The questions people actually ask before signing up. Honest answers, no marketing language.
"Unlimited" hosting is sold by overselling: hundreds of customers stuffed onto one server, with the assumption that most won't use much. When one customer gets busy, everyone else slows down. That's the reason shared hosting has a bad reputation for reliability.
We do the opposite. Each plan has clear limits, enforced in the code that runs the platform — not in fine print. You see your usage live in your portal, you know exactly what you're paying for, and your neighbour can't take down your site by getting suddenly popular.
Your plan caps the number of websites (separate domains served from your account) and mailboxes (named email accounts at your domain). Both have generous headroom — Flameling 1/5, Ember 3/20, Blaze 10/50, Inferno 20/100.
What doesn't count: subdomains within your existing website, email aliases that forward to one of your mailboxes, DNS records, customer admins on the portal, or backups you take.
See the plans page for the full breakdown.
You get a clear message — never a silent throttle, never an overage charge. Disk hits the limit? Uploads stop with an error explaining why. Mailbox cap reached? Creating a new one is rejected with a message telling you to delete one or upgrade.
The Plan Usage banner at the top of your portal shows live counts so you'll see headroom shrinking before you hit anything. Cells turn amber at 80% used.
Upgrade any time from the Hosting Plan card in your portal. We pro-rate — you only pay the difference for the remainder of your billing cycle. Renewals are billed at the published rate for your tier (subject only to the annual +2.5% cap, applied fleet-wide on 1 May), so an upgrade today doesn't quietly become more expensive than it should at renewal.
If a server is at capacity for your new tier, we may need to migrate you to another machine — that's a scheduled, planned move with you in the loop, not an automatic surprise.
Today: bring your own. If you already own a domain at GoDaddy, Namecheap, Cloudflare, Porkbun, etc., point its nameservers at ours: ember.truecorehosting.com, litespeed.truecorehosting.com, spark.truecorehosting.com. No glue records needed — our nameservers are fully-qualified hostnames. Your portal's Domains tab shows propagation status live.
Don't have a domain yet? The signup form has a one-click handoff to Porkbun — the registrar we use ourselves. You register there (typically £8–12/yr for a .co.uk or .com), come back, and complete signup with the domain you just bought. Porkbun bills you directly; we don't take a cut.
Roadmap. Native in-portal search, register, transfer, and auto-renew is on the way — we'll announce it on the updates page when it ships. Until then, the Porkbun handoff is the supported path.
Yes — but only when it's confident. flame-assist runs every ticket through automatic checks (DNS, SSL, mail records, 404s, error logs). If it finds the cause and it's a clear-cut answer (like a 404 from a deleted file, or DNS not yet propagated), it auto-resolves the ticket and emails you a [Resolved] message explaining what it found.
If it can't be sure, the ticket stays open and a real engineer picks it up. Either way, you get a fast first response — usually within seconds.
Always. Reply to the resolution email and the ticket re-opens immediately — no separate "request human" button to find. Same if you reply to a triage email or comment in the portal.
flame-assist is there to make obvious diagnostics instant, not to gatekeep. The intent is "answer the easy half before a human even sees it" — never "the bot decided your problem doesn't matter".
Yes — email ticket@truecorehosting.com from your registered contact address. We match the sender to your account and open a ticket automatically. Replies thread back into the same ticket from either email or the portal.
If you email from a different address than the one on your account, we'll send back a "we don't recognise this address" reply — log into your portal and use the form there, or email from your signup address.
flame-assist runs instantly on every ticket, regardless of plan. For tickets that need a human:
Urgent issues (site down, email not delivering) get triaged up the queue when you mark them URGENT in the subject.
Full-outage incidents are paged 24/7 — but via internal monitoring, not human eyes on a ticket queue overnight. Our watchman + sentinel pair detect site-down conditions usually before customers do, and pages route directly to the on-call engineer (currently me) with a response regardless of clock time. We don't promise overnight human review of every incoming ticket — that's the burnout-prone version of "24/7 support" we'd rather not pretend to offer. We do promise that if your site goes fully dark, someone is woken up.
Signup charges £1 today. That's not a recurring charge — it's a one-time card-validation gate, and your card has to be valid for us to start provisioning. You get full access to whatever plan you chose for 14 days.
On day 14, we charge the prorated amount for the rest of the calendar month (e.g. if your trial ends on the 29th, you'd be charged for 29th–end-of-month). From the 1st of the following month onwards, you're billed the full plan price on the 1st of each month, batched fleet-wide.
The three dates and amounts are shown to you at signup and again in the Billing tab of your portal — no surprises buried in the small print. If you want to walk away during the 14 days, you can — for the cost of the £1 (which is the performance charge for the trial service rendered, and is not refundable).
The 14-day trial is the evaluation window. We'd rather you decide during the trial — when you can walk away for £1 — than after we've provisioned your stack, allocated your isolation slot, and started billing. Refunding partial months after conversion would also mean every cancellation triggers operational work without revenue to cover it, and the "no refunds, no theatre" framing is cleaner for everyone.
If you've been billed in error — duplicate charge, billing bug, fraud — contact hello@truecorehosting.com. That's not a refund-policy issue, it's a "we messed up, fix it" issue and we'll fix it.
