Migrating Dedicated Servers Away From a Provider: A Practical Checklist

• Migrating dedicated servers is mostly about data, DNS, and downtime windows, not the new provider’s marketing.
• Ask any candidate provider for exact specs on hardware, network, and support before you sign anything.
• Fixed, predictable pricing matters more once you’ve been burned by surprise invoices.
• Test support response time yourself. Don’t take a number on a website at face value.
• A clean migration plan beats a rushed one every time. Build in overlap time.
• No provider is universally “easiest to migrate to”. It depends on your stack and your timeline.

If you’re planning to leave a provider like OVHcloud, Hetzner, Leaseweb, or anyone else, the pain rarely comes from the new provider itself. It comes from everything attached to your current setup: DNS records, IP allocations, firewall rules, backup schedules, licensing, monitoring integrations. The new host is often the easy part. The migration plan is the hard part.

That’s actually good news. It means you have real control over how smooth this goes, regardless of which provider you pick next.

There’s no single number that tells you how easy a migration will be. Instead, look at these factors:

• Hardware transparency. Can you see exactly what CPU, RAM, storage, and network you’re getting, or are you buying a vague tier name? Ambiguous specs make it harder to size your new server correctly the first time.
• Provisioning speed. Some providers offer instant delivery for standard configurations. Others need custom (maatwerk) builds for specific hardware or GPU requirements. Know which one you need before you compare timelines.
• Network and DDoS baseline. Check what’s included by default versus what costs extra. A server with no DDoS protection included is a different risk profile than one with protection built in from day one.
• Support model. Is support in-house or outsourced? What’s the average response time, and is it 24/7? A ticket that sits for six hours during a migration window can turn a planned maintenance task into an outage.
• Contract and billing clarity. Fixed monthly pricing means what you agree today is what’s on next month’s invoice. Variable pricing, usage surcharges, or bundled add-ons you didn’t ask for make budgeting and comparison harder.
• Compliance posture. If you handle payment data or need audited controls, ask directly about SOC 1, SOC 2 Type II, or PCI-DSS status. Don’t assume; get it in writing.

Use this checklist with any provider you’re evaluating, including your current one:

1. Exact CPU model, RAM, storage type (NVMe, SSD, HDD), and network port speed for the server you’d actually order.

2. Whether DDoS protection is included by default, and at what capacity (measured in Gbit/s).

3. Total network capacity of the provider’s backbone, not just your own port speed.

4. Average support response time, and whether that support is in-house, 24/7/365.

5. Whether pricing is fixed monthly or can change with usage, currency, or hidden fees.

6. Data center location and ownership: does the provider own the facility, or lease space from someone else?

7. Contract terms: notice period, early termination fees, data export process.

8. Relevant compliance certifications if you need them (SOC, PCI-DSS, ISO, etc.), verified with actual documentation.

9. Whether instant delivery is available for your configuration, or if it requires a custom build with a longer lead time.

10. How IP address transfers and reverse DNS changes are handled during migration.

Once you’ve picked a provider, the migration itself follows a predictable shape:

• Inventory everything. List every service, cron job, cert, and integration tied to the old server.
• Stand up in parallel. Get the new server running alongside the old one. Don’t cut over until you’ve tested.
• Lower DNS TTLs early. Do this days before cutover so changes propagate fast when you need them to.
• Migrate data in stages. Sync static data first, then handle the final delta during a short maintenance window.
• Keep the old server live for a buffer period. A week or two of overlap costs little and saves you if something was missed.
• Test failure scenarios, not just the happy path. What happens if DNS propagation is slow? If a service fails to start?

When you’re comparing options, it helps to have a concrete bar rather than vague promises. Worldstream, for example, publishes fixed monthly pricing, an average 7 minute support response time available 24/7/365, and 20 Gbit/s DDoS protection included as standard on its own European data center infrastructure. Use standards like that, whichever provider you’re actually evaluating, as a reference point for what “clear and predictable” should look like in a contract.

Don’t start with “which provider is easiest.” Start with your own inventory: what’s running, what depends on what, and what your acceptable downtime window is. Then run the checklist above against two or three candidates. The provider that answers every question directly, in writing, without hedging, is usually the one that makes the migration itself boring. Boring is what you want.