My Home Assistant automations stopped failing once I stopped trusting my sensors

Published Sep 16, 2026, 8:00 AM EDT Anurag is an experienced journalist and author who’s been covering tech for the past 5 years, with a focus on Windows, Android, and Apple. He’s written for sites like Android Police, Neowin, Dexerto, and MakeTechEasier. Anurag’s always pumped about tech and loves getting his hands on the latest gadgets. When he's not procrastinating, you’ll probably find him catching the newest movies in theaters or scrolling through Twitter from his bed. When I first started out with Home Assistant, I always wrote “happy path” automations. I was naive enough to assume that when a door opened, the Zigbee sensor would instantly broadcast that state, Home Assistant would receive it within milliseconds, and the light would turn on. But I quickly learned things aren't quite so rosy. Smart homes are prone to latency, battery drain, and packet loss. The right way to configure your home automations is to follow Murphy’s law, which says "anything that can go wrong will go wrong." Once you design your setup around the assumption that sensors will be delayed, go offline, or miss a trigger entirely, your automations become much more reliable. Your smart home sensors can and will fail Better be prepared for the same As much as I would love to tell you that every sensor and smart home device works exactly as expected, I would be lying. I was lying to myself when I first started building my smart home too. I assumed everything would just work, but over time I learned just how many ways things can go south, especially with sensors. For example, a Zigbee or Z-Wave motion sensor might fire, but a burst of 2.4GHz Wi-Fi interference can cause the packet to get lost. If your automation triggers only when the sensor changes from off to on, you may have missed your window. Battery-powered sensors can also sleep or drop offline. If your automation depends on a strict condition, such as only turning on the heater when the temperature is below 65°, and the sensor is currently reporting as unavailable, the automation can fail altogether. Someone in the Home Assistant community explained this perfectly with a simple example. Imagine it is a cold night, and you have an automation that turns on a space heater when your bedroom drops below 64°F, then switches it off once the temperature reaches 68°F. Now the Zigbee temperature sensor on your nightstand dies in the middle of the night and suddenly goes unavailable. Your automation is basically blind at that point. It has no fresh temperature reading to work with, so the condition that is supposed to shut the heater off never gets triggered. A few hours later, you wake up in a room that is way hotter than you ever intended. Always build a failover sensor Remember Murphy's law To stop your automations from failing randomly, you need to move toward defensive configuration. Instead of triggering an action at one exact moment, I trigger it based on a state that has been sustained for a period of time, or check the condition periodically. For example, rather than telling Home Assistant to check whether the garage door is open at exactly 10:00 PM and close it, I trigger the automation when the garage door has been open for 15 minutes. For critical automations, add a secondary trigger that acts as a sweeper. You can use a Time Pattern trigger that runs every 10 or 15 minutes and re-evaluates the conditions. To put this more concretely, let’s go back to the Home Assistant community example we discussed earlier. The sensor on the nightstand is still the one I would trust most because it is closest to where I am sleeping. The problem is that I do not want the entire climate automation to depend on that one battery-powered sensor staying online all night. So I would give Home Assistant a couple of backup readings to fall back on. The hallway thermostat is not in the bedroom, but it could be close enough to give a reasonable estimate of the temperature. You might also have a smart plug in the room with a built-in temperature sensor. Its reading is not perfect because the plug itself generates some heat, but it is still useful as a secondary data point. The failover sensor looks like this: template: - use_blueprint: path: TheThinkingHome/sensor_failover.yaml input: primary_entity: sensor.nightstand_temperature backup_entities: - sensor.hallway_thermostat_temperature - sensor.bedroom_smart_plug_temperature backup_weights: "80, 20" sensor_name: Master Bedroom Temperature unique_id: master_bedroom_temperature_failover unit_of_measurement: "°C" device_class: temperature state_class: measurement Under normal conditions, nothing really changes. If the sensor disappears, though, the failover entity can keep producing a usable temperature from the backups instead of leaving the automation unavailable. Choose the right hardware too A lot depends on that You can write the most defensive automation in the world, but unreliable hardware will still make your smart home harder to live with. A lot of the random delays and missed updates I blamed on Home Assistant early on were really problems with the devices. I now prefer devices that work locally over ones that depend on a vendor’s cloud. Zigbee, Z-Wave, Matter, Thread, and ESPHome devices can continue talking to Home Assistant even when your internet connection goes down. One of my colleagues at XDA saw a big improvement in presence detection after ditching traditional PIR motion sensors for mmWave radar, which can detect someone even when they are sitting still. Get more out of your smart home By combining Home Assistant with your smart home devices, you can automate a lot. You also don't have to create every automation yourself. Tools such as an AI automation suggester can help you build useful automations based on the devices already in your smart home. You can even connect AI to Home Assistant and let it help build those automations for you.