UptimeRobot's free plan is now positioned for personal, non-commercial use. If you have been monitoring 30 client sites on it since 2019, you are outside the terms. Moving to a paid tier runs a few hundred dollars a year. The bigger question is what those pings were confirming: that a homepage returned a 200.
What changed with UptimeRobot's free tier?
UptimeRobot's terms of service reserve the free plan for personal, non-commercial use, and the company has been enforcing that with long-time free accounts. Free monitoring of a site you invoice a client for is commercial use by any plain reading. Check the current terms yourself before you decide anything, because these documents get revised.
Nothing technically broke, and the pings still fired. What changed is that a load-bearing piece of an agency's service delivery turned out to be sitting on a consumer freebie, which is a specific kind of uncomfortable when a client asks who is watching their site at 3am and the answer is "a free account I opened six years ago."
Free tools have a habit of doing this. The same pattern shows up with free WordPress malware scanners and their blind spots: the tool is fine at what it does, and the gap between what it does and what you told the client it does is where the trouble lives.
What does it cost to stay on UptimeRobot?
For an agency with 25-50 sites, UptimeRobot's paid tiers at list pricing sit in the $30-$70/mo range depending on monitor count, check interval, and how many people need seats. Call it $400-$840 a year. That is not a budget crisis. It is a line item that should have been in the care plan price all along.
Here is how the common replacements compare. Prices are list figures at the time of writing and change often, so confirm before you commit.
| Option | Typical monitor allowance | List price range | What you still don't get |
|---|---|---|---|
| UptimeRobot paid | 10-50+ | $7-$70/mo | Application-level checks, WordPress context |
| Better Stack | 10 free, scales up | Free-$30+/mo | WordPress-aware failure detection |
| Pingdom | Per-check pricing | $10-$50+/mo | Cheap at small scale, steep with synthetics |
| StatusCake | 10 free, more on paid | Free-$25+/mo | Server-side visibility |
| Hyperping | 10-50 | $29+/mo | Plugin, cron, and email failure detection |
| Uptime Kuma (self-hosted) | Unlimited | $6-$12/mo VPS | A monitor nobody monitors, plus your setup time |
| Monitoring included with hosting | All sites on the plan | Bundled | Depends entirely on the host |
Uptime Kuma is the one most developers reach for, and it is a good piece of software. The catch is structural. You now run a monitoring server, which means you need something watching the watcher, you own its updates and its TLS certificate, and when your VPS provider has a bad night your alerting goes down at exactly the moment you need it. We have written before about what happens when a host's support team wipes server data, and a self-hosted monitor lives on infrastructure with the same failure modes as everything else.
Why uptime pings miss the failures that cost money
An HTTP check asks one question: did this URL return a 200 in under N seconds? That question catches hard outages and almost nothing else, and hard outages are not where most client revenue leaks.
According to ITIC's 2024 Hourly Cost of Downtime survey, 90% of mid-size and large enterprises report that a single hour of downtime costs more than $300,000. Small business numbers are smaller in absolute terms but not in proportion, which we broke down in what website downtime costs per minute. The part those figures understate is partial failure, where the site is up, the monitor is green, and the money stops anyway.
Some of the failures a 200 response cannot see:
- The checkout breaks but the homepage doesn't. A payment gateway plugin update changes an API call, orders stop completing, and the store looks perfectly healthy from outside. According to the Baymard Institute, the documented average cart abandonment rate across 49 studies is 70.19% (2025), which means nobody notices a few days of zero conversions until the monthly report lands.
- Form submissions stop sending. PHP mail gets blocked, or the SMTP credentials rotate, and every lead for two weeks goes into a void. This is common enough that we wrote a whole piece on why your host's free mailboxes are not a business email plan.
- WP-Cron stalls. Scheduled posts don't publish, subscription renewals don't process, backups don't run, and nothing returns an error to the outside world.
- A redirect fires only for mobile user agents or Google referrers. Classic SEO spam behavior. Your desktop browser and your uptime monitor both see a clean page.
- Performance degrades below the alert threshold. The site never goes down, it just takes 6 seconds to load, and the monitor's timeout is set to 30. Third-party scripts are a frequent culprit, which we measured in what Microsoft Clarity costs a WordPress page.
We covered the detection side of this in depth in what uptime monitors miss on a WordPress site, and the practical setup in our guide to WordPress uptime monitoring, SLAs, and incident response. For a business owner, ping monitoring works like a smoke detector in the hallway: worth having, but it is not a fire inspection.
