Is Your Smart Home Actually Healthy? Building a Home Assistant Health Dashboard

A smart home can look perfectly healthy while quietly falling apart behind the scenes.

Your dashboard shows the living room temperature, the hallway lights respond to motion, and your electricity consumption appears perfectly reasonable. Everything is green. Everything is wonderful.

Except that one Zigbee sensor stopped reporting three days ago. Your last successful backup was created before you changed half your configuration. An integration has disconnected, and an automation is still waiting for an entity that no longer exists.

Welcome to smart home administration, where everything works beautifully until you need it most.

The solution isn’t necessarily another integration, another dashboard card, or a complicated automation. Sometimes, what you need is a dedicated Home Assistant Health Dashboard: one place to see whether the system behind your smart home is doing its job.

In this guide, we’ll build a practical monitoring dashboard using Home Assistant’s built-in features, the System Monitor integration, backup information, and a few carefully chosen automations. No expensive monitoring service required, and no need to turn your smart home into a miniature data centre.

What Should a Health Dashboard Actually Monitor?

Before adding cards, decide what you’re trying to detect.

A useful dashboard should answer five questions:

1. Is Home Assistant itself running comfortably?
2. Are important integrations and devices available?
3. Are backups completing successfully?
4. Are batteries and other maintenance-sensitive devices approaching a problem?
5. Have automations or configuration changes introduced issues?

These are different kinds of health checks. A server can have plenty of free memory while a Zigbee sensor is offline. A backup can complete successfully while an automation is broken.

Consequently, one sensor cannot tell you everything.

I recommend separating your dashboard into a few clear sections:

• System: CPU, memory and storage.
• Devices: availability and battery levels.
• Backups: recent backup activity and failures.
• Maintenance: repairs, updates and troubleshooting reminders.

Start small. A dashboard that highlights five meaningful problems is usually more useful than one displaying fifty metrics that nobody checks.

Step 1: Monitor the Home Assistant Host

Home Assistant already provides a useful starting point through the built-in System Health integration.

Navigate to Settings → System → Repairs, then open the three-dot menu and select System information. Here you can inspect information about your installation, Home Assistant version, operating system, architecture and other system details.

This is useful when troubleshooting, but it isn’t a continuously updated performance dashboard.

For that, we can use the System Monitor integration.

Install System Monitor

1. Open Settings → Devices & services.
2. Select Add integration.
3. Search for System Monitor.
4. Add the integration and select the sensors you want to monitor.

System Monitor can expose CPU usage, memory consumption, disk usage, network statistics and other host information. Available sensors depend on your installation and environment, and the integration’s entities are disabled by default until you enable the ones you need.

Official documentation: System Monitor.

For a first dashboard, enable the sensors relevant to your hardware, such as:

• CPU usage.
• Memory usage.
• Disk usage.
• Network activity, if useful for troubleshooting.

Add them to a standard Entities or Gauge card.

For example, a Gauge card can display a CPU usage sensor:

type: gauge
entity: sensor.processor_use
name: CPU Usage
min: 0
max: 100
needle: true

This assumes your installation exposes an entity with that ID. Entity IDs can differ, so select the actual sensor from your own entity list.

The gauge is only a visualisation. It doesn’t determine whether the CPU usage is healthy, and a short spike isn’t automatically a problem.

Look for sustained high utilisation, unexpected memory pressure or storage that is approaching capacity. A temporary spike while creating a backup or updating Home Assistant may be entirely normal.

Step 2: Find Devices That Have Gone Missing

This is where a large smart home becomes interesting.

When you have dozens or hundreds of Zigbee, Z-Wave, Wi-Fi and Matter devices, checking every device individually isn’t practical.

Home Assistant represents the state of an entity using values such as on, off, a measurement, unknown or unavailable. The last two deserve attention, but they don’t always mean something is broken.

An entity may temporarily be unavailable during startup, an integration reload, a network interruption or maintenance. Some entities are intentionally unavailable when a device is asleep or a particular feature isn’t supported.

Create a device availability section

Start with your most important devices:

• Motion sensors responsible for lighting.
• Leak detectors.
• Door and window sensors.
• Temperature sensors used by heating automations.
• Network equipment and other critical infrastructure.

Add these entities to an Entities card so their current states are easy to inspect.

For example:

type: entities
title: Critical Device Status
show_header_toggle: false
entities:
  - entity: binary_sensor.hall_motion
    name: Hallway Motion
  - entity: binary_sensor.kitchen_leak
    name: Kitchen Leak Sensor
  - entity: sensor.living_room_temperature
    name: Living Room Temperature

Replace the example entity IDs with your own.

