The Three-Wave Pattern
AWS migration programs for mid-market to enterprise organizations succeed more reliably when structured as three sequential waves than as a single big-bang program. The waves have different goals, different teams, different timelines, and different success criteria. Treating them as one program produces the timeline overruns and cost surprises that AWS migrations are notorious for.
Forrester's 2024 cloud migration research tracked the success rates of different program structures and found wave-based programs delivered against expectations roughly 30 percent more often than single-program approaches (Forrester, "Cloud Migration Trends 2024"). The difference is not because the waves are smaller; it is because the waves allow learning between them.
Real Estate Marketing Attribution
A single attribution mistake led to a 22% pipeline drop. Here’s how real estate teams fix it with full-funnel visibility.
If you are scoping an AWS migration for an organization above the small-business scale, the pragmatic playbook is three waves over 24-36 months. Each wave delivers value independently and informs the next.
Wave One: Foundation and Quick Wins
The first wave runs 4-8 months and establishes the foundation for everything that follows. The wave's goals are technical foundation, operational practice, and demonstrated capability.
The foundation work includes AWS account structure (multi-account, often using AWS Control Tower), security baseline (IAM, encryption, network architecture, audit logging), and FinOps framework (tagging, cost monitoring, allocation). Without this foundation, subsequent waves accumulate technical debt that gets expensive to remediate.
The quick wins involve migrating workloads that benefit from cloud immediately and have low migration complexity. Backup and disaster recovery infrastructure that benefits from cloud's geographic redundancy. Development environments that benefit from cloud's flexibility. Non-critical applications scheduled for refresh anyway. The quick wins demonstrate that the migration can deliver value and build organizational confidence.
The operational practice includes establishing cloud operations capability: monitoring, incident response, change management adapted for cloud, vendor management with AWS. The practice cannot wait for production workloads to mature; it has to be operational when production workloads arrive.
Wave one's success criteria are: foundation in place and audited, 10-25 percent of workloads migrated, operational practice running with measurable performance, executive confidence sufficient to fund wave two.
The wave fails when foundation work is skipped to accelerate migration. The skipped foundation work returns as remediation under time pressure during later waves.
Wave Two: Strategic Migration
The second wave runs 8-14 months and handles the bulk of strategic migration. The wave's goals are migrating the majority of remaining workloads using the foundation and patterns established in wave one.
Workload selection follows the Six R framework (rehost, replatform, refactor, repurchase, retire, retain) with honest assignment. Each workload moves through assessment, decision, execution, and transition phases on a timeline appropriate to its complexity. Multiple workload streams run concurrently.
The wave includes complex workloads that wave one deferred. Production applications, mission-critical systems, regulatory-sensitive workloads. These require more careful planning than wave one's quick wins. The wave one foundation supports them.
Application modernization happens during wave two for workloads where modernization is justified. Containerization, serverless adoption, managed service migration. The modernization work is workload-specific; not every workload modernizes during migration.
Wave two's success criteria are: 60-85 percent of workloads migrated, operational practice mature enough to handle production load, cost trajectory aligned with business case projections, capability to absorb remaining workloads in wave three.
The wave fails when workload selection ignores the Six R framework and defaults to rehost for everything. Rehost-only migrations consistently deliver less value than the business case promised.
Wave Three: Optimization and Cloud-Native
The third wave runs 8-12 months and completes the migration with optimization, decommissioning of legacy infrastructure, and cloud-native adoption for strategic workloads.
Optimization work spans the migrated estate. Cost optimization through right-sizing, reserved capacity, and architectural refinement. Performance optimization for workloads that need it. Reliability hardening for workloads where production has revealed gaps. Security improvements based on accumulated learnings.
Decommissioning of legacy on-premises infrastructure happens during wave three. The cost savings from decommissioning typically exceed the cloud's incremental cost, which is part of how migration ROI gets realized. Workloads that remain on-premises after wave three are usually deliberate retains rather than failed migrations.
Cloud-native adoption for strategic workloads continues throughout the wave. Workloads identified during waves one and two as candidates for deeper modernization get the engineering investment they justify. This is where AI integration, advanced analytics, and modern architecture often happen.