How to price monitoring into a care plan instead of eating it
The reflex when a free tool starts charging is to look for the next free tool. That reflex costs agencies more than the subscription does, because it treats monitoring as overhead instead of as a deliverable you are already selling.
Run the arithmetic on a 30-site book. Paid monitoring at $40/mo is $1.33 per site per month. If your care plans run $75-$150/mo, monitoring is between 1% and 2% of that revenue. The client is paying for the outcome, which is that somebody finds out before they do. Juniper Web Studio built $2,900/mo in recurring profit from 28 previously dormant client sites using that framing, and the monitoring line was never the thing that made or broke the margin.
What does break the margin is unbilled emergency work. One Saturday spent diagnosing a checkout that has been dead since Wednesday, at an agency rate of $125/hr for four hours, is $500 of labor you either eat or bill to an already unhappy client. That is more than a year of monitoring, spent in one afternoon, on a failure a ping check was structurally incapable of catching.
If you are rebuilding your service tiers anyway, it is worth reading how agencies are positioning maintenance retainers in web agency client acquisition in 2026. Monitoring sells better as evidence than as a feature: a monthly report showing what was caught and when.
What to do in the next two weeks
Start with an inventory. Open your UptimeRobot dashboard, export the monitor list, and mark which ones belong to paying clients. That list is your commercial exposure and also your working monitor count, which is usually lower than the dashboard suggests because half of them are staging URLs and a domain you sold in 2022.
Then decide which of three paths fits your business:
Pay UptimeRobot for the tier that covers your live client count, add the cost to your next care plan review, and keep the workflow you already know. This is the lowest-friction option and there is nothing wrong with it.
Self-host Uptime Kuma on a small VPS if you have someone who will patch it. Budget an hour for setup, an hour a quarter for maintenance, and an external check pointed at the Kuma instance itself so you find out when the monitor dies.
Or move monitoring to the layer that can see inside the site. A host running on the same box as WordPress sees PHP errors, failed cron events, database query time, file integrity changes, and SSL expiry across every domain on the account. An external pinger sees none of it.
Monitoring that comes with the hosting
TopSyde includes 24/7 monitoring on every plan, starting at $89/mo per site. That covers uptime, but the part that matters more is server-side: TopSyde Sentinel watches file integrity and behavior patterns for malware, and our monitoring tracks the failure types an HTTP check cannot reach, including cron health, PHP fatals, SSL expiry, and response time drift. You can see exactly what is covered on the hosting spec sheet and the SLA page.
For agencies, this removes a subscription, a login, and a renewal reminder from your operations stack. Our agency hosting program is built around that: monitoring, backups, and malware response included per site rather than assembled from six vendors with separate billing dates. Human support responds in under 2 hours during business hours, which is the other half of monitoring that free tools never provided at all.
Every plan carries a 30-day money-back guarantee, and we handle migration. If you want to see what the per-site math looks like against what you are currently paying across UptimeRobot, a backup plugin, and a security subscription, the pricing page lays it out.
Frequently Asked Questions
Can I use UptimeRobot's free plan for client websites?
UptimeRobot's terms position the free plan for personal, non-commercial use, and monitoring sites you bill a client for falls outside that. Read the current terms before you rely on any summary, including this one. If you have client monitors on a free account, move them to a paid tier or another tool.
How much does UptimeRobot cost for an agency with 30 sites?
At list pricing, agency-scale UptimeRobot tiers generally run between $30 and $70 per month depending on monitor count, check frequency, and seat count. That works out to roughly $1 to $2.30 per site per month, which most care plans absorb without a price change.
Is Uptime Kuma a good replacement for UptimeRobot?
Uptime Kuma is solid software and costs only the $6-$12/mo VPS it runs on. The tradeoff is that you become responsible for its uptime, updates, and certificates, and you should point an external check at the Kuma instance so a dead monitor does not go unnoticed for a week.
Does uptime monitoring detect a hacked WordPress site?
Rarely. Most WordPress compromises leave the site returning a normal 200 response, with injected spam links or conditional redirects that only trigger for specific user agents or referrers. Detection requires server-side file integrity checking, which is what TopSyde Sentinel handles on our platform.
What should a monitoring report to a client include?
Uptime percentage, incident count with durations, response time trend, SSL expiry dates, and any caught failures that were resolved before the client noticed. That last category is the one that justifies the retainer, and it is the one pure ping monitoring can never fill.
DevOps & Security Lead
12+ years DevOps, Linux & cloud infrastructure certified
Marcus leads infrastructure and security at TopSyde, managing the server fleet and AI monitoring systems that keep client sites fast and protected. Former sysadmin turned WordPress hosting specialist.
