# Cloud Exit Playbook

> Reversible, staged practices for measuring and improving exit capability without beginning with a migration.

---

LLMS index: [llms.txt](/llms.txt)

---

The playbook supports an operational conclusion: **cloud exit is a staged,
reversible engineering capability, not a leap into the dark.** Begin by
measuring recovery somewhere else; decide whether to migrate only after the
constraints are visible.

## Five practices {#practices}

1. [Exit drill](/playbook/exit-drill/) — restore a complete critical dataset elsewhere and time it.
2. [Data first](/playbook/data-first/) — separate custody and recoverability from application migration.
3. [Inventory](/playbook/inventory/) — map dependencies, semantics, contracts, and owners.
4. [Staging](/playbook/staging/) — move through replay, shadow, partial traffic, and reversible cutover.
5. [Negotiation](/playbook/negotiation/) — turn measured exit options into better contract terms.

The call to action is the first drill. It is useful even when the final decision
is to stay with the current provider.

---

Section pages:

- [The exit drill](/playbook/exit-drill/): Restore a complete critical dataset somewhere else, verify it, time it, and record every dependency.
- [Data first](/playbook/data-first/): Establish independent custody, export, verification, and restore before moving application traffic.
- [Dependency inventory](/playbook/inventory/): Map data, APIs, identities, control planes, contracts, people, and recovery ownership before planning a move.
- [Staged migration](/playbook/staging/): Move through replay, shadow, partial traffic, bounded cutover, and rollback with explicit gates.
- [Negotiate with an exit option](/playbook/negotiation/): Use measured portability, cost, and recovery evidence to improve terms without bluffing.