AWS Practice
AWS Contract Negotiation. A US Bank Case Study.
A regional bank faced an AWS EDP renewal sized to optimistic growth. Here is how the commit, the Savings Plans mix, and the egress terms were reset on the buyer side.
A regional bank faced an AWS EDP renewal sized to optimistic growth. Here is how the commit, the Savings Plans mix, and the egress terms were reset on the buyer side.
A midwestern bank was about to renew an AWS Enterprise Discount Program commit built on growth that never arrived, and the commit, not the rate, was the real risk.
The bank faced an EDP renewal built on a growth forecast its real usage never matched. The prior commit assumed an expansion that supervisory caution and a slower roadmap had quietly cancelled.
The account team framed the renewal as a rate conversation. The real exposure was the commitment size, because an unmet EDP commit is paid whether or not the workloads arrive.
AWS describes the Enterprise Discount Program through its cloud economics resources, and the underlying commitment mechanics sit alongside its Savings Plans documentation.
The commit was reset to trailing actual consumption plus a defensible growth band, not to the optimistic forecast. That single change removed the largest source of waste in the deal.
We modeled three years of real usage and built the commit from the floor of stable workloads, leaving variable demand on flexible pricing rather than locking it in.
How the AWS deal was restructured
| Element | Before | After |
|---|---|---|
| EDP commit | Sized to growth forecast | Sized to trailing actuals plus band |
| Savings Plans | Stacked, no target | Coverage target on steady workloads |
| Egress | Unmodeled | Negotiated and modeled in the commit |
| Exit terms | Absent | Portability and notice terms added |
The levers that moved the deal were commit right sizing, a coverage target, egress negotiation, and exit terms, in that order of impact. Rate alone was the smallest lever.
The bank also used its regulated status deliberately. Supervisory expectations on outsourcing and concentration risk made portability and exit terms a legitimate, non price demand AWS had to meet.
The standard advice is that a bigger EDP commitment earns a bigger discount, so you should commit aggressively to maximize the rate. We disagree. In roughly 7 out of 10 EDP renewals we have benchmarked, the larger commit destroyed value, because the customer paid against a floor it never reached and the extra discount never offset the unmet commitment. The headline rate looked better while the effective cost rose. The buyer side move is to size the commit to trailing consumption plus a defensible band, push variable demand onto flexible pricing, and treat egress and exit terms as first class negotiation items, not afterthoughts.
Source: Redress Compliance advisory engagement file, 2024 to 2025.
On an AWS renewal the discount you celebrate is meaningless if the commitment behind it is larger than the workloads you will ever run.
The renewal reset the commit to real consumption, fixed the coverage mix, and added egress and exit terms the prior deal lacked. The effective unit cost fell even though the headline discount changed little.
More importantly for a regulated institution, the contract now carried portability and notice terms that satisfied supervisory expectations on concentration risk.
An EDP commit is a multi year spend commitment to AWS in exchange for a discount. It is a floor you pay against, so the commitment size, not just the discount rate, decides whether the deal saves money.
An unmet EDP commitment is paid whether or not the forecast workloads arrive. Oversizing the commit to optimistic growth costs more than a smaller discount on a right sized commitment.
Size the commit to trailing actual consumption plus a defensible growth band, built from the floor of stable workloads. Leave variable demand on flexible pricing rather than locking it into the commitment.
Cover steady baseline workloads with Savings Plans for the deepest rate, leave variable demand on demand, and review the mix quarterly. Stacking plans without a target leaves both unused commitment and on demand leakage.
Yes. Data transfer egress is one of the least negotiated and most negotiable AWS line items, often 5 to 12 percent of the bill. It should be modeled and brought into the negotiation explicitly.
Regulated institutions face supervisory expectations on outsourcing and concentration risk. Portability, notice, and exit terms are legitimate non price demands that carry as much weight as the discount.
Eligible AWS Marketplace spend can be routed to count toward an EDP commitment. Directing qualifying purchases through Marketplace helps reach the commit without inflating it.
The AWS quarter end concession window is the strongest timing lever. Aligning the close to it, with three years of consumption evidence in hand, improves both the commit terms and the rate.
No. The effective cost fell because the commit was right sized and the coverage mix was fixed, not because the headline discount grew. A larger commit for a larger discount would have raised the effective cost.
How we size an EDP commit, structure the Savings Plans mix, and negotiate egress and support terms on the buyer side of an AWS renewal.
Used across more than five hundred enterprise engagements. Independent. Buyer side. Built for procurement leaders running the next renewal cycle.
