Backups are one of those things nobody thinks about until the day they matter more than anything else. A bad plugin update, a compromised password, a database migration that went sideways, or a client calling months later asking whether that page they deleted can come back. When that day arrives, only one question matters: how far back can we go?
Until recently, our answer was about 21 days. Today it’s roughly six months.
We’ve expanded backup retention across both our Turbo and Plaid shared hosting platforms. There’s no price increase, no add-on to purchase, and no ticket to open. If you’re on one of these plans, the longer retention is already running.
What Changed
Turbo hosting now runs three tiers of backups:
| Schedule | Frequency | Copies Retained | Coverage |
|---|---|---|---|
| Daily | Every day | 7 | Last week |
| Weekly | Every Saturday | 6 | Last 6 weeks |
| Monthly | 1st of each month | 6 | Last 6 months |
Tight day-by-day granularity for the recent past, monthly checkpoints stretching back through the last half year. Our new Basic plan runs on the Turbo platform, so it gets this exact same schedule. Entry-level pricing, same six months of coverage.
Plaid hosting trades some of that long tail for much more aggressive frequency:
| Schedule | Frequency | Copies Retained | Coverage |
|---|---|---|---|
| Every 6 hours | 4x daily | 42 | Last ~10 days |
| Weekly | Every Sunday | 24 | Last ~5.5 months |
Plaid captures your account four times a day, every day. If something breaks at 2 PM, we’re not restoring you to yesterday morning. We’re restoring you to this morning.
Every one of these backups is written off-site, to a datacenter physically distant from the servers your site runs on and still inside the United States. Today that means backups from our Denver infrastructure land in Phoenix. If a rack, a room, or a building has a very bad day, your recovery points are nowhere near it.
Why Retention Length Actually Matters
Most hosts advertise “daily backups” and stop there. That’s fine for obvious disasters, the ones you notice within hours. It’s useless for the quiet ones.
Two failure modes actually cost people their businesses:
The slow-burn compromise. A vulnerable plugin gets exploited in March. The attacker doesn’t deface anything. They quietly inject spam links or skim checkout data. You find out in May when Google flags the site. If your host keeps a week of backups, every copy you have is already infected. Six months of retention means there’s a clean version to go back to.
The delayed discovery. Someone deletes a directory of old client files, or a bulk operation nukes 400 posts nobody visits often. It’s a month before anyone notices. Seven days of backups doesn’t help. Six months does.
The gap between “we back up daily” and “we can restore you to any of the last six months” is the difference between a backup system that handles inconvenience and one that handles catastrophe.
Restore It Yourself, in Minutes
Longer retention is only half of it. The other half is getting your data back without waiting on anyone.
We run JetBackup 5, and every one of those restore points is available directly inside your cPanel. Log in, click JetBackup 5, pick a date, pick what you want back. You can restore a single file, a whole directory, one database, one email account, or the entire account including SSL certificates, DNS zones, cron jobs, and email forwarders. You can also download any backup to keep a copy of your own.
No ticket. No queue. No waiting for business hours. No fee.
That last point is not a given in this industry, which brings us to the part worth being blunt about.
What the Rest of the Industry Actually Does
We went and read our competitors’ own documentation. Not their marketing pages. Their knowledge bases and terms of service. Every claim below links to the source so you can check it yourself.
HostGator runs its automatic backup service once a week on a random day, and each run overwrites the previous one, so only one week exists at any given moment. Those backups are described as a courtesy and are not guaranteed. If you want one restored, or even want a copy of it, HostGator charges a one-time fee of $49.00. Accounts over 20 GB get no automatic backups at all. Their own documentation strongly encourages customers with critical business data to go buy a third-party backup service instead. (Backup Policy, Terms of Service)
Bluehost includes weekly backups on its entry Starter plan. Daily backups are the upsell, sold as the CodeGuard add-on. (CodeGuard)
GoDaddy sells Website Backup as a separate paid product rather than including meaningful backups with standard hosting, starting at roughly $3 per month for 5 GB of backup storage. GoDaddy doesn’t publish a plain-language backup policy page we could link, so this one comes from an independent review rather than their own documentation.
