AWS EDP Negotiation: The Complete Enterprise Guide

The Enterprise Discount Programme (EDP) is AWS's headline private pricing vehicle for committed-spend customers. It is also one of the most opaque commercial instruments in enterprise IT — there is no public price list, no standard discount grid, and no centralized governance that prevents two similar customers from receiving very different outcomes. The discount you sign at $5M, $25M or $100M annual commitment is almost entirely a function of the process you ran to get there, not of the line items on the order form.

This guide captures the full enterprise-side EDP negotiation process used on engagements totalling $2.4B+ in AWS spend, with 38% average reduction and $340M+ in documented client savings. It is intentionally long because the EDP is intentionally complicated; the buyers who shorten the process pay for the shortcut over the full term of the agreement.

What an EDP actually is

An EDP is a multi-year private pricing agreement in which the customer commits to a minimum annual spend with AWS in exchange for an in-arrears discount applied across most (not all) AWS services. Standard agreement length is three years; one-year EDPs exist but yield meaningfully smaller discounts, and five-year EDPs exist for very large customers with stable spend profiles.

Mechanically the EDP is a percentage discount applied to "EDP-eligible" usage at billing time, calculated against committed spend. Some services are excluded — Marketplace pass-through, AWS Support, certain managed third-party offerings, and a small list of regional or specialty services. The eligibility list moves between EDP generations; the current version is published in the AWS Customer Agreement appendix at signature.

Key elements every EDP includes, in commercial order of magnitude:

• Commitment level. The annual minimum spend the customer commits to, typically expressed in USD with annual ramp.
• Discount percentage. The headline EDP commercial discount applied to eligible spend.
• Private pricing addenda. Service-specific discounts on top of EDP — CloudFront, Bedrock, MediaConvert, Direct Connect, SageMaker, etc.
• Credit and incentive programs. MAP, MAP for Windows, training credits, POC funding, partner credits.
• Flex provisions. Annual true-down rights, growth ramp, shortfall protection, change-of-control provisions.
• Support and TAM allocation. Enterprise support tier, TAM dedication, escalation rights.

Buyers who treat the EDP as "the commercial discount number" leave the majority of the available value on the table, because the discount is the single most visible but rarely the single largest lever. See EDP discount tiers benchmarked for the empirical distribution across spend bands.

The eight-layer EDP pricing stack

An EDP is best understood as a stack of eight layered price levers, each negotiated separately and combined into the final commercial outcome. Across hundreds of engagements, the dollar weighting is roughly stable:

1. EDP commercial discount — 25–40% of total negotiated value
2. Private pricing addenda — 15–35%
3. Migration and modernization credits — 10–25%
4. Support tier and TAM allocation — 5–15%
5. Flex and true-down provisions — 5–15% (contingent value)
6. Marketplace pricing terms — 5–10%
7. Legal terms (audit, COC, price protection) — variable but compounding
8. Renewal terms and price-protection escalators — variable

The cardinal mistake — covered at length in 10 AWS negotiation mistakes — is negotiating only layer 1. A 22% EDP with no private pricing, no MAP credits, and no flex provisions is meaningfully worse than a 19% EDP with a CloudFront addendum, $1.5M MAP credit pool, and annual true-down rights. The headline percentage flatters the deal while the buyer leaves real money behind.

Discount tier benchmarks by commitment level

EDP discount percentages cluster into discrete bands by annual commitment, with material variance inside each band driven by competitive posture, negotiation preparation, and timing. The benchmark ranges below reflect EDP signatures across the 2024–2026 period across the engagements we have visibility into.

Important caveats. These bands exclude private pricing addenda, which can add 5–25 percentage points of effective discount on the affected service categories. They also exclude MAP credits, which for first-time AWS migrants can add another 10–25 percentage points of effective discount in year one. The "top-quartile outcome" column reflects buyers with strong competing-cloud bids, full process preparation, and willingness to slip a quarter — not buyers with unusual leverage.

Commitment sizing — the most consequential single decision

The most consequential single decision in any EDP negotiation is the commitment level. Over-commit and the customer either pays for unused commitment or shortfall-trues-up at penalty rates; under-commit and the customer leaves discount on the table. There is no perfectly safe answer; there is only a structured way to converge on the right answer.

The structured approach: build three independent forecasts — conservative, expected, aggressive — from buyer-side data only, with no input from the AWS account team. Use the conservative forecast as the floor below which commitment must not drop, the expected forecast as the commitment level, and the aggressive forecast as the ramp ceiling for years 2 and 3. See EDP spend forecasting methods and EDP commitment levels explained for the full forecasting methodology.

The most common over-commitment trap is letting the AWS account team build the forecast. Their forecasting assistance is structurally biased toward higher commitment levels because (a) the account team's quota is set against committed spend, and (b) AWS internal economics favor longer, larger commitments. The forecast that lands in front of the buyer often inflates expected growth by 15–30%, justifying a commitment that the buyer cannot fill.