Stripe automatically retries failed charges over about a week. During that window your site stays live and you get emails from both Stripe and us. If retries are eventually exhausted:
A single successful payment at any point before day 51 restores everything immediately — site, DNS, the lot. We'd much rather you fix the card than lose the account.
Backups are encrypted on the server, then sent to Backblaze B2 for offsite storage. The encryption key never leaves our infrastructure — Backblaze cannot decrypt your backups even if they wanted to. Each customer's snapshots are scoped to their account; you cannot see another customer's snapshots, even via the portal API.
Restore is two-step: stage first into a separate directory so you can review, then promote into your live site. A misclick can never overwrite your live files until you confirm.
See the updates page for our backup architecture details.
Connect via SFTP and download whatever you want. Your site files, your database (via the credentials in the portal Database tab), and your customer config are all yours — there's no proprietary export format and no lock-in.
If you want a complete archive, raise a ticket and we'll generate a tarball of your files plus a SQL dump of your database, on request. You don't need to ask permission to leave; the data has always been yours.
A handful, yes — under the Add-ons tab in your portal (available on Ember plan and up; Flameling is static-only and has no Add-ons tab). Today: WordPress (with SQLite, no separate database needed), Adminer (database UI), and a Starter Template for landing pages. These are vetted, kept up to date, and run in your sandboxed environment.
What we deliberately don't do is the cPanel-style 200-script catalogue. Most of those installers are abandoned, half-maintained PHP scripts that get shipped with known vulnerabilities — we've watched too many shared-hosting customers get hit by an out-of-date plugin on a one-click app they barely remembered installing.
If there's something specific you'd like added to the catalogue, raise a ticket — we'll evaluate, package it properly, and add it for everyone.
Kernel-level — packet drops happen at the network driver via XDP (eXpress Data Path) before the attack ever reaches userspace. Most cheap hosts run a userspace firewall that's already overwhelmed by the time it tries to filter a flood; ours runs below the kernel TCP stack, so a hostile packet costs essentially zero CPU before being dropped.
What this means in practice: a DDoS aimed at another customer on the same server cannot touch your performance. Repeated probing escalates automatically — first a soft block, then a longer ban, then a permanent fleet-wide ban that propagates to all our nameservers within seconds.
Limits worth being honest about: we're not a Cloudflare-tier global scrubbing network. If you're being targeted by a sustained 100+ Gbps attack you need a CDN tier in front of us — most customers never hit anything close to that, and our XDP layer absorbs the day-to-day probing/scanning that's the actual constant on the modern internet.
Not directly — Flameling is static-only and has no PHP runtime, so WordPress can't run there at all. If you know you want WordPress now, start on Ember (£20/mo) and skip the upgrade — the Add-ons tab gives you a one-click WordPress install (using SQLite, so no separate database to manage).
If you start on Flameling for a static site and later decide you want WordPress: raise a ticket. We'll upgrade you to Ember (pro-rated for the remaining cycle, renewals at the published Ember rate subject only to the annual +2.5% cap) and walk you through installing WordPress from the Add-ons tab. The reason this isn't a one-click upgrade in the portal yet is that WordPress isn't just "add a database" — it's a runtime decision per site, and we'd rather make it deliberate than automatic.
Node.js apps live in a different operational world than what we host. They're long-running processes that need a process manager, port mapping, memory budgeting, restart-on-crash, npm dependency updates, and frequent security patching of the runtime itself — that's a different product than "host my website".
If you have a Node app, we'd recommend a platform built for it (Fly.io, Railway, Render). For things people commonly reach for Node to do — a fast personal site, a blog, a small API — you can usually do it in Go, Rust, or Python with a single binary, far less surface area to maintain. Happy to chat through alternatives via ticket if you've got something specific.
No bait-and-switch. The half-price-first-year, then 3×-on-renewal model is the single most common complaint about budget hosts and we just don't do it. There is no "introductory price".
Prices are reviewed once a year. Any increase applies to all customers from 1 May, capped at +2.5% from the previous year — not 5%, not 10%, not "whatever the market will bear". The cap is locked into our terms. We notify by email 30 days before any change takes effect.
Why a fleet-wide date instead of your signup anniversary? It's predictable, and you get one honest "here's what's happening" notice rather than scattered per-anniversary surprises. Yes, that means if you sign up in April you'll see your first increase about a month later — the trade-off is that everyone's on the same calendar, which makes the policy easier to explain and easier to hold us to. You can cancel within 30 days before or after the change with no penalty.
No catch. Flameling is genuinely a small slice of a server (4 GB disk, 512 MB RAM, 1 site, 5 mailboxes) and we cap each server at 100 Flameling tenants. The maths works out at the price we charge. Most personal sites and small projects use a fraction of even Flameling's cap, so there's plenty of room to be generous on price.
The honest limitation: Flameling is static-only — no PHP runtime, no databases, no Add-ons tab. Fine for HTML/CSS/JS sites, brochure pages, portfolios, and anything from a static-site generator. If you need WordPress, PHP, or any server-side software, Ember is the next step up at £20/month.
Didn't find your question?
Get in touch