Website uptime is the time that a website or web service is available to the users over a given period. Represented as a ratio of the time available divided by the total time, providers calculate the ratio in monthly or yearly increments. Although 100% uptime is the goal, the industry considers 99.999% uptime as high availability.

Every website experiences downtime planned or otherwise. Every website provider wants to keep uptime as high as possible, and rightfully so with the competitive nature of the Internet. Knowing that some downtime is expected, most brands try to meet a goal of 99.999% uptime. You may see this goal referred to as “five-nines availability” or “high availability.”

Calculating the uptime percentage

At its most basic, a system's uptime is simply the total time minus the downtime, but most monitoring services provide this information as a percentage. The formula for calculating the uptime percentage is the simple ratio of uptime divided by the total time.

uptime ÷ total time = availability percent

Using the above formula, a 99.999% uptime only allows for 5.25 minutes of downtime per year. Unless the website or service is mission critical such as hospital systems, many providers consider a lower availability of 99.99 (52 minutes and 36 seconds of downtime per year) or 99.98 (1 hour, 10 minutes and 7 seconds of downtime per year) good uptime ratios. This calculation provides the actual availability for a system and doesn’t consider any other factors such as performance and function.

Uptime caveats

As noted above, the basic formula doesn’t take into account any other availability considerations. On the simplest level, uptime only refers to the fact that you can access a website or service. This basic definition leaves a lot of room for interpretation. Some people consider the term too ambiguous and that some providers abuse the definition by excluding usability and performance from the equation.

Limited availability

On Black Friday 2015, a popular brand’s website experienced extremely high volume bringing it down, so in 2016 the site prepared for the additional load on their servers. While the site never went down, once the servers reached a defined level of capacity, new visitors were detoured to an apology page that informed the users that the site was experiencing heavier traffic than usual and they will need to try visiting the site at another time. Although the site didn’t crash under the volume, the actual availability was limited. Because the site remained up and available as capacity allowed, the brand considered the time as uptime. Many visitors disagreed.

Reduced performance

Some still consider a site to be up even though the site suffers from poor performance. As long as the page doesn’t time out or return page errors, some may still consider it up although usability is diminished. Is a page or service that takes an exorbitant amount of time to perform its function still considered up? Under the most basic definition of uptime, yes, a poor performing site is still available.

Missing functionality

Some may consider a page up even though it can’t perform any or all of its intended functions. With the distributed nature of websites and web services, many things can go wrong within a page independent of the pages uptime. However, when does someone consider a site or service down when functionality is impaired? For example, an e-commerce site may still be up, but due to a simple scripting error, the submit payment button may not become enabled after the user selects a shipping method. Although the site remains “up,” the site’s visitors can’t complete the checkout process. Is this site still available?

Planned maintenance

Some brands do not consider planned maintenance as downtime. Although the site or service is not available during the scheduled outage, the provider reduces the total time before calculating the uptime ratio. Excluding planned maintenance is a common practice, so it is advised that a consumer should ask how often and when the site experiences scheduled maintenance.

Uptime and Service Level Agreements

Many brands work at maintaining high availability to protect their reputation and revenue streams while others guarantee their uptime with a service level agreement (SLA). An SLA is a contract that says that the company guarantees a certain level of availability. Hosting companies and SaaS providers often provide an SLA to their users, and if they fail to meet their obligations many credit their users a portion of their fees to make up for the failure.

For the most part, end users rely on the service provider to tell them if the service met the uptime requirements. Service Providers also use external uptime monitoring tools to track their uptime and to provide proof that they met their obligations to their end users. In return, some end users or organizations obtain their own third-party monitoring to verify a provider's uptime.

Conclusion

Uptime is a subjective term. With improvements in technology and Synthetic Monitoring, service providers can meet and exceed their uptime goals. End users do need to understand the definitions of what constitutes uptime before agreeing to a service or provider. If a service or provider is under consideration, end users can use third party  Website Monitoring tools to track the availability of the product before making a buying decision.