EDP renegotiation: $12.8M saved over 3 years.

The customer had signed a three-year EDP eighteen months earlier at a $14M annual commitment. By the end of year two, actual AWS consumption was tracking at $19M annually — well above the commit — driven by new lines of business migrating from a legacy data center. The original EDP discount tier had been set against the $14M commit and did not improve as actual spend grew. AWS had also capped Marketplace eligibility at 25%, which constrained the customer's ability to route third-party ISV spend through the EDP for incremental discount.

Internal procurement had attempted to open a renegotiation directly with the AWS account team three months earlier and had been told the EDP was “not eligible for mid-term repricing.” That position is technically AWS's default but is negotiable in practice when the customer can demonstrate consistent material overage and credible alternative platforms for incremental workloads.

What the customer needed

• A tier reset that priced the actual consumption profile, not the original commit
• Higher Marketplace eligibility to capture ISV spend that was bypassing the EDP
• Flex provisions for the new business lines whose growth curves were uncertain
• A renewal-cycle structure that preserved the leverage gained from mid-cycle repricing

We started with a full Private Pricing Addendum benchmark against twelve comparable EDPs in the $15M to $25M annual commit range. The benchmark established what tier discount and overlay structure the customer should be receiving at the new consumption level, and what equivalent customers had achieved on Marketplace eligibility, flex terms, and ramp protection.

Phase 1 — Position and benchmark (weeks 1-3)

We built the consumption forecast for the remaining EDP term, modeling three scenarios: organic growth only, organic plus one announced acquisition, and organic plus two acquisitions plus an AI workload tier. Each scenario produced a different optimal commit level. We then anchored the negotiation against the highest-credible scenario.

In parallel, we built a multi-cloud BATNA. The customer had no real intent to migrate, but we documented a credible Azure equivalent for the AI workloads and an on-premise option for the regulated workloads. The BATNA was never used as a threat — it was used as a forecast disclosure to AWS, framed as “here is where these workloads will land if the EDP structure does not improve.”

Phase 2 — Open the negotiation (weeks 4-6)

The opening ask was a full tier reset to the new commit level, Marketplace eligibility at 70%, flex carry-forward of 25%, and a renewal-cycle protection clause that locked the new tier as the floor for the next EDP. AWS's initial counter was a partial tier improvement, Marketplace at 40%, and no carry-forward. The gap was material but expected.

Phase 3 — Close (weeks 7-11)

We escalated past the account team to the AWS sales operations approver who actually sets PPA terms at this commit tier. The final agreement landed at the new tier reset, Marketplace eligibility at 60%, carry-forward at 20%, and a renewal floor at the new tier discount minus three percentage points. The legal redline cycle took two additional weeks; the final agreement was countersigned 11 weeks after kickoff.

The restructured EDP produced $12.8M in contract savings against the original three-year commitment, measured against the trajectory the customer was on before the renegotiation. The savings break down across four buckets.

Where the savings came from

• Tier reset on consumption — $6.2M from pricing the actual $19M run rate at the appropriate enterprise tier instead of the original $14M tier
• Marketplace eligibility uplift — $3.1M by routing $4.8M of annual ISV spend through the EDP with the new 60% eligibility cap
• PPA overlays — $2.4M from new service-specific overlays on RDS, OpenSearch, and Bedrock that had not been part of the original EDP
• Flex provision savings — $1.1M in modeled savings from the carry-forward and renewal-floor clauses, which protect the customer if growth slows in years two and three

What the customer did with the savings

Roughly half of the freed capacity was redirected into a new AI workload program on Bedrock, which the customer had previously been running as a contained pilot due to budget pressure. The remainder offset the rest of the cloud budget through the end of the fiscal year.

The renewal-floor clause is the highest-leverage item in the agreement. When the EDP comes up for renewal in eighteen months, the customer enters the conversation with the discount floor already established. AWS cannot reset back to the original tier; the floor protects the negotiation leverage gained in this cycle.