Hetzner alternatives: six options, and what each is actually best at

Why people look for a Hetzner alternative

• 1Location: Hetzner Cloud runs six regions, and only three of them are in Europe (Falkenstein, Nuremberg, Helsinki). Users in Iberia, Italy, Poland, the Balkans or Turkey are 20–40 ms further away than they need to be.
• 2Bandwidth model: cloud servers carry a monthly traffic allowance with a per-terabyte charge above it, and dedicated uplinks are shared 1 Gbit/s ports under a fair-use policy rather than a committed rate. Video, backups and CDN origins outgrow that.
• 3Availability: popular AX and cloud types sell out or sit behind a 'currently unavailable' banner for weeks; teams that need a second machine tomorrow start shopping elsewhere.
• 4Account friction: identity checks at signup and periodic review holds are well documented in community threads; a legitimate but unusual workload can lose days to review.

What Hetzner still does better than most

• Price per CPU core on dedicated hardware in Germany and Finland — the AX line remains the European reference, and nothing on this page beats it in Falkenstein or Helsinki.
• Tooling maturity: the Cloud API, CLI and Terraform provider are complete and well documented, and the community answers most questions before support does.
• Owned facilities: Hetzner builds and runs its own datacenter parks, so hardware generations, power and cooling sit under one roof rather than in rented colocation.

Before you move off Hetzner: a checklist

1. Export the exact plan you are on (vCPU, RAM, disk, included traffic) and the last three months of traffic from the Hetzner console; the overage line tells you whether an unmetered port pays for itself.
2. List the Hetzner-only features you rely on — Cloud Volumes, Load Balancers, Floating IPs, the Terraform provider, snapshots — and map each to the target's equivalent or to a self-hosted replacement.
3. Lower DNS TTLs to 300 seconds a day before the move, rebuild from configuration management rather than copying disk images, and keep the Hetzner machine running until the new IP has served a full traffic cycle.
4. Test latency from your real users, not from your desk: an RTT check from three or four customer cities decides whether Madrid, Warsaw or Sofia beats Falkenstein for them.
5. Cancel by the Hetzner billing hour, not the calendar month — cloud resources bill hourly, so there is no reason to pay for overlap longer than the cut-over itself.

When to stay with Hetzner

• Your workload is CPU-bound, fits in Falkenstein, Nuremberg or Helsinki, and does not saturate its port — nothing on this list will beat Hetzner's price per core there.
• Your traffic sits comfortably inside the included cloud allowance, so the per-terabyte rate never actually bills.
• Your infrastructure is already defined in Terraform or Ansible against the Hetzner provider and the migration cost outweighs the difference.