What 99.9% Uptime Actually Means (and What Downtime Really Costs)

Every hosting company advertises an uptime number, usually with a lot of nines in it. Few explain what the number means, what falls outside it, or what downtime actually costs a small business. Here is the honest version.

The maths behind 99.9%

Uptime is the percentage of time your website is reachable. The gaps between the numbers are bigger than they look. At 99% uptime, your site can be down for more than seven hours every month and the host has still kept its promise. At 99.9% — the target we work to — the allowance shrinks to about 43 minutes a month. That difference is not academic: seven hours of downtime can swallow an entire business day; 43 minutes usually passes unnoticed at 3am.

What downtime actually costs an SME

The obvious cost is missed enquiries: the customer who googled you, hit an error page and called the next business on the list. But the quieter costs add up too. If you run Google or Facebook ads, they keep spending while your site is down — you are paying to send people to an error page. If Google's crawler hits repeated outages, your rankings can suffer, compounding the damage after the site is back. And a customer who has seen your site down once will quietly trust it less with their card details.

Why websites go down

• ●Overload — a traffic spike from a promotion or a viral post exhausts a fixed-size server
• ●Failed updates — a plugin or server patch goes wrong and takes the site with it
• ●Neighbour problems — on shared hosting, another site on the same server gets attacked or hogs resources
• ●Expired renewals — hosting or domain payments lapse quietly, and the site vanishes with them
• ●Hardware failure — the single machine your site lives on simply dies

How serverless hosting changes the picture

Most of that list has one root cause: the website lives on one fixed machine. This is why we host every site serverless on Amazon Web Services — there is no single box to overload, patch badly or lose to a hardware fault. Computing power is provided per request and scales automatically, so the promotion that would crash a shared server is just a busy afternoon. We explained the full setup in why we host on AWS.

Monitoring is the other half. A 99.9% target means little if nobody notices the 0.1%. Our care plans include uptime monitoring, so if a site becomes unreachable we know immediately — not when a customer mentions it a day later. And renewals are handled by us on a schedule, so nothing expires because an email went unread.

Questions worth asking your current host

• ●What uptime do you actually deliver, and is it monitored independently?
• ●Is my site on shared hosting, and how many other sites share the server?
• ●What happens during a traffic spike — does the site scale or fall over?
• ●Who notices first when the site goes down: you, or me?

If the answers are vague, that tells you something too. Our hosting is included in care plans from RM50 a month, with the allowances published openly on our pricing page — or send us your URL and we will give you an honest read on your current setup within one business day.