Uptime monitoring service with SLA and client reports
Contracts promise 99.9 percent. Clients ask what they pay the retainer for. Leadership wants to know whether last month was better than the one before. UptimePilot turns every check into reports you can send: uptime percent, response time percentiles, a full incident log, a monthly SLA PDF and white-label client reports with your own logo.
What every uptime report includes
Each report covers a period you pick, a monitor group or a single monitor, and the same numbers your team sees in the dashboard. Nothing in a report is estimated or smoothed. It is built from the actual check results recorded by the probes.
- Uptime percent for the period, per monitor and per group
- Total confirmed downtime in minutes and the number of incidents
- Response time percentiles p50, p95 and p99, plus the average
- Incident log with start, end, duration, affected monitor, regions and the error recorded
- Maintenance windows listed separately, so readers see what was scheduled
- Domain and SSL certificate expiry dates for the monitors in the report
Starter includes uptime and response time reports. Team adds the monthly SLA report as a PDF. Business adds white-label client reports and the REST API for pulling the same data into your own tools.
How UptimePilot calculates uptime percent
An uptime number is only useful if everyone agrees how it was counted. UptimePilot uses one clear rule, shown at the bottom of every report.
Uptime percent equals the monitored time minus confirmed downtime, divided by the monitored time. Two details decide whether that number is fair.
- Confirmed downtime only. A monitor is marked down only after a second region confirms the failure. A single failed probe caused by a route between one region and your server does not count as downtime. It stays in the raw check log, not in the SLA number.
- Maintenance windows excluded. Time inside a maintenance window you scheduled is removed from the monitored time. A deploy at 2 a.m. that you announced does not lower the number your contract measures.
- Downtime starts at the first failed check that was later confirmed and ends at the first successful check after it, so the duration is measured, not rounded to the hour.
- Degraded states, such as slow responses over your threshold, appear in the report but do not count as downtime unless you set the monitor to treat them that way.
SLA uptime percentages and the downtime they allow
This table uses an average month of 30.44 days (43,834 minutes) and a year of 365.25 days. It is plain arithmetic and holds for any service.
| SLA target | Allowed downtime per month | Allowed downtime per year |
|---|---|---|
| 99% | 7 h 18 min | 3 days 15 h 40 min |
| 99.5% | 3 h 39 min | 1 day 19 h 50 min |
| 99.9% | 43 min 50 s | 8 h 46 min |
| 99.95% | 21 min 55 s | 4 h 23 min |
| 99.99% | 4 min 23 s | 52 min 36 s |
The gap between the lines is bigger than it looks. Moving from 99.9 to 99.99 percent cuts the allowed monthly downtime from about 44 minutes to under 5. A single slow rollback can use the whole monthly budget at 99.99. That is why check interval matters for SLA reporting. With checks every 30 seconds on Team and above, UptimePilot measures a short outage to within half a minute. With checks every minute on Starter, to within one minute. More on the math in the post on 99.9 uptime.
Response time percentiles instead of averages
An average hides the slow requests your users feel. If 95 requests take 200 ms and 5 take 6 seconds, the average is about 490 ms, which describes nobody. Percentiles tell the real story. The p50 (median) shows the typical response. The p95 shows what the slowest one in twenty checks looked like. The p99 shows the tail, where timeouts and overloaded workers live. UptimePilot reports all three per monitor and per region, so you can see whether EU Central visitors wait longer than US East ones. Detailed charts are on website performance monitoring.
Monthly SLA report as a PDF
On Team and above, UptimePilot prepares a monthly SLA report as a PDF for each monitor group. It shows the SLA target you set, the measured uptime, the remaining downtime budget, a list of incidents with durations, and the maintenance windows that were excluded. Engineering leads attach it to enterprise customer reviews. Finance uses it to check SLA credit claims. The PDF is ready on the first day of the month and stays available for the full data history of your plan.
White-label client reports for agencies and service providers
On Business and Enterprise, client reports carry your logo and colors instead of UptimePilot branding. Each client gets a report for their own monitor group: their sites, their uptime percent, their incidents and what your team did about them, plus upcoming SSL and domain renewals. Send it every month as part of the retainer. Agencies running many client sites will find the full workflow on uptime monitoring for agencies, and SaaS teams can link the same numbers from a public status page.
Data history for long term reporting
Reports can only cover what is stored. Starter keeps 90 days of check data, Team keeps 1 year, Business 2 years and Enterprise 3 years. Annual SLA reviews and year over year comparisons need at least Team. Business also exposes the same data through the REST API for your own dashboards.
Which plan fits your reporting needs
Every plan runs the same monitoring engine. Reporting grows with the plan. Compare all limits on uptime monitoring pricing.
- Starter, $49 a month or $24 a month billed yearly: uptime and response time reports, 50 monitors, 90 days of history
- Team, $99 a month or $49 a month billed yearly: adds the monthly SLA PDF, 200 monitors, 30 second checks, 1 year of history
- Business, $249 a month or $124 a month billed yearly: adds white-label client reports and REST API, 1,000 monitors, 2 years of history
- Enterprise, $599 a month or $299 a month billed yearly: adds SSO (SAML), audit log and invoicing, 3,000 monitors, 3 years of history
01
Does a failed check from one region lower my uptime percent?
No. Only confirmed downtime counts, which means a second region has to confirm the failure. Unconfirmed failures stay in the raw check log for diagnosis.
02
Are maintenance windows excluded from the SLA number?
Yes. Time inside a maintenance window you scheduled is removed from the monitored time, and the report lists those windows separately.
03
Which plans include the monthly SLA PDF?
Team, Business and Enterprise. Starter includes uptime and response time reports in the dashboard.
04
Can I send reports to clients under my own brand?
Yes, on Business and Enterprise. White-label client reports use your logo and colors and cover only the monitor group of that client.
05
How precise is the measured downtime?
It follows the check interval. With 30 second checks on Team and above, an outage is measured to within half a minute. With 1 minute checks on Starter, to within one minute.
06
Can I pull report data into my own tools?
Yes, on Business and Enterprise through the REST API, which returns monitors, check results and incidents.
Show the uptime you deliver
Set your SLA target, add your monitors and get your first report at the end of the month.