Availability brief · Clinic Sites
Web hosting for medical clinics — Monitoring a cached homepage will not tell you when you are down
The check most clinics run points at a page that is served from cache, which means it keeps returning 200 long after the database that powers your booking has stopped answering.
The short answer
Point your uptime check at a URL that cannot be cached, or the check is measuring your cache rather than your site. A clinic homepage sitting in LiteSpeed will happily return a healthy 200 while PHP is failing behind it, so the first genuine outage you hear about comes from a patient rather than from your monitoring. A small uncacheable endpoint — a status page, or any route excluded from the page cache — is what an honest check looks like.
Availability and speed then become two separate measurements with two separate fixes, and neither of them depends on which tier you bought. Every plan runs NVMe behind server-level LiteSpeed, with free SSL, a migration our engineers carry out, domain mailboxes and cover at any hour of the night, from $2.42/mo on a rate that does not change at renewal.
By the HostingFast team · Reviewed 24 August 2026
Free
Migration, run by engineers
24/7
Support cover, day and night
Daily
Backup cadence
$0
Joining fee
For a private clinic the website functions as a credibility check that happens before any first appointment, so an unreachable site is read as a warning rather than as a technical problem. That raises the value of knowing about a failure quickly, and lowers the value of a monitoring setup that only proves your cache is working.
It also changes what you measure. Most performance advice is about milliseconds; the clinic version is about whether the request completed at all, whether the enquiry form delivered, and whether the person on the other end got a reply. Those are availability questions, and they need different instruments from a Lighthouse run.
The good news is that both sets of measurements are cheap. Ten minutes of setup gives you a check that can see a real failure, and a further ten gives you a dated baseline you can hold every future change against.
Monitor something the cache cannot answer
Choose a URL that is excluded from the page cache and returns quickly when the stack is healthy — a lightweight status route, or the appointment page if it is genuinely uncached. Check it every minute from outside your own network. The point is that the response has to travel the full path: web server, PHP, database, back out again.
Then decide what counts as a failure. A non-200 status obviously does. So does a 200 that takes eight seconds, and so does a page that renders but no longer contains the booking link. Content checks catch the failures that status codes miss, and on a clinic site those are the expensive ones.
Where the milliseconds actually go
A request has a fixed anatomy and each stage fails differently. DNS resolution, then the TCP and TLS handshake, then the server's own processing, then the transfer of the bytes. `curl -w` will print all of them: namelookup, connect, appconnect, starttransfer and total, which lets you say precisely which stage is expensive rather than describing the site as slow.
On a cached page the processing stage should barely register — LiteSpeed answers from a stored copy without booting PHP. On the appointment route it will not, because that page runs the full stack every time. If starttransfer is high on a page you believed was cached, you have found a caching problem, not a hardware one.
Forms, workers and the failure nobody sees
Every uncached request occupies a PHP worker for its duration. An enquiry form that sends mail synchronously holds its worker until the mail server answers, which is fine at three submissions a day and is not fine when a clinic gets written about. Sending on a queue instead returns the page immediately and does the delivery afterwards.
Test delivery as a separate thing from submission. A form that says thank you and never delivers is the worst failure on a clinic site, and it produces no error anywhere you would normally look. Send a real test to an address you control, check headers on arrival, and repeat it after any DNS or mail change. Outbound mail here goes through MailChannels, and mailboxes on your own domain come with every plan.
Keep a dated log, not an impression
Write down each measurement with its date, its tool and its conditions: median TTFB from five curl runs, the field block from PageSpeed Insights, the monitor's uptime figure for the month. It takes a minute and it converts every later conversation from opinion into comparison.
That log is also what makes our uptime target meaningful to you. We target 99.9%, monitored around the clock, and if a month falls short because of a fault on our side the terms provide a pro-rated credit — but you can only ask for that if your own record shows what you saw and when. Support answers at any hour and will compare notes against the same numbers.
What we monitor, and what you should
Our side of the line is the platform: NVMe storage, LiteSpeed caching in the server, and a 99.9% uptime target watched around the clock from our own monitoring.
Your side is the application: whether the booking route still works, whether the form still delivers, whether last week's plugin update changed either. Daily backups and staging copies exist so that checking is cheap and reversing is cheaper.
