Real talk: if you have an elderly parent (or any loved one living alone), the worry isn’t abstract. It’s the 2 a.m. “what if something happened and nobody knows?” feeling.
People usually jump straight to cameras. I get it — it feels like the obvious answer. But after about a day you realize cameras are a terrible “wellness monitor”: you can’t stare at a feed all day, people are often out of frame, and the privacy tradeoff is real.
In this video, I show a different approach: a passive, privacy-respecting activity monitor built on Home Assistant. No wearables to charge. No buttons to press. Just “normal life” leaving small signals behind — and an alert when those signals stop.
Important disclaimers (please don’t skip)
- This is not a replacement for a medical alert service (Life Alert / Medical Guardian / etc.) if your loved one truly needs that level of response.
- Do this with consent. Passive doesn’t mean secret.
The simple idea
We’re going to build a system that answers one question:
“Has normal activity happened recently?”
And we do it with three building blocks:
- The brain: a home automation hub (I use Home Assistant).
- The signals: sensors that reflect normal routines (lights, motion, power draw, water flow, door open/close).
- The magic: a countdown timer that resets whenever activity is detected.
Why timers are the “aha” moment
Think of an old-school kitchen timer. You set it, it counts down, and if it hits zero… you notice.
Now imagine a timer that restarts every time your loved one does something routine: turns on a light, walks through the kitchen, turns on the TV, uses water, opens a door. If the timer keeps getting restarted, it never reaches zero. If it does reach zero, that’s your “something might be off” moment.
In the video I demo this with a single smart bulb and a short timer (so you don’t have to watch me wait 12 hours). But the real system is the same idea at a bigger scale: multiple signals, multiple timers, and more confidence.
What I like to monitor (no cameras required)
You don’t need fancy sensors. The trick is choosing signals that naturally match someone’s routine.
- Smart lights / switches: “A light turned on/off” is an easy activity marker.
- Smart plugs: power draw changes when a TV or coffee maker turns on.
- Motion sensors: placed in a few “must pass through” areas (kitchen, hallway, living room).
- Water usage: a whole-home flow sensor can be a strong signal (coffee, hand washing, showers).
Picking timer durations (avoid false alarms)
This is where people overthink it — so here’s the practical version:
- Start with a longer window (8–12 hours) so sleep doesn’t trigger alarms.
- As you add more signal types (lights + motion + power), you can safely shorten windows.
- Consider multiple timers: one “frequent activity” timer (kitchen/motion) and one “daily heartbeat” timer (water usage, TV power).
How the automation works (in plain English)
- When any chosen sensor changes (motion detected, light toggled, TV power rises)…
- Start (or restart) a timer to a fixed duration.
- If that timer ever reaches idle/zero…
- Send a notification: “No activity detected for X hours.”
That’s it. Once you “get” the timer-reset pattern, you’ll start seeing other uses for it all over Home Assistant.
What this gives you in real life
It doesn’t prove everything is okay. What it does is reduce uncertainty. If normal activity is happening, you’re less likely to worry. If activity stops for an unusual window, you get a nudge to check in.
Next steps
- If you’re starting from scratch, the easiest hardware path is a Home Assistant Green or a small always-on box (Raspberry Pi / mini PC).
- Start with 2–3 sensors in key rooms and one timer. Expand only after you see how it behaves for a week.
- If you build your own variation of this on another platform, I’d genuinely love to hear what you did.