One path: prove the savings, do the move, then keep the bill lean for good. The ongoing plan is the product — the assessment and migration are how you get onto it. And I tell you, honestly, the cases where you should not move at all.
A bill you fix once drifts back up — traffic grows, a feature re-introduces egress, a commitment lapses, a managed service turns on. So I don't hand you a report and leave; I run the new setup and keep cutting the bill, every month:
You keep 80% of every dollar I save. The base is a fraction of a full-time platform engineer (~$120k+/yr loaded); the 20% share means my upside only grows when your bill shrinks. Cancel anytime — everything is standard open-source you own, no lock-in to me.
Say AWS is billing you $25,000/month and the move lands you at $9,000/month all-in (flat-rate boxes + managed services) — $16,000/month saved.
"Proven savings" is always measured against your pre-engagement baseline bill — a fixed number we agree up front — not against a moving target I get to define. If a month's savings can't be shown on paper, they aren't billed.
Send me a recent cloud bill and I'll send back a one-page savings teardown within 24 hours, free, with the egress line broken out. If going further is worth it, you'll see exactly why; if it isn't, I'll tell you that too. No signup, no sales call — a real engineer reads your bill.
I read your actual cloud bill and architecture myself — a senior infrastructure engineer, not a sales team — and hand you a decision-ready report:
Credited back: if you go on to the managed plan, this $5,000 is waived against your first months — so for clients who continue, the assessment is effectively free.
You don't have to hand the report to someone else. The same engineer who wrote it carries out the migration end to end — re-platform the application, move the database safely, stand up the new infrastructure — for a fixed price quoted before any work starts, once the numbers are clear. I spent more than 20 years as a software developer, not only a systems engineer, so the application and the data layer are mine to handle, not something I sub out.
Most "leave the cloud" pitches sell a migration no matter what. I don't. If your bill is spiky, mostly idle, or dominated by managed services rather than traffic, moving it can cost more once you count the engineering time — and the assessment exists to find that out before you spend a cent on a migration. "Zero ongoing cost" is a myth; someone always runs the servers — that's the job the managed plan prices honestly. I won't put you on a plan that doesn't save you more than it costs. Often it does. Sometimes it wouldn't, and I'll say so.