Enterprise Discount Programs are AWS's premier commercial vehicle for large-scale committed spend. They are also one of the most consistently underoptimized contracts in enterprise IT. This engagement is a case study in why.
A Fortune 100 financial institution had signed a four-year AWS EDP with $38M in committed spend. Twelve months in, they engaged us to review the agreement after a cloud cost management audit flagged that actual spend was tracking significantly below the ramp commitments in years two through four. What we found went beyond a consumption gap: the discount tiers were structured incorrectly, the commitment ramp was front-loaded in a way that benefited AWS rather than the client, and two critical service categories had been excluded from the committed spend calculation — meaning the client was consuming billable AWS services that didn't count toward their EDP threshold.
After fourteen weeks, we had renegotiated the EDP in its entirety: improved effective discount rates across compute, storage, and networking; brought the excluded service categories into the committed spend pool; restructured the annual ramp to align with the client's actual cloud migration timeline; and inserted penalty-free exit provisions if AWS failed to maintain SLA commitments. Total value recovered: $11.4M over the remaining term.
The client's EDP had been structured with a blanket 18% effective discount across all services. Based on our benchmark database of comparable financial services EDP agreements at similar spend volumes, the market rate for this commitment level was 24–27%. The three-tier discount structure AWS had proposed — which the client had accepted without independent benchmarking — was calibrated to AWS's pricing model, not to the client's negotiating position. Over four years, that gap represented $4.2M in foregone savings.
The EDP commitment pool had been structured to include EC2, S3, and RDS — standard compute, storage, and database services. But the client's workloads were increasingly running on SageMaker (ML infrastructure) and Amazon Connect (cloud contact centre), neither of which counted toward the committed spend threshold. The client was accruing $4.8M in annual AWS spend on these services at list price, outside the EDP discount umbrella, while simultaneously struggling to meet EDP consumption thresholds that triggered penalty provisions.
The EDP's annual spend ramp required the client to increase committed consumption by 22% each year — a structure that made sense for AWS's revenue recognition model but bore no relationship to the client's cloud migration roadmap. The migration was a three-year programme with significant spend acceleration in years three and four. The ramp structure created under-consumption risk in years one and two, exposing the client to shortfall payments, while years three and four would have required renegotiation anyway. We rebuilt the ramp to match the migration curve.
AWS's standard position is that Enterprise Discount Programs are fixed commercial frameworks. In our experience, they are fixed until they are not.
The leverage in this engagement came from three sources. First, the consumption gap: AWS does not want an EDP customer to fall materially short of their commitment threshold, triggering penalty provisions and potential churn. A customer tracking below commitment is a relationship risk for the AWS account team. Second, the competitive dynamic: the client had a parallel Azure relationship for Microsoft workloads and had recently completed a GCP proof-of-concept for data analytics. AWS's account team knew that workloads were portable. Third, the migration roadmap: the client could credibly commit to accelerated spend in years three and four — but only if the economics of the agreement were restructured to reflect that commitment. We built a revised commercial proposal that gave AWS what it wanted (increased committed spend in years three and four, in exchange for improved economics in years one and two and throughout) while delivering the client a materially better agreement.
AWS account teams will tell you an EDP cannot be renegotiated mid-term. This is true only when the buyer has no leverage. A consumption shortfall, a credible multi-cloud posture, and a compelling ramp commitment change the conversation.
An 18% discount across EC2 and S3 is worth less than a 22% discount that includes SageMaker, Connect, and your actual usage patterns. Benchmark the scope of the commitment, not just the headline discount rate.
Standard EDP ramp schedules are designed to maximise AWS's revenue recognition and lock in year-over-year ACV growth. Your ramp should reflect your migration roadmap — and that requires independent negotiation.
We secured penalty-free exit rights if AWS fails to maintain defined availability SLAs across the client's critical workloads. This provision alone changed the risk profile of the agreement materially.
"We had assumed the EDP was a fixed instrument — that we had negotiated it, signed it, and were committed to the terms. The team showed us within two weeks that the structure of the agreement was working against us in three separate dimensions. The final outcome exceeded everything we thought was achievable."Chief Technology Officer — Fortune 100 Financial Institution
Our vendor-agnostic guide to negotiating AWS, Azure, and GCP enterprise agreements. Covers EDP structure, discount benchmarks, commitment pool design, and SLA provisions.
We review EDP structures and identify discount, scope, and ramp optimisation opportunities — typically within the first week of analysis.
Tell us about your current AWS commitment structure and we will provide an initial assessment of your optimisation opportunity.