Website Uptime, SLAs and Status Pages: What They Mean for Buyers

Every hosting company talks about uptime. It sounds like a simple measure: the percentage of time your site is online. In practice, uptime figures are only meaningful when you know how they are measured, what counts as downtime and what you receive if the promise is broken. This article explains how to read uptime claims and what actually helps when something goes wrong.

What uptime percentages mean

Uptime is expressed as a percentage over a period, usually a month or a year. Small differences in the percentage translate to very different amounts of allowed downtime. You can calculate it yourself: there are about 43,800 minutes in an average month, so a target of 99.9% allows roughly 44 minutes of downtime per month, 99.99% allows around 4 minutes, and 99% allows more than 7 hours. The numbers sound close but behave very differently if your site earns money.

Target	Approximate allowed downtime per month
99%	About 7 hours 18 minutes
99.9%	About 44 minutes
99.99%	About 4 minutes

Targets versus guarantees

An uptime target is a goal. An uptime guarantee or service level agreement (SLA) is a commitment with consequences, usually service credits. Read what counts. Common exclusions include scheduled maintenance, problems caused by your own site or account, attacks, issues outside the provider's network and force majeure events. Check whether the provider measures downtime itself and how you report an incident. Credits are generally small and rarely compensate for lost business, so the SLA is a signal of seriousness, not insurance.

What really affects your uptime

• The hosting infrastructure: hardware, network, power and the provider's operations.
• Your application: a plugin update or a code error can break your site while the server is perfectly healthy.
• Your DNS: if DNS fails or is misconfigured, nobody can find the site.
• Your certificate: an expired SSL certificate makes browsers block the page.
• Your domain registration: a lapsed domain takes everything down, including email.
• Traffic spikes and attacks: without protection, they can overwhelm resources.

A provider can only control the first item and part of the last. The rest is a shared responsibility.

What a status page is

A status page is a public page where a provider reports the current state of its services and records incidents and maintenance. A good one is hosted independently from the infrastructure it describes (otherwise it fails with it), is updated promptly during incidents, explains what happened, and keeps a history. You can judge a provider by how it communicates problems: a short, honest update with an expected resolution time is worth more than silence followed by an apology.

What a status page cannot tell you

It describes the provider's systems in general. It may not show that your specific site is down because of a local issue. That is why you should monitor your site yourself.

Monitor your own site

1. Use an external uptime monitor that checks your homepage and one critical page, such as checkout, every one to five minutes from outside your hosting provider.
2. Send alerts to more than one channel and one person, for instance email and a chat message.
3. Monitor the content, not just the response code: check that the page contains an expected phrase.
4. Monitor SSL expiry and domain expiry dates.
5. Keep a log of incidents with dates, causes and fixes.

Over a quarter, your own log tells you more about a provider than any marketing figure.

Reduce the impact of downtime

You cannot prevent all outages, but you can shorten them. Keep tested backups so you can restore quickly. Put a CDN in front of the site; static content may keep serving during a short origin problem, and it helps absorb abusive traffic (see the CDN page). Keep DNS with a reliable service and document your records. Set automatic renewals for domain and certificates. For critical systems, consider a VPS or dedicated server with a failover plan, though that adds cost and complexity.

Questions to ask before buying

• How is uptime measured, and by whom?
• Is there a public status page and where is it hosted?
• How are customers informed about maintenance?
• What is the support response process during an outage?

Why response time belongs next to uptime

A site can be technically "up" and still unusable. If pages take twenty seconds to load, visitors leave and monitoring based on a simple up/down check stays green. Track response time alongside availability and look at the trend: a slow creep upward over several weeks often warns of a growing database, a failing plugin or insufficient resources long before an outage. Treat unusually slow periods as incidents worth investigating, and add them to your log.

How we handle it at Zebra Hosting

We do not publish an invented uptime figure here, and you should be cautious about any provider that does so without explaining how it is measured. What we can say is that you can reach our support team in 17 languages when something looks wrong. See our hosting plans and ask about our maintenance and incident communication on the contact page.