WHM runbook · Beginner · 20 minutes of setup
How to monitor website uptime — Finding Out First Is the Whole Product
A client who discovers their own outage before you do has learned something about your service that no uptime figure will unteach.
Straight answer first
Monitor every client site from outside the infrastructure it runs on, send the alert to yourself rather than to the client, and check a page that proves their business works instead of a homepage that returns 200 while the database is down. The platform commitment is a 99.9% uptime target watched by monitoring day and night; your job is knowing about the exception before the phone does.
What follows is the order to work in, the mistake that generates the ticket, and the line to leave in the handover note. Where the platform already does a step for you, the runbook says so instead of charging your afternoon for it.
Written by the Hosting Seller staff · Checked 24 August 2026
Beginner
Hand on the keyboard
Quick
Billable time, per account
5
Stages in the runbook
24/7
Cover for you and for them
A beginner-level job, 20 minutes of setup on the clock, and a client who will judge the result by what appears on their screen rather than by how it was done.
Read the failure mode before the steps. It is the part that converts a routine change into an out-of-hours callout, and it is always avoidable.
Who the alert should reach
Route alerts to you, never to the client. An alert in a client's inbox is an outage they now have to interpret, at exactly the moment you would rather be fixing it than explaining it. They hear from you, in a sentence, once you know what is happening.
That means the monitoring account is yours, in your name, with their sites as checks inside it. It is one of the few pieces of a client's setup that legitimately belongs on your side of the line.
Checking something that proves the business works
A homepage returning 200 while the checkout errors is the outage nobody catches. Point the check at the page carrying the client's revenue — the booking form, the shop, the enquiry page — and verify a keyword inside it rather than only the status code.
One meaningful check per client beats five shallow ones. At forty accounts the free tiers run out, and paying for forty considered checks is far easier to justify than paying for two hundred pointless ones.
What you say while it is still down
Have the sentence written before you need it: what is affected, what is being done, when you will next update them. Sending that within a few minutes changes a client's experience of an outage more than the length of the outage does.
You can point at a 99.9% uptime target, which allows around forty-three minutes a month, and at a pro-rated credit on any month that falls short because of a fault at our end. Say what is true, ask for the credit where it applies, and pass it on.
Reading the history rather than the incident
A single blip tells you almost nothing. Repeated micro-outages at the same hour tell you something specific, and that pattern is what you take to a support desk with a reasonable expectation of an answer.
Most monitors track certificate expiry alongside availability, which is worth switching on for every account. SSL reissues itself before it lapses on our plans, but a client with a certificate elsewhere, or a domain you do not control, will eventually prove why the check exists.
What your clients get, under your name
The panel is cPanel, which the whole trade already knows. That means the tutorial a client finds online matches the screen they are looking at, and every support answer you give is one somebody else has already written down.
Staging copies come with the plans, so a change gets tested on a copy of a client site before it is tested on their customers.
- The one panel the whole trade knows
- Staging copies before anything goes live
- WordPress Toolkit, updates applied
- Seven-day guarantee on reseller plans
Why Hosting Seller
On every plan, as standard
Costed in your time, not just described
Each stage says what it takes out of your week, so the job can be quoted before it is agreed.
Alerts land with you, not the client
The account structure is part of the runbook, because who receives the alarm decides how the outage feels to the person paying you.
Escalation that runs both ways
Support answers you at any hour and answers your clients too, which is what keeps a small operation off a permanent on-call rota.
Limits named before you hit them
Where a ceiling exists on accounts, storage or bandwidth, the runbook names it rather than letting a client find it for you.
The holding message, written in advance
What you say in the first five minutes is treated as a deliverable rather than as something to improvise at the time.
What the client sees, spelled out
Every step names the client-facing consequence, so nothing arrives in their inbox as a surprise you then have to explain.
First Steps
From choosing to live
- 1
Open the monitoring account in your own name
One account with every client site inside it as a separate check. Alerts land with you and nowhere else, which is what lets you contact the client with information rather than with a forwarded alarm.
- 2
Monitor from outside the hosting entirely
A check running on the same infrastructure as the site misses network-level failures completely. The independence of the vantage point is the only thing that qualifies it for the job.
- 3
Check the page that carries the money
Test the client's booking page, shop or enquiry form with keyword verification on the response. A 200 from a homepage while the database is down will fool a naive check every time.
- 4
Split the alerting by severity
Email for everything, and something immediate — a push or an SMS — for the accounts where minutes cost the client money. An alert read four hours later is a postmortem, not a warning.
- 5
Add certificate expiry, then read the monthly history
Turn on certificate checks for every account, then review the monthly history instead of reacting to single incidents. Patterns are what you take to a support desk; blips are not.
In the Box
Packed with every plan
- PHP versions picked per site from the panel
- Webmail plus IMAP, POP and SMTP on every mailbox you create
- WHM for you, a cPanel of their own for every client
- You set the limits every client account gets
- Setup free, with nothing due before you start selling
- NVMe SSD underneath, LiteSpeed cache in front
- Real people on the counter, for you and for them
- 300 GB, 600 GB or 1.5 TB of bandwidth by tier
- Site migration done for you by our staff, at no charge
- Staging copies, so changes get tested before clients see them
Across the Counter
Things people ask us all the time
Should the client receive the alerts too?
No. Send them to yourself and contact the client with an explanation. A raw alert gives them the anxiety without the information, and it invites a phone call during the ten minutes you most need to be looking at the server.
How many monitors do I need at forty accounts?
One meaningful check per client site, not per page. Free tiers usually stop well short of forty, so budget for it and treat it as a cost of the retainer rather than an overhead you absorb quietly.
What do I tell a client while their site is down?
What is affected, what you are doing, and when you will update them next. Three clauses, sent quickly. Silence during an outage does more damage to a hosting relationship than the outage itself.
Is there a credit when the uptime target is missed?
The target is 99.9%, which allows roughly forty-three minutes a month, and the terms provide a pro-rated credit on any month that falls short through a fault at our end — ask and it is applied. It is called a target rather than a contractual SLA on purpose, and hardware and network faults are flagged to us by the platform, usually before the first customer notices.
Read next
WordPress Hosting
WordPress kept in order for you: LiteSpeed caching, staging copies and daily backups.
Dedicated Cloud
Reserved resources with cloud flexibility, the step up past a VPS.
Moving your site to another host? Start with this checklist.
A step-by-step order of work for a move your visitors never spot: which files to copy first, how to carry email across without losing one message, the right moment to repoint DNS, and the two mistakes behind nearly every hour of downtime people ring us about.
Your brand on the counter, ours out of sight.
From your first ten accounts to five hundred, without changing platform to get there.
See WordPress Hosting plans