Wave three's success criteria are: legacy infrastructure decommissioned, cost savings realized at the level the business case promised, cloud-native capability established for strategic workloads, organization operating in cloud as the default.
The wave fails when optimization gets deferred indefinitely. Optimization deferred is optimization lost; the cost savings depend on doing the work.
What Changes in 2026 From Earlier Migrations
Three things have shifted the AWS migration landscape from migrations that started in 2018-2022.
AWS service capabilities are broader. The service catalog has matured significantly. Workloads that required complex custom architecture in 2020 often run on managed services in 2026. The reference architecture shifts toward managed services over time.
The talent market has changed. AWS-certified engineers are more available than in 2020, but the senior architecture talent that determines migration success remains scarce. The hiring strategy has to account for this.
AI workloads have entered the migration scope. Migrations planned in 2020 did not need to account for AI workload patterns. Migrations planned in 2026 do. The architecture choices affect later AI deployment.
These shifts make 2026 migrations easier in some ways (more managed services, more documentation, more reference patterns) and harder in others (AI workload complexity, more sophisticated business expectations).
The Cost Trajectory That Works
Successful migrations show a recognizable cost trajectory across the three waves.
Wave one usually produces minimal net cost savings. The cost of cloud workloads roughly equals the cost of the legacy workloads they replace at this stage because the migrated workloads have not been optimized yet. The wave's value is operational rather than financial.
Wave two produces meaningful cost variability. Migrated workloads start optimized; previously-migrated workloads from wave one get retroactive optimization; legacy infrastructure cost has not yet declined. The aggregate cost trajectory is roughly flat or slightly increasing as cloud cost grows faster than legacy cost declines.
Wave three produces the cost savings the business case promised. Legacy decommissioning accelerates as workloads migrate. Optimization compounds across the migrated estate. The cost trajectory finally turns clearly downward.
Programs that expect cost savings starting in wave one typically experience disappointment that triggers premature program changes. Programs that expect the trajectory described above stay the course and realize the savings.
Real Estate Identity Resolution
Duplicate records are hiding your best leads. Identity resolution reveals true buyer intent and fixes your pipeline.
Call to Action
What Logiciel Does Here
Logiciel works with mid-market to enterprise IT leadership scoping or executing AWS migrations. The work is typically structured around the three-wave playbook with adaptations for the specific organization's starting state, regulatory environment, and workload portfolio.
The Cloud Migration Patterns That Deliver ROI framework covers the Six R workload selection. The Cloud Infrastructure Modernization Loops framework covers the iterative pattern that emerges from the three-wave structure.
A 30-minute working session is enough to assess your candidate migration against the three-wave playbook.
Frequently Asked Questions
Can I compress the timeline below 24 months?
For small portfolios or organizations with existing cloud expertise, yes. For mid-market to enterprise with significant workload portfolios, below 24 months is usually rushed and produces quality tradeoffs. The 24-36 month envelope reflects realistic migration of meaningful portfolios.
What if my organization has not started migration yet in 2026?
The playbook works fresh. The shifts in AWS capability since 2020 actually make starting in 2026 in some ways easier than starting in 2020. The patterns and reference architectures are more mature; the experience base is broader.
How does this compare to multi-cloud migration?
The three-wave structure applies. Multi-cloud adds complexity that affects wave two specifically (workload-to-cloud assignment) but the wave structure still works. Pure single-cloud migrations are simpler; multi-cloud migrations need more deliberate wave planning.
Should I use AWS Professional Services or a third-party partner?
Often both. AWS Professional Services brings AWS-specific expertise; third-party partners bring industry and operational expertise. For complex enterprise migrations, the combination usually outperforms either alone.
How does AI workload migration fit into the three waves?
AI workloads typically enter during wave two for early adoption and during wave three for production scaling. The wave one foundation has to support AI workloads (Bedrock access, GPU capacity planning, data architecture). The architecture decisions affect AI deployment downstream. Sources: - Forrester, "Cloud Migration Trends 2024" - AWS Cloud Adoption Framework, 2024 - McKinsey, "Cloud's trillion-dollar prize is up for grabs," 2024
