Ops Brief
Hosting with 24 7 support — Grade the desk on substance, at an hour that inconveniences it
At two in the morning the difference between hosts is not the marketing page — it is whether the reply engages with your stack trace or tells you the server is up.
The short answer
Test the desk before you depend on it: send a genuinely technical pre-sales question at an awkward hour and grade the answer on substance, not on speed alone. Our targets are a two-hour first reply, and one hour when a site is down, with humans on duty at every hour.
Then reduce how often you need it. The best support outcome at 3am is a problem you solve yourself in four minutes: a restore from the panel, a PHP branch switched back, a cache purged, a resource meter that tells you which site is misbehaving. A host that gives you those tools is doing more for your night than one that answers fast.
Below: what a fair test looks like, what a target is and is not, what falls inside the scope of the desk, and the specific things you should never have to open a ticket for.
By the HostingFast team · Reviewed 24 August 2026
NVMe
Storage, every tier
Free
Domain, first year
99.9%
Uptime, monitored
Flat
Renewal pricing
Every host in this market prints 24/7 somewhere. The claim is nearly free to make, which is exactly why it carries so little information on its own.
What carries information is a reply you can grade. This page is about how to obtain one before you are a customer, what to compare it against, and how to arrange your account so that fewer nights depend on it at all.
The pre-sales test, and how to grade it
Send something real and specific at an inconvenient hour. Good examples: how outbound mail is handled and what that means for SPF; whether a long single-process mysqldump will survive the process limits; what happens to OPcache after a deploy. Bad example: 'do you offer 24/7 support'.
Grade three things. First, latency — did a human respond inside the stated target. Second, substance — did the reply engage with the mechanism or restate a knowledge-base article. Third, honesty — did it say 'no' or 'not on that plan' where that was the true answer.
Remember what you are measuring. Pre-sales behaviour is the ceiling, never the floor: the desk is at its most attentive when you have not yet paid. A reply that is thin now will not improve in month nine.
A target is not an SLA, and the difference matters
Our first-reply targets are two hours, and one hour when a site is down. Those are operational targets we hold ourselves to, and we call them targets rather than contractual guarantees because that is what they are.
The contractual number is uptime: 99.9%, and if a month falls below it through a fault on our side, the terms entitle you to a pro-rated credit. Ask and it is applied. Hardware and network faults surface through platform monitoring, usually before the first customer notices.
Be suspicious of anyone printing a response-time guarantee without saying what happens when it is missed. A commitment with no remedy attached is a sentence, not a term.
Scope: the questions that should be in bounds
The practical ones. Mailbox authentication and DNS records. A certificate that has not issued. A cron that is not firing. A restore. A WordPress update that has left the site white. A PHP branch that needs switching back.
'The server is up, closing this' is the failure mode to look for. It is technically accurate and completely useless, and it is what separates renting a datacentre from hiring a host.
Where the boundary honestly sits: writing your application code, debugging a bespoke plugin line by line, or administering an unmanaged VPS you have taken root on. A desk that pretends otherwise is over-promising, and you will discover it during the incident.
The things you should never have to ask for
Restores. Daily backups on every account with self-service restore from the panel, files or databases or both, at three in the morning with no queue in front of you.
Cache and version control. Purge the LiteSpeed cache yourself, restart PHP for the account after a deploy, switch a site's PHP branch back if an update went badly. All from the panel, all reversible.
Diagnosis. The resource meter in cPanel shows CPU, memory and entry processes against your allocation, and SSH gives you the error log with `tail -f`. Knowing which site is misbehaving before you open the ticket turns a two-hour exchange into a two-line one.
Turbo at $4.66 a month carries all of that: 25 sites, 20 GB of NVMe, 1 TB of bandwidth, 50 MySQL databases, SSH with Git and Composer, renewal at the same $4.66.
Why we would rather you needed us less
A ticket costs us more than a self-service restore costs you, so the incentive is to hand you the tools rather than the queue. That is why the panel does restores, purges and version switches without a human in the loop.
We are recommending our own plan on this page and would rather say it plainly. The test protocol above works on any host you are considering, including the one you are leaving.
