Uptime Report and Uptime Reporting With Website Uptime and SLA Reports per Endpoint

Uptime percentage is the number of successful checks divided by the total number of checks in the period, multiplied by 100. An endpoint checked every 60 seconds gets 43,200 checks in 30 days. If 43,170 of them passed, the report shows 99.93 percent. Downtime minutes are measured separately, from the first failed check of each incident to the first successful one after it.

The two numbers answer different questions, which is why the report prints both. The percentage is what an SLA is written in. The minutes are what a client actually felt, and they are what a hosting provider asks for when you claim a service credit. A report that only shows 99.9 percent hides whether that was one 40 minute outage on a Monday morning or forty blips nobody noticed.

Check frequency sets how much you can trust either figure. At 5 minute checks an outage can be anywhere from a few seconds to almost 10 minutes long and still look the same in the data. At 60 seconds the measured duration is within a minute of the truth. If the report is going to be held against a 99.9 or 99.95 percent commitment, that resolution is the difference between evidence and an estimate, and it is the main reason to run SLA reporting on Professional rather than Starter.

Most disputes over an uptime report come down to a few minutes, so it helps to know the budget before the month starts. These are the allowances for a 30 day month and a 365 day year.

At 99.99 percent the whole monthly budget is smaller than one 5 minute check interval. A report built on 5 minute checks cannot honestly confirm or deny a number that tight. For 99.9 and 99.95 percent, one minute checks give you enough resolution to say which side of the line the month landed on. The uptime monitoring for SaaS page walks through the same math for teams that publish an SLA to their own customers.

An agency or freelancer on a maintenance retainer usually sends one page a month per site. The parts that clients read are short: the uptime percentage for the month, the number of incidents, how long each one lasted, and one line on what was done about it. The first three come straight off the report. The fourth is yours to write, and the incident log gives you the times to write it against.

Two details make that page hold up. First, measure from outside the hosting. A report produced by the same server that went down, or by a plugin running inside the site, stops recording at exactly the moment you need it. Checks here run from our monitoring nodes, so the record keeps going when the site does not. Second, keep the raw data. A client who questions 99.7 percent in March should be able to get the list of incidents that produced it, which is what the CSV export is for.

If you run many client sites, Enterprise has no endpoint limit and lets every report and export go out under your agency name. The website monitoring for agencies page covers the alerting side of the same setup, including sending each client's alerts to the right people.

It is better to know the limits before you promise a client a format. The report is opened in the panel when you want it. It is not emailed to a list on a schedule, and there is no public link a client can open without an account, so the delivery step is you sending a PDF. Each report covers one endpoint, so ten sites are ten reports rather than one combined document.

There is no setting to exclude planned maintenance from the percentage. If you pause an endpoint during a maintenance window, no checks run and nothing is counted against you for that time, but an outage that is already open when you pause is logged up to the last check. Checks come from our monitoring network rather than from a list of cities you choose, so the report does not break availability down by region. If you need scheduled branded emails, a combined multi site PDF or per region figures, a tool built around client reporting or synthetic monitoring will fit better, and the comparison of uptime monitoring tools for SLA reports names the ones that do each of those.

A report can only reach as far back as the plan keeps history. Starter keeps 30 days, Professional keeps 90 days and Enterprise keeps everything with no limit. Ask for a 90 day report on Starter and the page tells you it stopped at 30 days instead of quietly showing a shorter window as if it were a quarter.

Retention decides which reports are possible at all. A monthly client report works on any plan as long as you pull it before the month rolls off. A quarterly review needs Professional. A year on year comparison for a hosting renewal, or evidence for an annual audit, needs Enterprise. The website downtime tracker page goes deeper on the incident record itself and why the retention window matters more than most buyers expect.