Cloud Repatriation
Your cloud bill grows every month and you are not sure what half of it is for. I move your systems off AWS, GCP, or Azure onto infrastructure you own: a fraction of the cost, EU data sovereignty, and no vendor lock-in, with high availability and backups you have actually restored from. I run five production systems this way, so this is not theory.
Move off AWS, GCP, or Azure to infrastructure you own and control. Hyperscaler bills scale with success, and the managed-service premium adds up fast: the same workload often costs three to five times more than owned infrastructure. For most teams the lock-in and the cost are not buying anything they actually need. The catch is that leaving safely requires knowing what you are doing, networking, storage, backups, high availability, or you trade a big bill for an outage. That safe migration is the work.
Every deliverable, upfront.
Cost and dependency audit: what you run, what it costs, what is safe to move
Migration to owned infrastructure (Hetzner, or your provider of choice)
High availability: no single point of failure for the parts that matter
Backups plus a restore you have watched work end to end
CI/CD so deploys stay boring after the move
EU data residency and sovereignty, no US processor in the chain
A written runbook so your team operates it without me
How it runs.
Days 1–3, Audit: I map what you run and what it costs, and hand you a migration plan with the projected saving. You get this even if we stop here.
Days 4–8, Migrate: move workloads to owned infrastructure with HA and backups, validated in staging before anything cuts over.
Days 8+, Cut over and hand off: a zero-downtime switch where possible, a restore drill, and a runbook your team owns.
This is for you if…
Teams whose cloud bill has quietly become a real line item
Companies that need EU data sovereignty for customers or regulation
Products paying a managed-service premium they do not use
German SMEs can often fund part of this
This kind of engagement is frequently eligible for BAFA consulting funding (up to 50%) and, in NRW, the MID programme. Public funding runs through 2026. You apply, eligibility depends on your company; I'll point you to the right programme on the call.
01Is leaving the cloud actually cheaper?
Often three to five times cheaper for the same workload, if you are not leaning heavily on managed services. But I do not ask you to take that on faith: the audit in the first three days gives you the real number for your specific setup before any migration starts.
02Isn't self-hosting risky or less reliable?
Not when it is done properly, high availability, backups, and a restore you have watched work. The real risk is a botched migration, and that is exactly what the staged, staging-validated process is built to prevent.
03Will there be downtime?
Migrations are staged and validated in a staging environment before anything cuts over, with a zero-downtime switch where the architecture allows it. Nothing moves in production until it is proven in staging first.
04What if we still need some cloud services?
Hybrid is completely fine. The goal is to stop overpaying for what you do not need, not purity. If a specific managed service earns its cost, it stays; everything else moves to infrastructure you control.
Still unsure whether this is the right shape of work?
Book a call