Website uptime monitoring is a small service that answers one question every few minutes: did the server respond? It is genuinely useful, it costs almost nothing, and it is also narrower than most people assume.
Understanding what it does not catch is what turns it from a reassuring green light into something that actually protects the business.
What a basic uptime check really tests
A monitor requests your homepage on a schedule and looks at the response. If the server answers, the check passes. That is the whole test.
So a site passes while the contact form silently stops sending. It passes while the checkout throws an error at the payment step. It passes while every page after the homepage returns a 404. It passes while the site is serving a security warning to visitors. In each case the server responded, so the monitor is satisfied and you hear nothing.
Uptime monitoring answers “is the machine on”. Most of what actually costs a business money happens while the machine is on.
What to monitor instead
Four things, in the order they matter for a small business site.
- A page deeper than the homepage. Point at least one check at a service page or a product page. Homepages are often cached and served from somewhere other than the application, so they can stay up while everything behind them is down.
- A keyword on the page. Most monitoring tools can check that a specific phrase is present in the response. Pick something that only appears when the page has rendered properly — a heading, a price, a button label. This is the single upgrade that catches “the server answered with a broken page”.
- The SSL certificate expiry date. A certificate that lapses produces a full-screen browser warning, which is worse than being down because visitors remember it. Most monitors will warn you weeks ahead.
- The domain expiry date. Rarer and far more damaging. Worth a calendar reminder even if your monitor does not offer it.
How often, and who gets told
One-minute checks sound better than five-minute checks and are usually worse in practice. The shorter the interval the more likely a brief network hiccup is reported as an outage, and a monitor that cries wolf gets muted within a fortnight. Five minutes is a sensible default for a business site.
More important is the confirmation step. A good monitor re-tests from a second location before alerting, so a problem at one network does not wake you up. Ask for that before you ask for frequency.
And decide who the website downtime alert goes to. An alert sent to an address nobody reads at 2am is the same as no alert. If nobody is going to act on it overnight, send it somewhere it will be seen in the morning and stop pretending otherwise.
What free monitoring gives you, and where it stops
Free tiers are genuinely worth using and it helps to know their shape. Most give you a handful of checks at five-minute intervals with email alerts, which covers a small business site properly.
What you usually give up: keyword checking, so you only learn the server answered; SMS or phone alerts, so a 2am outage waits for the morning; and history, so you cannot answer how much downtime you had last quarter. For a brochure site none of that matters much. For a store taking orders through the day, keyword checking on the checkout page is the one worth paying for.
A second thing worth knowing: monitoring from a single location reports network trouble at that location as your outage. If your alerts are frequent and never reproduce, that is usually what is happening rather than a genuinely flaky site.
Monitoring a site that takes payments
An ordinary uptime check on a store is close to useless, because the pages that matter most are the ones a monitor cannot safely load. It will not add a product, reach the payment step, or complete an order.
What helps: point a keyword check at a product page and watch for the price or the add-to-cart text, so a catalogue that fails to render is caught. Point a second check at the cart or checkout URL and watch for a phrase that only appears when it has loaded properly. And put a calendar reminder on placing a real test order yourself, monthly, including a refund.
The reason is simple. Payment gateways change, plugins update, certificates rotate. A checkout can stop accepting money while every automated check on the site stays green, and the first report usually comes from a customer who has already given up.
What to do when the alert arrives
Before assuming the worst, check whether the site is down for everyone or only for you — try it on mobile data with wi-fi off. A surprising share of alerts are network problems at one end, not site problems.
If it is genuinely down, the useful question is what changed. An update, an install, a settings change or a payment that failed at the host will explain most outages. Our guide on why WordPress websites keep breaking works through the usual causes in order.
Where monitoring fits
Monitoring is a smoke alarm. It is cheap, it is worth having, and it does not put fires out. What it buys you is finding out from a machine at 9am rather than from a customer at 4pm.
The other half is somebody acting on what it reports, which is what maintenance that includes monitoring is for. Our website maintenance checklist covers the monthly checks that catch the things a monitor never will.