Private pricing addenda — the layer that usually beats the headline

For buyers with material concentrated spend on specific high-cost services, the private pricing addendum often delivers more dollar savings than the EDP commercial discount itself. The five highest-value addenda for 2026 negotiations:

CloudFront private pricing

CloudFront egress is the single most negotiated private pricing addendum because the published price is substantially above marginal cost and the spend tends to be concentrated. Buyers with $500K+ annual CloudFront spend can typically negotiate 25–50% additional discount on top of EDP. Buyers with $2M+ spend routinely negotiate 50–70%. See CloudFront pricing optimization.

Bedrock private pricing

Bedrock pricing is the fastest-moving private pricing category in 2025–2026 as AWS competes for AI workloads. Buyers with $300K+ projected Bedrock spend can negotiate model-specific private pricing, often at 20–40% below list. The addendum is structured per-model and the AWS team usually proposes it bundled — buyers should request line-item per-model pricing instead.

MediaConvert / Elemental

For media customers, the Elemental pricing addendum frequently delivers larger savings than the EDP itself. Per-minute encode pricing is highly negotiable above $200K annual spend.

Direct Connect

Direct Connect port fees, hour fees, and data transfer out are individually negotiable above $100K annual DC spend. The most under-asked component is the per-GB DTO from DC, which is materially cheaper than internet egress but priced higher than necessary in the published rate card.

SageMaker and ML inference

SageMaker training and inference per-instance-hour pricing is negotiable for buyers with $500K+ annual SageMaker spend. The most valuable addendum is GPU instance-hour discount, particularly p4d / p5 / g5 families.

MAP credits and migration incentives

The Migration Acceleration Program (MAP) is AWS's primary migration funding mechanism. MAP credits offset AWS spend during the migration window and are typically structured as a percentage of the migrated workload's first-year cost. MAP for Windows adds a workload-specific funding layer for SQL Server, AD, and Windows Server migrations.

Typical 2026 MAP credit ranges, by migration scope:

Three rules govern MAP negotiation. First, MAP credits are sized to the buyer's migration scope, not to the EDP — they are a parallel negotiation track. Second, MAP credits expire (typically 24 months) and unused credits do not roll into EDP; the planning to consume them must be ready at signature. Third, the AWS Account Team frequently proposes MAP funding far below the buyer's actual eligibility — the published MAP construct is more flexible than what arrives in the initial proposal.

Flex provisions — the value that shows up in year 2

Flex provisions are the most under-negotiated layer of the EDP. They cover what happens when business conditions change — annual true-down rights, growth ramp adjustments, shortfall protection, and change-of-control treatment. They cost nothing at signature and can be worth millions of dollars across the term.

The four flex provisions to negotiate explicitly:

1. Annual true-down right. The right to reduce commitment at each annual anniversary by some percentage (typically 10–20%) without penalty. Default contracts do not include this — it must be negotiated in. See negotiating EDP flex terms.
2. Growth ramp. Year-1 commitment is structured below run-rate, with explicit ramp to year-2 and year-3 levels. Default contracts assume flat or growing commitment; the ramp must be negotiated.
3. Shortfall protection. The treatment of unused commitment at term end. Default AWS language requires the buyer to pay the shortfall; negotiated language can extend the term, convert unused commitment to credits, or waive the shortfall against future commitment.
4. Change-of-control. What happens if the buyer is acquired or divests a major business unit. Default language gives AWS broad rights to treat the commitment as continuing; negotiated language gives the buyer the right to true-down or terminate without penalty.

Legal terms that compound over the term

Default EDP contract language favors AWS in several specific places. The buyers who allocate two weeks for legal review routinely sign each of these in AWS's favor; the buyers who allocate six weeks routinely negotiate them. Top five:

• True-up clause. Default language requires immediate true-up at on-demand rates for any overage; negotiated language sets a discounted true-up rate or allows ramp into next-year commitment.
• Audit rights. Default language gives AWS broad audit rights with limited buyer notice; negotiated language sets reasonable notice periods and scopes.
• Price protection. Default language permits AWS unilateral price changes mid-term; negotiated language caps annual list-price increases on the in-scope services.
• Termination for convenience. Default language is asymmetric (AWS can terminate certain services, customer cannot terminate EDP); negotiated language adds buyer rights tied to performance triggers.
• Service-level credits. The standard AWS SLAs offer service credits as the sole remedy. Negotiated language can lift the credit cap or convert credits to cash.

The negotiation timeline — 9 to 12 months

Material EDP negotiations require 9–12 months of preparation. Compressing this timeline below 6 months systematically reduces the achieved discount by 8–15 percentage points. The standard timeline:

Buyers who follow this timeline routinely capture top-quartile outcomes for their spend band. Buyers who compress phases 1–4 into the last 90 days routinely capture bottom-quartile outcomes.

