Four ways to find out what you are actually sitting on.
All four are read-only, run against your own data, and end with a model you keep whatever you decide to do next. Three of them are funded by AWS.
Same method, different question.
Every assessment we run follows the same discipline: scope it honestly, measure rather than assume, price every option side by side, and publish the assumptions underneath.
Optimization and Licensing Assessment
The full estate assessment, and the one most engagements start with.
Discovery across on-premises and cloud, utilisation measured over time, right-sizing against real load, a complete AWS cost model with purchasing options priced against one another, and a Microsoft, Oracle and VMware entitlement analysis layered over the top.
Read the full page →Detached Storage Assessment
The line item that quietly becomes the business case once compute is right-sized.
SAN, NAS, file servers and object storage mapped to S3, EBS, EFS and FSx, with a provisioned-against-used reclamation target and a cost comparison against the current environment. AWS puts storage at up to 45% of total migration opportunity, and on one of our own engagements it finished as 57% of the recommended annual cost.
Talk to us about storage →Cloud-to-cloud Assessment
Azure to AWS, mapped construct by construct rather than machine by machine.
A cloud-native estate has no conventional server inventory to collect. Managed database instances, app service plans, container node pools and PaaS databases each need an AWS target chosen and an assumption stated. We publish every one of those mappings, including the ones we price conservatively because the equivalence is imperfect.
See two engagements →GenAI Readiness Assessment
Whether your data is in a state where generative AI is worth attempting.
Where your data actually sits, what condition it is in, what it would take to make it usable, and what the resulting workload would cost to run on AWS. The same discipline as the other three: measure first, price honestly, and say plainly where the answer is not yet.
Talk to us →Why AWS pays for this, and what the catch is.
AWS funds these assessments because a migration built on a proper model is more likely to happen, more likely to land on the right services, and more likely to stay. A badly modelled migration that gets reversed costs AWS more than the assessment did.
The one thing worth saying plainly: an assessment needs an estate large enough to be worth modelling. If you are below that line we will tell you at the scoping call rather than after the collection. The exception is GenAI readiness, which is newer. It is scoped and quoted before anything starts, and we tell you at that point whether funding applies.
Not sure which one you need?
Most conversations start with an OLA and narrow from there. A thirty-minute scoping call is usually enough to work out which assessment answers the question you actually have.