Understanding Uptime and SLA Reports
Last updated: 2026-02-25
Uptime is one of the most important metrics for any online service. Here’s how we calculate it and how to use it effectively.
What Is Uptime?
Uptime is the percentage of time your service was available over a given period. It’s calculated as:
Uptime % = (Total time - Downtime) / Total time × 100
For example, if your service was down for 1 hour in a 30-day month (720 hours):
Uptime = (720 - 1) / 720 × 100 = 99.86%
Uptime Standards
Here’s what different uptime levels mean in real terms:
| Uptime % | Downtime per Month | Downtime per Year |
|---|---|---|
| 99.9% | ~43 minutes | ~8.7 hours |
| 99.95% | ~22 minutes | ~4.4 hours |
| 99.99% | ~4.3 minutes | ~52 minutes |
| 99.999% | ~26 seconds | ~5 minutes |
Most businesses aim for 99.9% or higher.
Viewing Uptime Statistics
On each monitor’s detail page, you’ll find:
- Current uptime - Percentage over the last 30 days
- Response time graph - Historical performance trends
- Uptime bars - Visual day-by-day availability
- Downtime log - List of outages with duration
SLA Reports
SLA (Service Level Agreement) reports provide detailed uptime documentation. These are useful for:
- Internal reporting - Share with management
- Customer commitments - Prove SLA compliance
- Trend analysis - Spot reliability patterns
Generating an SLA Report
- Go to your monitor’s detail page
- Click SLA Report or Generate Report
- Select the date range
- Download or view the report
What Affects Uptime?
Your calculated uptime includes:
| Counts as Downtime | Does NOT Count |
|---|---|
| Failed monitor checks | Scheduled maintenance (if configured) |
| Timeout errors | Paused monitors |
| HTTP error codes (5xx) | Checks during planned windows |
Reading the Uptime Chart
The uptime chart on your status page shows:
- Green bars - Service was fully available that day
- Yellow bars - Partial outage or degradation
- Red bars - Significant downtime occurred
- Gray bars - No data (monitor wasn’t active)
Hover over any bar to see exact numbers.
Improving Your Uptime
| Strategy | How It Helps |
|---|---|
| Use multiple servers | Redundancy prevents single points of failure |
| Add load balancing | Distributes traffic across servers |
| Monitor proactively | Catch issues before users notice |
| Set up alerts | Respond faster to outages |
| Review trends | Find patterns and fix root causes |
Exporting Uptime Data
You can export your uptime data for external analysis:
- Go to Data Export
- Select the monitors and date range
- Download in your preferred format
Quick FAQ
Q: If I resume a monitor set to 30 seconds, will it still run every 30 seconds?
A: Yes. It keeps the frequency you selected. You may see a short delay right after resuming, then checks continue at the configured cadence.
P95 Latency SLA Alerts (Team+)
Beyond binary uptime, you can set a P95 response-time threshold on any monitor. When the rolling 24-hour P95 latency exceeds that threshold, you receive an alert even if overall uptime is 100%.
Configure it on the monitor edit screen under Advanced → P95 SLA threshold (ms).
This is useful for API-level SLAs that specify latency targets (e.g., “p95 < 500 ms”).
Error Budget & SLA Burn-Rate Alerts (Team+)
If you define an SLA target (e.g., 99.9%), StatusPage.me tracks your monthly error budget — the total downtime you can afford. When the current hourly burn rate would exhaust that budget in less than 2 days, a burn-rate alert fires.
Example: 99.9% SLA → 43.8 min/month budget. If you’re down 5 min/hour, that’s 6.8× the sustainable rate — budget exhausted in ~6 hours.
This concept is inspired by Google SRE error budgets and provides an early warning before an actual SLA breach.
See also: Error Budget & SLA Burn-Rate Alerts