Home › Uptime monitoring
Uptime monitoring: what it is and how to do it right
Uptime monitoring is how you find out a website or API is down before your customers do. A monitor checks each URL at a regular interval from outside your infrastructure, records whether it responded correctly, and alerts you the moment it doesn’t — turning “a customer emailed us that the site is broken” into “we were paged 40 seconds after it happened.”
How website uptime monitoring works
At its core, an uptime monitor does four things on a loop: send a request to your URL, evaluate the response, store the result, and alert if something’s wrong.
- Check on a schedule. Every 1–5 minutes the monitor requests your URL from a server on the public internet, just like a real visitor.
- Evaluate the response. It confirms the site returned a healthy status code (2xx/3xx), responded within a timeout, and — for deeper checks — that an expected keyword is present or an API returned the right shape.
- Record the result. Every check builds the uptime history (24-hour, 7-day, 30-day percentages) that powers status pages, SLAs and reports.
- Alert on failure. When a check fails — ideally confirmed by a quick re-check to rule out a network blip — it notifies you and opens an incident.
What to look for in an uptime monitor
Check frequency
A 5-minute interval can miss short outages; 1-minute checks catch them. Look for a re-check step before an alert fires so a single dropped packet doesn’t page you at 3am.
Endpoint and keyword checks
Real monitoring is more than “did it return 200?” Broken pages often still return 200. The monitor should let you assert on a keyword in the response or check an API endpoint with a custom method, headers and body.
Status pages and reports
A status page lets customers check for themselves instead of emailing you, and an uptime report gives stakeholders or clients the numbers. Some tools charge extra for these or reserve white-label status pages for their top tier — worth checking before you commit.
Alerting and incidents
Alerts should reach you where you already are — email, Slack, Discord — within about a minute. For teams, an incident inbox with acknowledge/assign/resolve and an on-call rotation turns a noisy alert into a handled event.
SSL and certificate expiry
An expired certificate takes a site down for every visitor with a browser warning. A good monitor warns you days before expiry.
How UptimeDesk does uptime monitoring
UptimeDesk checks your websites and public endpoints every 60 seconds, re-checks after 15 seconds to rule out blips, and alerts by email, Slack or Discord within about a minute of a confirmed outage. Each monitor can assert on a status code or a keyword, and API checks support custom method, headers and body. Every group of monitors gets a branded, optionally password-protected status page and a monthly PDF report under your own name. When something breaks, it opens an incident you can run like an on-call team — with AI triage that points at the likely failure layer (DNS, SSL, hosting or your app). There’s a free plan of 5 monitors to start, then one all-inclusive paid plan — see pricing.
UptimeDesk vs other uptime monitoring tools
Weighing options? These honest comparisons cover pricing and features against the popular tools:
- UptimeRobot alternative — Looking for an UptimeRobot alternative? UptimeDesk gives you 1-minute checks, branded status pages, monitor groups and monthly PDF reports in one simple, all-inclusive plan.
- Better Stack alternative — A simpler Better Stack alternative: uptime monitoring, branded status pages, monthly PDF reports and a shared incident inbox in one all-inclusive plan — without the observability suite.
- Pingdom alternative — A cost-effective Pingdom alternative: 1-minute uptime checks, branded status pages, monthly PDF reports and a shared incident inbox in one all-inclusive plan — plus a free plan Pingdom doesn’t offer.
- PagerDuty alternative — A lightweight PagerDuty alternative for small teams: UptimeDesk monitors your sites AND gives you a shared incident inbox with on-call escalation — one flat, all-inclusive plan, no per-seat pricing.
- Datadog alternative — A right-sized Datadog alternative when you just need uptime monitoring, status pages and reports — one flat, all-inclusive plan, no usage-based observability bill or agents to run.
Run client sites? See uptime monitoring for agencies, freelancers and marketing agencies.
Frequently asked questions
What is uptime monitoring?
Uptime monitoring is the practice of automatically checking a website or endpoint at regular intervals from outside your network, recording whether it responded correctly, and alerting you when it doesn’t. It tells you a site is down — ideally before your users do — and builds the uptime history behind SLAs and status pages.
How often should a site be checked?
Common intervals are 1 to 5 minutes. Shorter intervals catch outages faster; a 5-minute check can miss a brief blip. UptimeDesk checks every 60 seconds on Pro (5 minutes on the free plan) and re-checks after 15 seconds before declaring an outage to avoid false alarms.
What’s the difference between uptime monitoring and observability?
Uptime monitoring answers “is it up, and was it up?” from the outside. Observability platforms (logs, metrics, traces) explain what’s happening inside your systems. Many teams want the former without the cost and complexity of the latter.
Does uptime monitoring include a status page?
Not always — some tools charge extra for status pages or reserve white-label ones for higher tiers. UptimeDesk includes a branded, optionally password-protected status page per monitor group.
How do I get alerted when a site goes down?
A good monitor sends alerts within about a minute of a confirmed failure. UptimeDesk alerts by email, Slack and Discord, opens a trackable incident, and can escalate through an on-call rotation if no one acknowledges it.