Know your app is down before your users do
Navy checks your services from 18 regions, alerts the right person, and keeps your customers posted on a status page.
See pricingCustomers
Fewer pages, faster fixes
4,000 teams use Navy to hear about outages first, page the right person and keep their customers posted.
Read customer stories92%
fewer false alarm pages in the first month
Tessly64%
fewer support tickets during incidents
Corvane
“Navy's on-call schedules and escalation policies replaced three separate tools, and every incident now reaches the right engineer the first time.”
Lucía Moreno
Head of Platform, Varnel
“During a major outage our status page took more traffic than our homepage, and it stayed fast and accurate the whole time.”
Henrik Lund
SRE Lead, Ombrik
0
certificate expiries in the 18 months since rollout
Halvik1,200
monitors managed in Terraform, none edited by hand
SelvoHow it works
From a failed check to the right phone in 40 seconds
Check
Every 30 seconds from the regions you pick, over HTTP, ping, TCP or DNS.
Confirm
A failed check runs again from three other regions, so one bad network path never pages anyone.
Alert
The person on call hears first, then your team and your customers, in that order.
Recover
When checks pass again the incident closes itself and the status page says so.
Escalate
Nobody acknowledges in five minutes? The backup gets called, then the lead.
Inform
Subscribers get an email and the status page posts an update without anyone typing it.
Review
Every check, page and reply lands on one timeline, ready for the postmortem.
Report
Monthly uptime and response time per monitor, ready to send to customers with an SLA.
Status pages
Tell customers before they ask
Every monitor can feed a public status page, so the support queue stays quiet while your team works on the fix.
See status page plans- Updates itself
- Components go down and come back with your monitors, so the page is right before anyone edits it.
- Your domain, your brand
- Host it on status.yourdomain.com with your logo, colors and a private mode for internal tools.
- Subscribers
- Customers follow the components they care about and get every update by email or Slack.
- Incident history
- Every past incident stays on the page with its full timeline, so trust builds over time.
On call
Wake up one person, not the whole team
Alerts follow your rota and climb the escalation policy until someone answers. Rotations cross time zones, sync to everyone's calendar, and a shift swap takes two clicks.
See every alert channel- 12:04Paged
Samuel Ortiz
On call this week, reached by push and SMS.
- 12:06Acknowledged
Samuel answers
Escalation stops here. Nobody else is woken up.
- 12:09Skipped
Maya Lombardi
Backup, called if nobody answers within 5 minutes.
- 12:19Skipped
Marco Bianchi
Engineering lead, called after 15 minutes.
Network
The network behind every check
Every region runs on more than one cloud provider, so an outage at one of them never takes your monitoring down with it.
- Check regions
- 18
On six continents, each on two cloud providers.
- Check interval
- 30s
On every plan, from the very first monitor.
- Median time to alert
- 40s
From the first failed check to a phone ringing.
- Checks a month
- 2B+
Across every region, every minute of the day.