Coordinator Guide

At the start of each day, for every project you cover:

1. Open the NOC Dashboard (sidebar → Network Operation Center → your project → NOC Dashboard).
2. Read the Daily Report morning briefing — a once-per-day snapshot with a health verdict and the sites that recovered or newly went down overnight.
3. Note the offline count and online % in the KPI strip so you know where the day stands.

Three views tell you what needs attention, in order:

• KPI strip — the headline offline count for the project.
• Outage Triage — the live list of currently-down sites, already grouped into area-wide clusters vs. isolated faults, with a suggested next action.
• Sites Map — down sites are color-coded, so you can see at a glance whether an outage is geographically clustered.

Prioritize sites that have been down the longest — these are the ones most likely to need a real fix rather than a brief flap.

For each down site, work the cheapest action first:

1. Check the provider cloud/controller for the device status and logs. If it's reachable or flapping, a remote reset often clears it.
2. If that doesn't bring it back, call the site contact / beneficiary. Ask what's happening on the ground — power brownout, weather, an unplugged router — and have them power-cycle the router.

Most outages are resolved at these two steps. A truck roll is expensive — never jump straight to it.

Escalate when the remote reset didn't work and the on-site power-cycle also failed — but before you assume a physical fault. When you escalate, hand off the full picture:

• Project, site/group name, and device
• How long it's been down
• What you already tried (remote reset? contact reached? power-cycled?)
• Anything the beneficiary reported (power, weather, damage)

The full ladder you're working within:

When three or more sites within about 10 km go down together, the Outage Triage flags it as an area-wide cluster. These almost always come from a shared cause — a power utility outage or a backhaul provider issue — not the sites themselves.

For clusters, hold dispatch and coordinate with the power/backhaul provider. Sending an installer to one site won't fix a regional outage.

• Every down site has a recorded action (reset / called / escalated).
• Long-running outages have a ticket or a clear next step.
• Area-wide clusters are noted with the suspected shared cause.
• Anything handed to IT Support has the full context above.