This is deliberately a curated list rather than an attempt to display every entity in your installation. A dashboard full of expected offline entities, unused sensors and optional features can obscure the devices that actually matter.

For a larger installation, create separate lists for critical devices, environmental sensors and less important equipment.

You can also use an Entity Filter card to show entities in particular states, but you must provide the entities you want it to evaluate. It isn’t an automatic inventory of every unavailable entity in Home Assistant.

The goal is to make a real problem visible, not to turn every sleeping battery-powered sensor into an emergency.

Step 3: Keep an Eye on Battery Levels

Battery-powered devices are among the most common causes of unexpected maintenance in a smart home.

A sensor can work perfectly for months and then stop reporting because its battery is depleted. The problem is particularly frustrating when the device is tucked behind furniture or installed in a place you rarely visit.

Create a dedicated battery section using the battery sensors provided by your devices.

type: entities
title: Battery Watch
show_header_toggle: false
entities:
  - entity: sensor.hall_motion_battery
    name: Hallway Motion
  - entity: sensor.front_door_battery
    name: Front Door Sensor
  - entity: sensor.leak_sensor_battery
    name: Leak Sensor

Again, these are example entity IDs.

Where your devices provide battery percentage, you can sort the list by the lowest reading or use a suitable visualisation to make low levels easier to spot.

Don’t assume every battery sensor reports an accurate percentage. Some devices report voltage, some update infrequently, and others use a percentage that changes in large steps.

For critical devices, consider setting a low-battery threshold appropriate to the device rather than relying on one universal percentage.

A smoke detector or leak sensor deserves more attention than a decorative remote control.

Step 4: Verify That Your Backups Are Working

Backups are one of the most important parts of Home Assistant maintenance.

The Backup integration supports creating and restoring backups across Home Assistant installation types. You can configure automatic backups through the user interface and inspect their status using the provided entities.

Open Settings → System → Backups to review your backup configuration and available backups.

For dashboard monitoring, look for the entities supplied by the Backup integration, including the last successful automatic backup where available.

You can then display the relevant entities in a dedicated card.

For example, use an Entities card with the actual backup sensors exposed by your installation:

type: entities
title: Backup Status
show_header_toggle: false
entities:
  - entity: sensor.backup_manager_state
    name: Backup Manager
  - entity: sensor.backup_last_successful_automatic_backup
    name: Last Successful Backup

The IDs above are illustrative; verify the exact IDs in your instance before using this YAML.

A successful backup is not the same thing as a tested recovery. Review the backup location, retention settings and whether your backups survive a failure of the machine running Home Assistant.

If every copy is stored on the same physical disk, a disk failure can take out both the installation and its backups.

Official documentation: Home Assistant Backup.

Step 5: Get Notified When an Automatic Backup Fails

A dashboard is useful when you look at it. An alert is useful when you don’t.

Home Assistant’s Backup integration provides an event entity for the last automatic backup task. Its event_type attribute can indicate whether the task completed, is in progress or failed.

That gives us a straightforward way to notify ourselves when an automatic backup fails.

The following example uses the mobile notification service for your phone. Replace notify.mobile_app_your_phone with the notification service available in your installation.

alias: Home Assistant Backup Failure Alert
description: Notify me when an automatic backup fails
triggers:
  - trigger: state
    entity_id: event.backup_automatic_backup
conditions:
  - condition: template
    value_template: >
      {{ state_attr('event.backup_automatic_backup',
                    'event_type') == 'failed' }}
actions:
  - action: notify.mobile_app_your_phone
    data:
      title: "Home Assistant Backup Failed"
      message: >
        An automatic backup has failed.
        Check Settings > System > Backups for details.
mode: single

This automation assumes that your installation exposes the event entity as event.backup_automatic_backup and that your mobile notification service has the example name shown above. Confirm both entity IDs in Developer Tools before enabling the automation.

When the backup event changes, the condition checks whether its event_type attribute is failed. If it is, Home Assistant sends a notification.

The automation does not create a backup or repair the failure. It simply makes the problem harder to miss.

For additional reliability, consider checking the last successful backup as well. A backup process that hasn’t failed but has stopped running can be just as concerning.

Official documentation: Backup integration.

Step 6: Add Repairs and Updates to Your Routine

A healthy installation isn’t just one with plenty of free memory and devices that respond to commands.

Configuration problems and pending updates matter too.

Home Assistant provides its Repairs interface for reviewing detected issues. Some integrations can report problems there, and custom integrations such as Spook can add their own diagnostic findings.