- A 99.9% target, monitored, with a credit when it is missed
- NVMe storage and LiteSpeed caching on every tier
- Daily backups restorable from the panel in minutes
- MailChannels handling outbound mail delivery
Why HostingFast
Standard on every plan
A check that can see a failure
Monitoring an uncacheable URL means the request travels the whole path. A cached homepage returning 200 proves nothing about the stack behind it.
Stage-by-stage timing
curl reports namelookup, connect, appconnect and starttransfer separately, so "the site is slow" becomes a named stage with a number attached.
Mail that is tested, not assumed
Mailboxes on your own domain with MailChannels handling outbound delivery — and a test you can repeat after every DNS change.
An uptime target with a remedy
99.9%, monitored around the clock. If a month falls short through a fault on our side, the terms provide a pro-rated credit.
Rollback that takes minutes
Daily backups restore files, databases or both from the panel, so a plugin update that broke the booking route is a short problem.
No setup charge to absorb
Nothing to pay to get started, and renewals bill at the rate you ordered at — so the plan you sized on evidence stays that price.
Quick Start
From order to online
- 1
Put a monitor on an uncacheable route
One minute interval, from outside your network, with a content check as well as a status check. This is the single highest-value ten minutes available on a clinic site.
- 2
Take a stage-by-stage baseline
curl -w against your homepage and your appointment page, five runs each, medians kept. Now you know which stage is expensive on which route, and you have a dated record of it.
- 3
Re-run both after any change
Migration, plugin update, DNS edit, theme change. Same routes, same tools, same conditions — and our engineers move the site free, so the migration itself gets measured too.
Built In
Loaded onto every plan
- Uptime monitor pointed at a URL excluded from the page cache
- Content check as well as status check, so a broken booking link registers
- curl -w baselines captured for both a cached and an uncached route
- Enquiry form delivery tested to a real mailbox, not just the thank-you page
- Synchronous mail sending moved off the request where the plugin allows it
- Mailboxes on your own domain, with outbound handled by MailChannels
- 99.9% uptime target, monitored around the clock, with a pro-rated credit if missed
- Daily backups restored at least once so the recovery path is proven
- A dated log of every measurement, its tool and its conditions
- Free engineer-run migration, with before and after readings on the same routes
Frequently Asked
What people ask us most often
What should an uptime monitor actually point at?
A route that cannot be served from cache, checked from outside your network at a one-minute interval, with a content assertion as well as a status code. A cached homepage answers 200 from a stored copy even when PHP and the database are unreachable, so it certifies your cache rather than your site. Add a content check for a string that only appears when the page rendered properly.
How do I tell DNS latency apart from server latency?
Run curl with the full timing template and read the stages separately: namelookup for DNS, connect for TCP, appconnect for the TLS handshake, starttransfer for the server's own processing. A slow namelookup is a resolver or record problem and nothing to do with hosting; a slow starttransfer on a page that should be cached is a caching problem. Naming the stage is most of the fix.
Our enquiry form is slow to submit. Where does the time go?
Usually into a synchronous send: the form waits for the mail server to accept the message before it will return the page. Where the plugin supports queuing, move delivery off the request and the response returns immediately. If it does not, the fallback is to reduce what else runs on that route — every plugin that hooks the submission is holding a PHP worker open alongside it.
What uptime do you commit to, and what happens if a month misses it?
We target 99.9%, monitored around the clock. If a month falls below that because of a fault on our side, our terms entitle you to a pro-rated credit — ask and we apply it. We call it a target rather than a contractual SLA deliberately, because that is the honest description of what a shared platform can promise, and your own monitoring record is what makes the conversation quick.
Keep reading
Moving Companies — Hosting Guide
Peak-week load, uncached routes and the numbers to take before the vans move.
Interior Designers — Hosting Guide
Colour-true photography at a weight a mobile connection will actually carry.
How to Test Your Website Speed
The measurement method in full: tools, thresholds and the order to apply them.
Web Hosting — Built for Medical Clinics
The shared platform underneath all of this — NVMe, LiteSpeed and free migration.
Dedicated Cloud
Isolated resources for when the uncached routes are genuinely your ceiling.
Changing hosts? Run through our checklist first.
A straightforward sequence for a switch your visitors never feel: which files move first, how to shift email across without losing a single message, the right moment to repoint DNS, and the two mistakes behind almost all the downtime we get asked to rescue.
Measure availability, not just speed.
A 99.9% uptime target monitored around the clock, NVMe and LiteSpeed on every tier, and daily restores you can rehearse before you need them.
View Web Hosting — Built for Medical Clinics plans