Introduction: Why SLA Uptime Matters
When negotiating hosting contracts for large-scale websites, few metrics are as critical yet as widely misunderstood as Service Level Agreement (SLA) uptime guarantees. These seemingly simple percentages—99.9%, 99.99%, 99.999%—have profound implications for your business operations, customer experience, and bottom line.
As discussed in my companion article on Hosting Large-Scale Web Sites: The Ultimate Contract Review Guide for CTOs , understanding the practical meaning of these percentages can mean the difference between an acceptable level of service interruption and a business-threatening outage that falls within contractual limits.
This guide translates abstract uptime percentages into concrete downtime durations across different time periods, helping you make informed decisions about the SLA guarantees you need and how they should be measured and enforced.
The Critical Importance of Measurement Windows
The table below illustrates why you should ensure that the “SLA Uptime Measurement Reset Window” is the same duration as your billing cycle or shorter. Without this alignment, a provider could experience a significant outage early in an annual contract and still meet their yearly SLA by maintaining perfect uptime for the remaining months—leaving you with no contractual remedy for the business impact of that outage.
SLA Uptime Percentage to Downtime Conversion Table
| Availability % | Common Name | Downtime per year | Downtime per month (30 days) | Downtime per week | Downtime per day |
|---|---|---|---|---|---|
| 90% | “one nine” | 36.5 days | 72 hours | 16.8 hours | 2.4 hours |
| 95% | 18.25 days | 36 hours | 8.4 hours | 1.2 hours | |
| 97% | 10.95 days | 21.6 hours | 5.04 hours | 43.2 minutes | |
| 98% | 7.3 days | 14.4 hours | 3.36 hours | 28.8 minutes | |
| 99% | “two nines” | 3.65 days | 7.2 hours | 1.68 hours | 14.4 minutes |
| 99.5% | 1.83 days | 3.6 hours | 50.4 minutes | 7.2 minutes | |
| 99.8% | 17.52 hours | 86.4 minutes | 20.16 minutes | 2.88 minutes | |
| 99.9% | “three nines” | 8.76 hours | 43.2 minutes | 10.08 minutes | 1.44 minutes |
| 99.95% | 4.38 hours | 21.6 minutes | 5.04 minutes | 43.2 seconds | |
| 99.99% | “four nines” | 52.56 minutes | 4.32 minutes | 1.01 minutes | 8.64 seconds |
| 99.999% | “five nines” | 5.26 minutes | 25.92 seconds | 6.05 seconds | 0.864 seconds |
| 99.9999% | “six nines” | 31.54 seconds | 2.59 seconds | 0.605 seconds | 0.0864 seconds |
| 99.99999% | “seven nines” | 3.15 seconds | 0.259 seconds | 0.0605 seconds | 0.00864 seconds |
Practical Implications of SLA Percentages
Common Industry Standards
- 99.9% (“three nines”): Often the minimum acceptable standard for business applications, allowing up to 8.76 hours of downtime per year.
- 99.99% (“four nines”): Typical for mission-critical applications, allowing just under an hour of downtime per year.
- 99.999% (“five nines”): The gold standard for telecommunications and critical infrastructure, allowing only 5.26 minutes of downtime per year.
Selecting the Right SLA for Your Business
When determining what SLA to require, consider:
- Business Impact of Downtime: Calculate the direct financial impact of website downtime (lost sales, reduced productivity) and indirect costs (damaged reputation, customer dissatisfaction).
- Cost vs. Benefit Analysis: Higher availability guarantees typically come with significantly higher costs. For example, moving from 99.9% to 99.99% availability might double your hosting costs, while the final step to 99.999% could triple or quadruple them.
- Application Criticality: Not all applications require the same level of uptime. Consider a tiered approach where your most critical systems have higher SLA requirements than less essential ones.
- Industry Expectations: Consider what your competitors offer and what your customers expect. An e-commerce site might need higher availability than a content-based website.
Beyond the Percentages: Important SLA Considerations
Scheduled vs. Unscheduled Downtime
Most SLAs exclude scheduled maintenance from availability calculations. Ensure your contract:
- Limits scheduled maintenance to specific windows that minimize business impact
- Requires your approval for maintenance activities
- Caps the total scheduled maintenance time allowed per month/quarter/year
Measurement Methodology
Understand exactly how uptime is measured:
- Measurement Frequency: More frequent checks (e.g., every minute vs. every 15 minutes) provide a more accurate picture of availability
- Measurement Points: Monitoring should occur from multiple geographic locations to identify regional issues
- Success Criteria: Define what constitutes “available” (e.g., server responds within 2 seconds with correct data)
Service Credit Structure
Effective SLAs tie downtime to meaningful compensation:
- Progressive Credits: Credits should increase as availability decreases (e.g., 10% credit at 99.8%, 25% at 99.5%, 50% at 99%)
- Calculation Base: Ensure credits are calculated against your total monthly bill, not just a portion of it
- Application Process: Credits should be applied automatically, not require complex claim procedures
- Cash Option: Credits should be convertible to cash, especially near the end of a contract
Case Study: The Cost of Inadequate SLA Terms
A major online retailer negotiated a hosting contract with a 99.9% annual uptime guarantee. In November, during the critical holiday shopping season, they experienced a 6-hour outage. While this represented significant lost revenue (estimated at $2.1 million), it constituted only 0.068% of annual time—well within the 0.1% downtime allowed by their SLA.
Had they negotiated a monthly measurement window instead, this single outage would have exceeded their monthly allowance by 2.8 hours, entitling them to significant service credits and giving the provider stronger incentives to prevent such outages.
Conclusion: Align SLA Measurement with Business Reality
When negotiating SLAs, remember that the measurement window is as important as the percentage itself. A 99.9% uptime guarantee measured monthly provides better protection than a 99.95% guarantee measured annually, as it prevents providers from “averaging out” significant outages over a longer period.
Ultimately, your SLA should reflect your business requirements, with stricter guarantees for systems that directly impact revenue or customer experience. By understanding the real-world implications of different availability percentages and measurement windows, you can negotiate contracts that provide meaningful protection against costly downtime.
This article is part of a series titled “Guide for the CTO: A compilation of articles on how to lead and manage technologies, projects and people”.