A client emails on Tuesday asking why their site was down on Saturday. You didn’t know it was. Nothing in the setup was broken, exactly. There just was never a standard for what gets monitored when a site joins the roster, so client number thirty-one got a homepage check and nothing else, and the checkout page that actually failed was never watched at all.
Start With a Standard, Not a Dashboard
Most advice for agencies starts with picking a tool that shows every client on one screen. That’s the easy part, and it’s the part that scales by itself.
What doesn’t scale by itself is consistency. When each site gets whatever monitoring the person onboarding it felt like adding, your coverage is a function of who was busy that week. Two years later nobody can tell you which client sites have certificate monitoring, and the answer is discovered the day one expires.
Write down the standard once, apply it to every site including the small ones, and revisit it yearly. Four checks per client is a defensible default.
The Four Checks Every Client Site Needs
- The homepage, from multiple regions. The baseline. It catches hosting failures, origin outages, and expired domains. One vantage point isn’t enough to trust — a check failing in a single location usually means a network problem near the checker, not a site that’s down.
- The certificate. Certificate expiry is the outage you can see coming weeks out and still manage to miss. It takes the whole site down, browsers show a warning that frightens customers, and the fix is trivial if you know in time. We wrote about why this one keeps happening in SSL certificate expiration.
- The page that makes money. The checkout, the booking form, the quote request, the donate button. A homepage check passes happily while the payment step throws a 500. Watching for specific text or elements on that page — the price, the Add to Cart button, the form itself — catches the failures that cost the client something. FlareWarden’s content monitoring does this without writing selectors by hand.
- Whatever runs on a schedule. Nightly backups, invoice generation, feed imports, cache warmers. Cron jobs fail silently by design; nobody gets an error page when a backup doesn’t run. A heartbeat check is the only way to notice absence.
Four monitors per client, multiplied out, is also how you size a plan honestly. Twenty-five clients at four checks each is 100 monitors, which is FlareWarden’s Business plan at $29/month. Fifty clients is 200, which lands on Growth at $79/month. Prices from our pricing page, checked August 2026.
Route Alerts by Who Fixes Them
The default setup sends everything to one inbox, usually the founder’s. Then the founder goes on holiday.
Route by responsibility instead. The developer who maintains a client’s application should get that client’s application alerts. The person who owns hosting relationships should get certificate and domain warnings, because they’re the one who can renew things. Role-based permissions are available on every FlareWarden plan, including free, so restricting who can change monitors without restricting who can see them costs nothing.
One thing worth deciding early: clients should almost never be on the alert list. They’ll receive a notification for a ninety-second blip, panic, and email you about it while you’re already fixing it. Give them a status page instead.
The Monthly Report Clients Actually Read
Retainer clients want evidence they’re getting value. A raw uptime percentage doesn’t provide it, because 99.94% means nothing to someone who doesn’t know what good looks like.
A report that earns the retainer covers four things:
- What the uptime figure was, and what it means in minutes. “99.94%, about 26 minutes of downtime” is legible. A percentage alone isn’t.
- Every incident, with cause and duration. Including the boring ones. “Two minutes, host network blip, resolved automatically” builds more trust than silence.
- What you did about it. The alert time, the response, the fix. This is the line that justifies the invoice.
- What you changed to prevent a repeat. Certificate auto-renewal, a caching change, a hosting upgrade recommendation.
Send it whether or not anything happened. A month with nothing to report is the product working, and saying so monthly is how a client learns to value the quiet.
Status Pages Do the Explaining for You
A per-client status page cuts the “is it just me?” emails during an incident, and it does something less obvious: it makes your monitoring visible. A client who can see their own uptime page understands what they’re paying for in a way no report achieves.
On FlareWarden these can run on the client’s own domain from the Business plan upward, so the page looks like theirs rather than a vendor’s. The Agency plan removes the page count and team seat limits entirely, which matters once you’re past fifteen clients. What to put on the page, and how honest to be during an incident, is its own discipline — status page strategy covers the version that builds trust rather than eroding it.
For the full agency setup, including white-label details and per-client onboarding, see our agencies page.
What Breaks at Fifty Sites
Two things, reliably.
The first is alert volume. Fifty sites generate real incidents plus a background hum of blips, and a team that gets paged for every one stops reading any of them. Cross-region verification before an alert fires removes most of that hum by refusing to report a failure that only one location can see. No other single setting does as much for a multi-client setup.
The second is drift. Sites get redesigned, checkout URLs change, staging domains get monitored forever after launch. Put a quarterly audit on the calendar: every monitor still points somewhere real, every client on the roster has the full four checks, every certificate check covers the domain actually in use. It takes an hour and it’s the difference between a monitoring setup and a monitoring theatre.
Key Takeaways
- Write the standard down. Consistent client website monitoring beats a clever dashboard, because coverage gaps come from ad-hoc onboarding, not bad tools.
- Four checks per client site: homepage from multiple regions, certificate, the revenue-bearing page, and anything scheduled.
- Route alerts to whoever can fix the thing and keep clients off the alert list — give them a status page instead.
- Report monthly even when nothing happened. Uptime in minutes, every incident with its cause, what you did, what you changed.
- Audit quarterly. Redesigns and domain changes silently retire monitors, and nobody notices until the check that mattered has been green for months on a URL that no longer exists.
Managing sites for more than a handful of clients? Start free with FlareWarden — 15 monitors and cross-region verification with no card, so you can put the standard on a few client sites before rolling it out to all of them.