Six things worth checking before you commit
Check frequency you can live with
Five-minute checks mean up to five minutes of silence before anything even looks. For a marketing site that's fine. For anything taking payments, it isn't.
Alerting that stays readable
A tool that pings you every five minutes during an outage trains you to ignore it. Look for one alert on the transition and a slow nudge after.
Flat, predictable pricing
Per-seat and per-check pricing is built for teams with budgets. You want a number that doesn't move when you ship another side project.
Setup measured in seconds
If configuring the monitor takes longer than deploying the thing it watches, you'll skip it on the next project. That's how sites end up unmonitored.
Response time, not just up/down
Most outages announce themselves first as latency creep. A response-time chart turns a 3am incident into a Tuesday-afternoon fix.
Nothing you won't use
On-call rotations, escalation policies, and RUM are real features for real teams. When the team is one person, they're just configuration you maintain.
The honest shortlist
UptimeRobot is where most people start, and it gives you the most monitors per dollar — 50 free, or one-minute checks on 10–50 monitors for around $10/month. If raw count is what you need, it's the value pick and we'll say so plainly. The catch is the free tier's five-minute interval, which is the thing you'll eventually outgrow.
BetterStack is the most polished product of the group and the right answer the moment you have teammates. On-call schedules, escalation policies, and hosted status pages are genuinely excellent. They're also the features you'll never open if the escalation policy is “wake me up”.
Pingdom brings real user monitoring and scripted transaction checks, which nothing else here matches. Its entry tier is more affordable than its enterprise reputation suggests — synthetic monitoring starts at $18/mo, so the question isn't cost, it's whether you'll use the depth. Real user monitoring is priced separately on top. There's also no permanent free tier, just a 14-day trial.
PulseCheck is the narrow one on purpose: uptime, response time, and error rates, with one-minute checks from $9/mo and alerts that don't pile up. We don't do status pages, on-call rotations, or RUM — if you need those, one of the tools above is a better buy, and we'd rather say so here than after you've paid.
A reasonable default
If you're monitoring one production site and a couple of side projects, start free somewhere and put your most important URL on a one-minute check. That single decision — interval, not vendor — is what determines whether you hear about an outage from your dashboard or from a customer.
Then set alerts to a channel you actually read. Slack works well because outages land next to the work; SMS is worth it only for the one or two things where a missed hour costs real money.