What to monitor first
- HTTP status on your top revenue paths.
- DNS records — silent DNS changes are common during migrations.
- SSL certificate expiry — preventable, frequently forgotten.
- Blacklist / DNSBL for any IP or domain that sends email.
Uptime monitoring is the simplest insurance policy a business can buy: a quiet system that watches your sites and APIs and tells you when something changes — before customers do. This guide explains how to set it up calmly and what to measure.
Good alerts are descriptive, infrequent, and actionable. Bad alerts use red warning emoji, fire on every transient blip, and train your team to mute the channel. Aim for an alert volume small enough that every one gets read.
Treat the alert feed as a weekly habit, not a real-time stream. Most changes are benign: a registrar TTL change, a CDN rotation, a planned cert renewal. The ones that aren't will stand out clearly.
Run through these steps in order. Each step links to the right diagnostic tool.
Once a day catches outages within hours and is fine for marketing sites. Hourly is right for ecommerce, SaaS, and anything that affects revenue. Sub-minute polling is rarely useful — it produces noisy alerts.
Yes. Monitoring asks 'is it up?'. Observability asks 'why is it slow?'. Start with monitoring; reach for observability when your team is large enough to investigate.
Two consecutive failed checks before alerting cuts most false positives. For SSL, warn 14 days before expiry. For blacklists, any positive listing.
No. Route alerts to the on-call owner or a shared inbox, never to everyone. Calm alerting builds trust; noisy alerting trains people to ignore it.
Monitor it continuously
Save this check, get alerts on changes, and track multiple domains in one dashboard.