras record audio and which do not.
The two-year test
A smart home system should be judged two years after installation. Are the batteries changed, are the devices still supported, and does the app still work on the phones in the house? A system that depends on a subscription a family forgot to renew is not security.
We plan for that from the start: devices with replaceable batteries, a written schedule of what needs updating and when, and a preference for hardware that keeps its local functions when a cloud service does not answer.
Related:Residential security servicesResidential alarm systems
Frequently asked questions
Do smart devices work without internet?
Some functions do and most cloud features do not. A local controller keeps scenes, automations and local recording running through an outage, while remote viewing and app control stop until the connection returns. That is the first question to ask of any platform.
Can we mix devices from different brands?
Yes, if the platform supports the standard those devices use. Mixing is more reliable for sensors and lighting, and is least reliable for cameras and locks, where vendors keep the useful features behind their own app.
How many cameras does a house need?
Usually fewer than people expect. A doorbell on the main approach, one camera on the driveway or side gate and one on a rear door cover most houses. Interior cameras are a choice about privacy, not a default.
Is a smart lock worth it for security?
For control and records, yes: codes can be issued and revoked, and the lock reports when a door was opened. For physical strength, the door and frame matter more than the lock, and a smart lock on a weak door is still a weak door.