- Two-hour first reply target, one hour when a site is down
- 99.9% uptime with a pro-rated credit if a month misses
- Restores, purges and PHP switches without a ticket
- A pre-sales test you can run tonight
Why HostingFast
Standard on every plan
Humans at every hour
Staffed rather than queued: a two-hour first-reply target, and one hour when a site is down, at any time of day or night.
A remedy attached to the uptime number
99.9% is the target, and a month that falls short through a fault on our side entitles you to a pro-rated credit under the terms.
Scope that includes the practical jobs
Mail, DNS, certificates, cron, restores and WordPress problems — not 'the server is up, closing this'.
Self-service restores
Daily backups on every account with a panel restore that takes minutes, so 3am does not have to involve anyone else.
Diagnostics before the ticket
The cPanel resource meter and SSH log access mean you arrive with evidence rather than with a description.
Testable before you buy
Send a hard technical question at an awkward hour. The reply is the most honest specification any host publishes.
Quick Start
From order to online
- 1
Write a question that has a wrong answer
Something specific about mail, process limits or caching. A question anyone can deflect measures nothing.
- 2
Send it at an hour that inconveniences the desk
Late evening or the small hours. Business-hours performance tells you about business hours only.
- 3
Grade latency, substance and honesty
Fast and wrong is worse than slow and correct. A straight 'no, not on that plan' scores higher than a hedge.
- 4
Then set up so you need it less
Know where the restore button is, how to purge the cache and how to switch the PHP branch back. Before you need them.
Built In
Loaded onto every plan
- A staffed desk at every hour, not a form that opens at nine
- A two-hour first-reply target, one hour when a site is down
- 99.9% uptime as the target, with a pro-rated credit if a month falls short
- Daily backups with self-service restores from the panel
- LiteSpeed cache purge and PHP restart from the control panel
- Per-site PHP version switching, reversible in seconds
- Full SSH access for reading logs and reproducing faults
- Mailboxes on your own domain, included in the plan
- Zero setup charges — there is no joining fee
- A refund window of 30 days on hosting plans and 7 on reseller
- Your existing site moved across by our engineers, free of charge
Frequently Asked
What people ask us most often
What makes a fair pre-sales test of a support desk?
A question with a wrong answer, sent at an awkward hour. Ask how outbound mail and SPF are handled, or what happens to a long single-process mysqldump under the account's process limits. Then grade latency, substance and honesty. Pre-sales is the best a desk will ever treat you, so read it as a ceiling.
What are the reply targets, and are they contractual?
Two hours for a first reply, one hour when a site is down. We call those targets rather than an SLA because that is honestly what they are. The contractual figure is uptime: 99.9%, with a pro-rated credit under the terms if a month falls short through a fault on our side.
Which problems should I be able to solve at 3am without a ticket?
A restore from the daily backup, a LiteSpeed cache purge, a PHP restart after a deploy, and switching a site's PHP branch back after a bad update. All are in the panel. The resource meter and SSH log access also let you identify which site is at fault before you write to anyone.
What is genuinely outside the scope of hosting support?
Writing your application code, debugging a bespoke plugin line by line, and administering an unmanaged VPS where you hold root. Mail, DNS, certificates, cron, restores and platform-level WordPress problems are all in bounds. A desk that claims no boundary at all is over-promising.
Keep reading
Node.js App Hosting
Where the support boundary shifts, because a long-running process fails differently from a PHP request.
Hosting for Freelance Web Designers
When the desk is effectively your out-of-hours cover for client sites, scope matters more than response time.
Uptime (Glossary)
What 99.9% works out to in minutes per month, and how the pro-rated credit is calculated and claimed.
Website Builder
The drag-and-drop builder at $2.99 a month, for sites where the support questions are design rather than stack.
WordPress Hosting
Managed WordPress with Toolkit Deluxe, staging and Redis, where most out-of-hours tickets originate.
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.
Ask us something hard. Tonight.
Turbo at $4.66 a month with humans on the desk at every hour, and a panel that lets you fix most of it yourself.
View Website Builder plans