Skip to main content

Monitor Your Fleet

Once your monitors are registered in Vertex and you can sign in to Beacon, this page orients you around the day-to-day work. Each area has a fuller guide linked from here.

Everything works the same whether your monitors are AirQo hardware or come from another manufacturer.

The Short Version​

You want toGo to
Check whether the fleet is healthyThe Fleet Dashboard
Find out what's wrong with one monitorDevice Details
See what's been flagged recentlyAlerts
Chart the raw sensor dataDevice Data Analysis
Compare cohorts or regions over timePerformance Analysis
Plan a repair tripMaintenance
Produce a PDF for a funderReports

A First Week​

If Beacon is new to you, this order builds understanding fastest.

Day one — establish the baseline. Open the dashboard and note your Average Online Rate with no filters applied. That single number is what you'll compare everything against later. Then set the Fleet Filter to each of your cohorts in turn and note theirs. Differences between cohorts are usually the first real finding.

Then — find your worst performers. Switch to the comparison view and look at Needs Attention and the Uptime Range (Min–Max). The range matters more than the average: a cohort at 90% could be uniformly mediocre or mostly excellent with one dead monitor, and those need completely different responses.

Then — diagnose two or three. Open your worst monitors and read their Performance tab. Look at Daily Uptime, Data Frequency, Sensor Health, and Battery Voltage together rather than separately — the combination is what identifies a cause. Health Metrics Explained has a table of common patterns.

Then — plan a visit. Take what you found to the maintenance map and see whether the problem monitors cluster geographically. If they do, one trip may fix several.

Finally — set a rhythm. Check Alerts filtered to Critical and Today as a routine, and widen to Warning over 30 days weekly to catch slow degradation while it's still cheap to fix.

What Beacon Doesn't Do​

Beacon is a monitoring surface. Registering, deploying, grouping, and recalling monitors, and choosing who can see your data, all happen in Vertex. If something about a monitor's identity, location, or grouping is wrong, fix it in Vertex and the change follows through.

Team members and roles are managed on AirQo Nexus, not in either product.

What's Next​