This is particularly useful after changing devices, renaming entities or updating integrations.

Rather than attempting to duplicate the Repairs interface inside a dashboard, add a maintenance reminder or a Markdown card with a short checklist.

For example:

type: markdown
title: Weekly Maintenance
content: |
  ## Smart Home Health Check

  - Review **Settings → System → Repairs**
  - Check failed automation traces
  - Verify the last successful backup
  - Review low-battery devices
  - Investigate persistent unavailable entities
  - Check pending updates before installing them

This card is intentionally simple. It doesn’t claim to know whether your system has problems; it reminds you to review the places where Home Assistant reports them.

If you use Spook, this is a natural companion to the Spook guide. Spook can help identify certain configuration issues, while your health dashboard gives you a broader overview of system resources, devices and maintenance.

They complement one another rather than replacing one another.

Step 7: Put Everything Together

Now it’s time to organise the information into a dashboard that makes sense at a glance.

Home Assistant’s Sections view is a convenient way to arrange cards into groups without relying on a complicated collection of nested stacks.

Create a new dashboard or add a view dedicated to maintenance. Choose the Sections layout and organise the cards into these groups:

Section	Suggested cards
System	CPU, memory and disk usage
Devices	Critical sensors and availability
Batteries	Battery levels for important devices
Backups	Backup manager and last successful backup
Maintenance	Weekly checklist and selected update entities

Keep the most important information near the top. If storage is almost full or a critical leak sensor is unavailable, you shouldn’t need to scroll through twenty cards to discover it.

You can also use conditional visibility to show selected cards only when a particular condition is met. For example, a warning card can be displayed when a monitored entity enters an unexpected state.

For the initial version, however, a clear layout with a handful of useful cards is more than enough.

Official documentation:

• Home Assistant Dashboards
• Sections view
• Dashboard cards

What About Zigbee, Z-Wave and Network Health?

If you run a larger smart home, you may want to go beyond the basic dashboard.

Your Zigbee coordinator, Z-Wave network, MQTT broker, Wi-Fi access points and other infrastructure can all affect how reliably Home Assistant operates.

However, there is no universal sensor that tells you whether an entire Zigbee or Z-Wave network is healthy. Each integration exposes different information, and some useful diagnostics are available only through integration-specific tools.

A sensible approach is to monitor the components you actually use.

For example:

• Check that your Zigbee and Z-Wave integrations are loaded and functioning.
• Monitor the availability of your MQTT broker if your setup depends on it.
• Review network equipment for connectivity issues.
• Use integration-specific diagnostics when devices repeatedly become unavailable.
• Investigate patterns rather than reacting to a single temporary interruption.

A device that becomes unavailable once during a restart may be perfectly normal. The same device disappearing every few hours deserves investigation.

For Zigbee, also remember that a device’s availability can depend on its reporting behaviour, routing and power-saving design. Don’t confuse an inactive battery-powered device with a failed network.

Common Mistakes to Avoid

A health dashboard should reduce uncertainty, not create more of it.

Here are a few traps worth avoiding.

Monitoring everything. Every additional metric adds visual clutter. Focus on things that can affect reliability, safety or maintenance.

Treating every unavailable entity as a failure. Some entities are optional or intentionally unavailable. Investigate the device’s normal behaviour before creating alerts.

Relying on a dashboard alone. A warning displayed on a screen won’t help if nobody looks at that screen. Use notifications for important events.

Ignoring storage and backups. A system can appear healthy right up until it runs out of disk space or loses its configuration. Check both the backup history and where the copies are stored.

Assuming a successful backup guarantees recovery. Verify that your backups are accessible and plan how you would restore the installation.

Installing too many monitoring integrations. Start with built-in tools. Add third-party components only when they provide information you genuinely need.

Final Thoughts

Home Assistant makes it possible to build an impressive smart home, but reliability depends on more than the number of devices you can connect.

A dedicated health dashboard gives you a place to monitor the infrastructure behind your automations, notice maintenance issues and investigate problems before they turn into a frustrating evening of troubleshooting.

Start with system resources, critical devices, batteries and backups. Add network-specific diagnostics as your installation grows, and use notifications for failures that genuinely need your attention.

Most importantly, keep the dashboard useful. The aim isn’t to achieve a perfect sea of green icons. The aim is to know when something needs attention and have enough information to understand what to do next.

After all, a smart home should make life easier. It shouldn’t require its own full-time IT department.

And if something does go wrong, at least you’ll know where to start looking.