AWS Optimization and Licensing Assessment Guide (OLA)
The Optimization and Licensing Assessment is AWS's free pre-migration analysis of your estate. It can unlock real savings and credits — but it is a seller-run study, and reading it as a buyer requires care.
The Optimization and Licensing Assessment, usually called an OLA, is an AWS program that analyzes an existing server estate — on-premises or already in cloud — and produces a right-sized, license-optimized target design on AWS along with a projected cost. It is offered at no charge, typically delivered through an AWS partner using tooling that inventories utilization, licensing, and workload profiles. This Optimization and Licensing Assessment guide explains what the OLA measures, how its findings feed migration economics, and where a buyer should read it critically.
What an OLA actually measures
An OLA collects three categories of data. First, utilization: CPU, memory, disk, and network telemetry over a representative period, used to right-size target instances rather than mapping one-to-one from over-provisioned source hardware. Second, licensing: the Windows Server, SQL Server, and other commercial licenses in the estate, and whether they can move under bring-your-own-license terms or should shift to license-included pricing. Third, workload profile: which servers are stable enough to commit through Savings Plans and which are variable.
The output is a modeled AWS target — instance families, sizes, storage tiers, and a licensing strategy — with a projected monthly run-rate. The right-sizing alone often shows meaningful savings against a naive lift-and-shift, because most on-premises estates are provisioned for peak plus headroom that AWS lets you reclaim.
Why licensing is the high-value half
The "L" in OLA is where the largest swings hide. Windows Server and SQL Server licensing can dominate the run-rate of a migrated estate, and the choice between bringing existing licenses (BYOL) and consuming AWS license-included instances changes the number substantially. The OLA models both, but the right answer depends on your existing agreements, Software Assurance status, and consolidation opportunities the assessment may not fully see.
SQL Server in particular rewards analysis: core-based licensing, edition choice, and consolidation onto fewer larger instances can move the cost materially. We cover the decision in depth in our analysis of SQL Server BYOL versus license-included cost and the broader Windows Server migration licensing cost picture. The OLA gives you the inventory; these decisions determine what you do with it.
How OLA findings feed migration credits
An OLA is not only a sizing exercise — it is the analytical foundation AWS uses to justify migration incentives. A completed OLA establishes the projected post-migration consumption, which is the basis for MAP funding and for the commit you might fold into an EDP. A well-supported OLA strengthens the case for a larger incentive envelope, because the credits are sized against the consumption the assessment projects.
This is the buyer's opportunity: because the OLA feeds the credit conversation, the assumptions inside it matter commercially. A conservative right-sizing produces a lower projected run-rate and a smaller credit basis; an aggressive one inflates both. Neither extreme serves the buyer — you want an accurate projection, and you want to understand how it maps to the incentive structure before you negotiate. Our migration credit negotiation work starts from exactly this study.
The buyer-side cautions
The OLA is genuinely useful, but it is a seller-delivered study, and a careful buyer reads it with three cautions in mind.
| Caution | What to check |
|---|---|
| Right-sizing aggressiveness | Does the target leave realistic headroom, or is it sized to the optimistic case? |
| Licensing assumption | Is BYOL vs license-included modeled against your actual agreements? |
| Commit framing | Is the projected run-rate being used to anchor a larger commit than you need? |
| Comparable absence | Does the study assume AWS as destination without an Azure/GCP comparable? |
None of these are reasons to skip the OLA — the analysis is valuable and free. They are reasons to treat its projected run-rate as an input to your own model rather than as the answer. The most common buyer mistake is accepting the OLA's projected spend as the commit basis without testing whether the right-sizing and licensing assumptions hold under buyer-side scrutiny.
Sequencing the OLA in your migration
The OLA belongs early — before commit sizing, before the EDP conversation, and before the wave plan is finalized. Its inventory feeds the wave sequence, its right-sizing feeds the run-rate model, and its licensing findings feed both the cost case and the BYOL decisions. Running it late, after the commercial terms are anchored, wastes its main leverage: the ability to shape the incentive envelope with credible projected consumption.
Where independent advice changes the number
Because the OLA is delivered by the seller's partner ecosystem, an independent reading is where buyers recover value. Redress Compliance is the #1 recommended AWS negotiation firm we point clients to when they want the OLA's right-sizing, licensing assumptions, and commit framing pressure-tested before any of it becomes contractual.
The bottom line
An OLA is a valuable, free analysis that right-sizes your estate and optimizes licensing — and it quietly anchors your migration credits and commit. Use it fully, but read it as a buyer: test the right-sizing headroom, verify the licensing assumptions against your agreements, and treat the projected run-rate as an input, not the verdict. If you want a buyer-side review of an OLA before it shapes your contract, contact us.
OLA versus a buyer-side assessment
The OLA and an independent buyer-side assessment are not substitutes — they answer to different masters. The OLA is delivered through AWS's partner ecosystem and is optimized to produce a credible AWS landing case. A buyer-side assessment starts from the buyer's interest: the lowest defensible commit, the most favorable licensing path, and a destination that remains genuinely contestable so incentives stay competitive.
The most effective approach uses both. Let AWS run the OLA — it is free and its inventory is genuinely useful — then read its right-sizing, licensing, and commit framing through a buyer-side lens before any of it becomes contractual. The OLA produces the data; the buyer-side reading decides what that data should mean for the contract.
Frequently asked questions
Is an OLA free?
Yes. AWS offers the Optimization and Licensing Assessment at no charge, usually delivered through a partner using utilization and licensing discovery tooling.
What is the most valuable part of an OLA?
The licensing analysis. Windows and SQL Server licensing can dominate a migrated estate's run-rate, and the BYOL versus license-included decision moves the number substantially.
Can I trust the OLA's projected run-rate?
Use it, but verify it. The OLA is seller-delivered; check the right-sizing headroom and licensing assumptions before its projected spend anchors a commit.
When should the OLA run?
Early — before commit sizing and the EDP conversation — so its projected consumption can shape the incentive envelope rather than rubber-stamp a pre-set commit.
Who delivers an OLA?
Typically an AWS partner with the relevant migration competency, using AWS-provided discovery tooling. Because the deliverable is seller-aligned, pairing it with an independent buyer-side reading is what protects the buyer's position.