---
# source: src/content/services/en/cloud-strategie.md
# route:  /en/services/cloud-strategie/
title: Cloud Strategy
group: consulting
audience: [enterprise, provider]
order: 5
tags: [cloud, strategy, migration]
vendors: [microsoft]
summary: A cloud assessment before the migration. Every application gets one of the 6 Rs, and rehost is rarely the right answer for all of them.
photoNeed: "A working session with a customer: a whiteboard with a real diagram, or two people over a laptop across a table"
stub: false
draft: false
---

## Where it starts

With an inventory. In the cloud assessment we record every application and service along
with its dependencies and connections. That service catalogue is the basis for everything
after it, and it is also the part almost nobody has in full.

The dependencies are what matter. Moving one application is easy; separating it from the
database three other systems also use is not.

## The cloud readiness assessment

We then check the applications for cloud fitness and assign each one of the **6 Rs**:

- **Rehost** – lift and shift. The virtual servers move from your own private cloud onto
  the public cloud provider's infrastructure, and the application stays as it is.
- **Replatform** – the architecture stays in principle, the foundation does not. An
  on-premises MySQL database becomes a managed database service, for example.
- **Refactor** – the application is redesigned in its essentials and, where possible, run
  cloud-native.
- **Repurchase** – the application is replaced outright by a SaaS product.
- **Retire** – the application is switched off, because the service is delivered another
  way or is no longer needed at all.
- **Retain** – nothing changes. Reasons are latency requirements, legacy applications, or
  an operating system that runs nowhere else.

Without this assignment a migration becomes a blanket move. That leads to running costs for
applications you could have switched off, and to applications that run more expensively in
the cloud than they did before.

**Retire** is the underrated category. Every inventory turns up systems nobody needs any
more that would otherwise have been migrated.

## What comes out of it

A future mode of operation per application, and therefore a migration you can plan and
budget rather than start and see how far you get.