Namecheap restricts its AutoBackup tool to the Stellar Plus and Stellar Business plans, giving 6 daily, 3 weekly and 5 monthly copies. The entry-level Stellar plan doesn’t get it. If your account is suspended or terminated and you need data pulled from archive, that’s a $15.00 service fee. (AutoBackup, backup copies, archive fee)
A2 Hosting keeps off-server backups for the last 30 days, and their own knowledge base describes that as best-effort. Restore tooling is tiered by plan rather than available to everyone. (Server Rewind)
DreamHost keeps an average of two weeks of website files and five days of database backups, and states plainly that long-term backups are not included in any hosting plan. (Hosting plan FAQs)
Hostinger keeps weekly backups for six weeks and daily backups for seven days. Daily backups are only included on Business plans and above. On lower tiers, you buy them. (retention, daily backups)
InMotion charges $2.99 per month for 10 GB of Backup Manager storage, plus $1.80 per month for each additional 10 GB. (Backup Manager storage)
SiteGround is the strongest of the group and keeps up to 30 daily copies on shared plans, with free restores. Even there, downloading your own backups requires their paid Premium Backup service. (Backup service, Premium Backup)
Thirty days is the best number in that list. Ours is six months, included, with self-serve restores, on every shared plan we sell, including the cheapest one.
Competitor details above were verified against the linked sources in August 2026. Policies and pricing change. If you’re reading this later and something looks off, check the links, and let us know so we can update it.
We Still Want You to Keep Your Own Backups
Let’s be straight about something, because it would be easy to oversell this.
Like essentially every host, our Terms of Service says that backups may be unavailable and that you should maintain your own off-server copies. That language exists at every provider in the comparison above, and anyone who tells you otherwise is selling something. Hardware fails. Software has bugs. Any single system can let you down, ours included.
The difference isn’t whether a host disclaims liability. It’s what happens next.
At several of the hosts above, “backups are a courtesy and not guaranteed” is followed by a $49 invoice to hand you the backup, or a plan tier you have to upgrade into before the restore button works, or an add-on you should have bought last year. That’s where we part ways. We run these backups because recovering your data is part of what you’re paying us for. There’s no fee to restore, no fee to download your own backups, and no tier where the restore button is greyed out. You do it yourself, from your own cPanel, whenever you want.
So keep your own copies too. Not because we expect to fail you, but because every serious infrastructure plan follows some version of the 3-2-1 rule: three copies, two different media, one off-site. Six months of off-site retention makes us a strong contributor to that plan. It shouldn’t be the entire plan. Any provider can have a catastrophic day, and the people who come through intact are the ones who didn’t have a single point of failure.
One Piece of Fine Print, Stated Plainly
There is a threshold, and we’d rather tell you about it here than let you find it in a policy document.
Accounts larger than 25 GB, or that generate more than 250 GB of changes per month, may be excluded from our backup systems. Backing up an account like that is genuinely expensive in storage and I/O, and every host has a line somewhere. In practice this is exceptionally rare.
What matters is what happens when we draw that line. Our Terms of Service commits us to emailing the primary address on file before or at the time an account is excluded, identifying the account and the reason, and working with you on options to bring it back into the backup system. In writing: we will not silently exclude an account from backups.
Compare that to a host whose knowledge base states that accounts over 20 GB simply aren’t backed up. The threshold is barely different. The experience is completely different. In one case you get an email and a conversation. In the other, you find out six months later, at the worst possible moment, that nothing has been running.
No Catch
We’ve run MDDHosting independently since 2007. No private equity, no board asking why we’re spending money on storage instead of returning it to shareholders, no habit of carving basic reliability into a higher tier so we can upsell you later.
Six months of off-site retention costs us real money in storage. We think it’s worth it, because the alternative is telling a customer their oldest recovery point is from last Tuesday and their site has been compromised since spring. That’s a conversation we’d rather never have again.
Questions about any of this, or want a hand walking through a restore? [PLACEHOLDER: support/contact link.] Our support team is US-based and staffed by the people who actually run these servers.