“Move to the cloud” sounds like one decision, but in practice it is dozens of smaller ones: one per application. A payroll system, a customer portal and a forgotten reporting server rarely deserve the same treatment. The 7 Rs give teams a shared vocabulary for those decisions.
This article uses the definitions in AWS Prescriptive Guidance, which documents each strategy with its common use cases. The ideas apply to any cloud, but AWS’s wording is the reference here.
The seven strategies
AWS Prescriptive Guidance lists seven migration strategies:
| Strategy | Also known as | In one line |
|---|---|---|
| Retire | Decommission or archive the application | |
| Retain | Keep it where it is, for now | |
| Rehost | Lift and shift | Move it to the cloud without changing the application |
| Relocate | Move servers or instances to a cloud version of the same platform, or between cloud accounts, VPCs or Regions | |
| Repurchase | Drop and shop | Replace it with a different product, often SaaS |
| Replatform | Lift, tinker and shift; lift and reshape | Move it with some optimization, such as a managed database |
| Refactor or re-architect | Redesign it to use cloud-native features |
Retire
Some applications are no longer worth running. AWS suggests retiring when there is no business value in keeping or moving an application, when you want to remove the cost of maintaining it, or when it depends on unsupported software that adds security risk. The guidance also points to usage data, such as no inbound connections for an extended period, as a signal.
Retiring is often the cheapest and safest outcome of a migration assessment, which is why discovery comes first.
Retain
Retain means keeping the application in its current environment, at least for now. AWS lists reasons such as data residency requirements, high-risk systems that need detailed planning, dependencies on other applications that must move first, recent investment in an upgrade, specialized hardware with no cloud equivalent, and waiting for a vendor’s SaaS version.
Rehost
Rehosting, or lift and shift, moves the application to the cloud without changing it. AWS notes that it lets you migrate large numbers of machines from physical, virtual or other cloud sources, and that applications are easier to optimize once they are already running in the cloud. The trade-off is that you do not get cloud optimizations on day one.
Relocate
Relocating moves a set of servers to a cloud version of the same platform, or moves instances and objects between VPCs, Regions or accounts, without buying new hardware, rewriting applications or changing operations. AWS describes it as the quickest way to migrate and operate a workload in the cloud, because it does not change the application’s overall architecture.
Repurchase
Repurchasing, or drop and shop, replaces the application with a different product, often moving from a traditional license to SaaS. AWS stresses assessing the new product against your business, security and compliance requirements before buying, then planning for training, data migration, integration with your authentication services and network configuration.
Replatform
Replatforming, or lift, tinker and shift, moves the application with some targeted optimization. AWS’s example is moving a Microsoft SQL Server database to Amazon RDS for SQL Server. Other common moves include adopting managed or serverless services, upgrading the operating system, or moving workloads into containers without code changes.
Refactor or re-architect
Refactoring redesigns the application to take full advantage of cloud-native features for agility, performance and scale. AWS lists triggers such as a monolith that slows delivery, a legacy application nobody knows how to maintain, or low test coverage that makes change risky. AWS also calls refactoring the most complex and costly of the strategies, and for large migrations recommends it only when the other strategies are not acceptable, preferring to modernize after migrating.
Choosing a strategy per application
A simple decision path helps keep assessments consistent:
- Should it exist at all? If not, retire.
- Can it move now? If not, because of hardware, regulation or dependencies, retain and set a review date.
- Is there a product that does this better? If yes, consider repurchase.
- Does it need to change to deliver business goals? If yes, and the case is strong, refactor.
- Is there a quick win from managed services? If yes, replatform.
- Otherwise, rehost or relocate, and plan modernization after the move.
Questions to ask for each application
- Who owns it, and who uses it?
- What does it depend on, and what depends on it?
- What data does it hold, and are there residency or retention rules?
- What is its current utilization?
- What would failure during migration cost?
- Is the source code available and understood?
- Is there an end-of-support date driving the timeline?
Combining strategies over time
Strategies are not one-time labels. A common path is to rehost an application first to exit a data center, then replatform its database onto a managed service, and later refactor the parts that change most often.
For that last step, incremental approaches reduce risk. The Strangler Fig pattern, described in Microsoft’s Azure Architecture Center, routes requests through a facade so that pieces of functionality move to new services one at a time while the old system keeps running. It suits large, complex systems where a single cutover would be risky.
A migration readiness checklist
- Discovery: an inventory of applications, servers, dependencies and utilization.
- Strategy per application: one of the 7 Rs, with a short written rationale.
- Landing zone: accounts, networking, identity and guardrails ready before workloads arrive.
- Security and compliance: data classification and controls agreed for each workload.
- Migration waves: applications grouped by dependency and risk.
- Cutover and rollback plans: tested, with clear go and no-go criteria.
- Operations: monitoring, backup and cost reporting in place on day one.
- Post-migration review: retire source systems and revisit candidates for replatform or refactor.
How Devacon can help
We plan and carry out cloud migrations on AWS, Azure and GCP, from discovery and per-application strategy to landing zones, migration waves and cutover. A typical engagement produces an application-by-application plan using the 7 Rs and then delivers the first migration wave. Learn more about our cloud and infrastructure services.
Sources
Facts in this article were checked against the sources above on November 17, 2026. Standards and products change, so check the source before relying on a detail. Related service: Cloud & Infrastructure.