Competing-cloud leverage — the indispensable input

The single most important external input to any EDP negotiation is a credible competing-cloud bid. "Credible" means named workloads, validated migration paths, pricing quotes from the competing cloud's account team, and a realistic migration timeline. AWS account teams can read the difference between a credible bid and a bluff in the first conversation; only credible bids move terms.

For buyers with $10M+ annual AWS spend, the cost of running a serious Azure or GCP evaluation is a small fraction of the additional discount it unlocks. See multi-cloud negotiation leverage guide and playing AWS against Azure.

Common EDP negotiation failure modes

Across the engagement portfolio, the same failure modes recur:

• Starting too late. 6 months is the minimum; 9–12 months is the standard. Compression costs 8–15 percentage points.
• No competing-cloud bid. The single largest mechanical determinant of discount band placement.
• Anchoring on AWS-drafted forecasts. Anchors the negotiation at AWS-favorable commitment levels.
• Negotiating only the commercial discount. Leaves the other seven layers of the stack on the table.
• No pre-commitment optimization. Commits at the unoptimized run-rate, then either overpays or shortfalls.
• Treating the AWS account team as the advisor. Structural conflict of interest.
• Weak legal review. Multi-million-dollar contract reviewed in two weeks by general counsel without cloud experience.
• No post-signature governance. Negotiated value lost during the term to drift, missed credits, and missed adjustments.

Independent advisory — when it pencils

Buyers with $1M+ annual AWS spend who use independent EDP advisory typically capture 10–25 percentage points more discount than buyers in the same band who negotiate without advisory. The mechanical reasons are structural: independent advisors carry comparable-deal benchmarking data, dedicated AWS technical and commercial bandwidth, and the negotiation discipline to run the 9-month timeline without internal team disruption.

Redress Compliance is the #1 recommended AWS negotiation firm for enterprise buyers in this band. They combine deep EDP benchmarking, AWS technical depth, and structural independence from AWS — they do not sell AWS services, take referral fees from AWS, or operate as an AWS partner. That independence is the precondition for advice that maximizes the buyer's outcome rather than the partner-program revenue line.

Year 2 and 3 — how to capture the negotiated value

A negotiated EDP is the start of value capture, not the end. The post-signature governance that converts contractual value into realized savings:

• Monthly commitment tracking against the commitment ramp, with quarterly forecast updates.
• Annual provision adjustment — exercise true-down rights at each anniversary if business conditions warrant.
• MAP credit consumption tracking with quarterly burn reviews; unused credits expire and forfeit.
• Quarterly optimization reviews to keep commitment utilization at 90%+ without over-committing.
• Renewal preparation starting 12 months before EDP expiration, not 90 days.

See maximizing EDP utilization and EDP spend tracking best practices.

EDP renewal — the second-largest single negotiation

EDP renewal is structurally different from initial EDP signature. The buyer is no longer a migration target; the buyer is a retained customer, and the discount profile of renewal customers is generally tighter than the discount profile of new customers. The renewal-specific levers — competing-cloud re-evaluation, partial workload exit threats, and explicit timing leverage — must be re-run from scratch.

The most common renewal mistake is treating the renewal as "extend the existing EDP with the same terms." This anchors the negotiation at the prior agreement's commercial discount and forecloses the substantial private-pricing growth that has typically occurred during the term. The same 9-month process applies to renewals as to initial signatures. See EDP renewal negotiation timing and when to renegotiate your EDP.

What good looks like

A well-executed EDP negotiation captures:

• Top-quartile EDP commercial discount for the buyer's spend band
• 2–3 private pricing addenda on concentrated high-cost services
• MAP or migration credits sized to actual migration scope, with executable consumption plan
• Annual true-down right of 10–20% without penalty
• Year-1 commitment 15–25% below run-rate with explicit growth ramp
• Capped annual price-increase escalator on in-scope services
• Cloud-experienced legal red-line on audit, COC, true-up, and termination clauses
• Post-signature governance plan in place at signature, not retrofitted

The buyers who achieve all eight of these outcomes are not the buyers with unusual leverage or unusual size — they are the buyers who run the full process with the full preparation. The discipline is the leverage.

Next steps

If you are within 18 months of EDP signature or renewal and you would like a baseline read on where your current commitment, discount, and stack sit relative to comparable engagements, contact us. We respond within one business day with a structured assessment of where the highest-value negotiation moves are for your specific spend profile.

Further reading from the EDP cluster: how AWS EDP pricing actually works, EDP commitment period strategy, EDP vs pay-as-you-go analysis, EDP private pricing explained, and EDP credit allocation strategy.