WordPress Uptime Monitoring

Plugins cannot report the outages that matter

This is the decision that separates WordPress monitoring tools, and it is worth getting right before comparing feature lists. A monitoring plugin runs inside WordPress, which means it depends on exactly the stack it is supposed to be watching. When PHP runs out of memory, the database connection drops, the host suspends your account for a billing problem, or a DNS record expires, the plugin is down too. Nothing is sent, and the dashboard stays green.

External monitoring checks your site the way a visitor does, from a machine that has nothing to do with your server. It notices the failures a plugin structurally cannot: white screens of death, 500s from a bad update, expired certificates, DNS problems, and the host going away entirely. That is why the honest answer to "which WordPress uptime monitor plugin should I install" is usually that the uptime check itself belongs outside WordPress, whatever else you run inside it.

What actually matters when comparing tools

• Check interval: many free tiers check every 5 minutes. For a site taking payments, that is long enough for an outage to begin and end unrecorded.
• Retry verification: a single failed request is usually a network blip, not an outage. Tools that alert on the first failure train you to ignore alerts.
• Control over what counts as healthy: configurable expected status codes, and the ability to switch a check from HEAD to GET. WordPress hosts and security plugins fairly often mishandle HEAD requests, which shows up as a monitor that alerts on a site that is perfectly fine.
• SSL expiry warnings: silent until it is very loud.
• Alert channels you actually read: email, Slack, Discord, webhooks.
• A status page: if you run client sites, this is what stops the "is the site down?" emails during an incident.

The WordPress URLs worth monitoring

Most people add the homepage and stop, which hides a whole class of outage because the homepage is usually the most aggressively cached page on the site. A cache can happily serve a homepage for hours after the database behind it has fallen over.

• / — the baseline check
• /wp-login.php — confirms PHP and the database are alive, since the login page cannot be served from a static cache
• /wp-json/wp/v2/posts — the REST API, essential if you run a headless front end or a mobile app against WordPress
• /wp-admin — tells you your editors can still work, which is a separate outage from visitors being unable to read
• A deep content URL — a single post or product page, to catch database and template failures the homepage cache is hiding

WooCommerce needs more than a homepage check

WooCommerce stores fail in a specific way. Cart and checkout pages are excluded from full-page caching by design, so they hit PHP and the database on every request. Under a traffic spike, they are the first thing to slow down and the first thing to break, while the cached homepage and catalogue pages keep responding normally. A monitor watching only the homepage will report 100% uptime through a checkout outage.

Monitor the cart and checkout paths directly, and monitor any payment webhook endpoint your store receives callbacks on. If your store is on a busy day, the difference between finding out from a monitor and finding out from a customer is measured in lost orders.

Setting this up in PULSX

1. Add your homepage as an HTTP monitor, expecting a 200
2. Switch the check to GET if your host or a security plugin rejects HEAD requests, a common source of false alarms on WordPress
3. Add /wp-login.php to catch PHP and database failures behind the cache
4. Add cart and checkout if you run WooCommerce
5. Point alerts at email, Slack, Discord, or a webhook
6. Publish a status page if you look after sites for clients

Checks run from EU infrastructure, and SSL expiry warnings are included on every HTTPS monitor. The free tier covers 10 monitors with 5-minute checks and one status page, with no credit card. Pro adds 60-second checks and retry-verified alerts, where a failing check is retried before the monitor is confirmed down, at £9.99 per seat per month.