A five point tier step rarely paid for the commitment increase it demanded
The Enterprise Discount Program ladder does not slide, it steps, and each threshold is worth roughly five points. Reaching the next one usually costs a far larger percentage increase in commitment, which is why the renewal is a sizing decision wearing the costume of a discount negotiation.
Prepared by Redress Compliance · August 15, 2026 · AWS advisory. Based on 25 to 35 AWS EDP renewals advised, 2024 to 2025.
Executive summary
The ladder is built to pull you up a step. The EDP pays 5 to 20 percent, stepped by commitment threshold: 5 to 10 percent at $1M to $5M, 10 to 15 percent at $5M to $25M, and 15 to 20 percent above $25M.
Each step is worth about five points, and the commitment increase required to reach it is usually a much larger percentage than five, which is the arithmetic nobody puts on the page.
Commitments sized to reach a higher tier left 10 to 25 percent of spend unused but still owed, and the shortfall is paid in cash rather than in a reduced discount.
Flexibility beats rate at renewal. Rollover, ramp, and a defined shortfall cure period change the outcome more than the headline number does.
Marketplace routing eases the arithmetic but does not rescue it, because the contribution toward commit is capped at 25 percent.
The ladder, and what each step costs
| Annual commitment | Discount band | What the step demands | When it pays |
|---|---|---|---|
| $1M to $5M | 5 to 10 percent | Entry to the program | Consumption already at the floor |
| $5M to $25M | 10 to 15 percent | A large multiple of the entry commitment | Growth is funded and dated |
| Above $25M | 15 to 20 percent | Commitment well beyond most estates | Consumption already exceeds the threshold |
| Any step up | About 5 points | Commitment you must consume in full | Only when the gap is small |
The marginal test, in one line. A step is worth about five points on the whole commitment. Reaching it requires committing more, and every dollar of that increase you fail to consume is paid anyway. So the question is never whether the higher tier has a better rate, because it always does. The question is whether the additional commitment is spend you were going to make regardless. If it is, take the step. If reaching the threshold means committing beyond your own trailing consumption, the five points are being funded by spend that will not happen, and the effective rate lands below the tier you left.
What to negotiate alongside the rate
- Rollover of unused commitment between years, which converts a hard penalty into a timing problem and matters most where demand is volatile.
- A ramp schedule that grows the commitment with verified consumption rather than with the forecast, protecting year one.
- A defined shortfall grace or cure period, so a temporary miss does not become a cash event.
- Review points on major change, covering divestiture, acquisition, or a workload migration that alters the base.
- Confirmed scope, including which spend counts toward the commitment and how Marketplace transactions qualify under your own private terms.
- The offer shape itself, since packages mix credits with rate discounts and change quarter to quarter, so compare on effective cost rather than on headline percentage.
The AWS EDP negotiation brief
Commitment sizing, the discount thresholds, the flexibility provisions, and the buyer side moves across the AWS estate.
Get the brief →The renewal is a sizing decision dressed as a discount negotiation
The advice from the account team is consistent and internally logical: commit to the highest tier you can reach, because the rate is best there. Every part of that sentence is true and the conclusion is still usually wrong, because it compares two rates while ignoring the two different commitments they sit on. A better rate on more money you must spend is not automatically cheaper than a worse rate on money you were going to spend anyway.
The structure of the ladder is what makes this trap so reliable. Because the bands are wide and the steps are worth roughly five points each, most estates sit somewhere in the middle of a band rather than just below a threshold. Moving up therefore means a large percentage increase in commitment to capture a modest rate improvement, and the improvement applies to a total that has just grown. That combination flatters the headline and quietly worsens the effective rate whenever any of the increase goes unconsumed. In the renewals we advised on, commitments set to reach a higher tier left 10 to 25 percent of spend unused but still owed, which is more than enough to erase a five point gain.
What makes the failure hard to see in advance is that the shortfall is paid in cash rather than in a smaller discount. A discount that quietly shrinks would be noticed in the rate conversation. A shortfall shows up later as a true up, in a different budget line, often in a different fiscal year, attributed to demand rather than to the commitment decision that created it. By then the tier looks like it was correct and the business looks like it underperformed.
The disciplined renewal inverts the order. Model three demand scenarios first, size the commitment to the conservative one, and require that case to make the agreement worthwhile on its own. Then fill the commitment with spend you would make regardless, including eligible Marketplace routing inside the 25 percent contribution cap. Only then look at which tier that commitment lands in, and treat the tier as an outcome rather than as a target. The upside is captured through rollover and ramp, not through a bigger pledge. The commitment mechanics sit in the EDP negotiation brief, the discount evidence in the benchmark brief, and the year round controls in the vendor management playbook.
- EDP, MACC and CUD commits sized from actual consumption, with three demand scenarios
- Every risky clause flagged with the exact quote, the page, and the replacement language
- A negotiation playbook, talking points, and a two page executive brief on day one
The renewal runway
Build the three cases
Conservative, base and high demand forecasts from billing detail rather than from the plan, with the conservative case tested first.
Map eligible spend
List everything that counts toward the commitment, including Marketplace routing inside the 25 percent contribution cap, before any tier is discussed.
Negotiate flexibility first
Settle rollover, ramp and the shortfall cure period alongside the rate, so the sizing decision is protected rather than merely optimized.
What the renewal file shows
Across roughly 25 to 35 AWS EDP renewals advised in 2024 and 2025, the tier ladder pulled buyers toward commitments they could not absorb:
On commitments sized to reach a higher discount tier, still owed in full and paid as cash rather than as a smaller discount.
Conservative, base and high. Buyers who modeled all three before agreeing a number rarely chased a threshold they could not fill.
The patterns: the tier treated as the objective, the shortfall treated as a demand failure, and flexibility terms negotiated after the number rather than with it.
The buyer side move is to let the tier be an outcome of the sizing. The wider library sits in the AWS practice.
Your first five moves
- Build conservative, base and high demand forecasts from billing detail, and test whether the conservative case alone justifies the agreement.
- List every dollar of eligible spend, including Marketplace routing inside the 25 percent contribution cap.
- Size the commitment to the conservative case, then read off whichever tier that lands in rather than reaching for one.
- Negotiate rollover, ramp and a shortfall cure period alongside the rate, not after it.
- Start six to nine months before expiry, so nothing is decided under deadline pressure. The AWS practice builds the model with you.
Frequently asked questions
How does the AWS EDP discount ladder work?
It steps rather than slides. The discount runs 5 to 20 percent overall, at roughly 5 to 10 percent between $1M and $5M of annual commitment, 10 to 15 percent between $5M and $25M, and 15 to 20 percent above $25M. Each threshold is worth about five points, and the offer usually mixes credits with rate discounts.
Why does stepping up a tier usually fail to pay?
Because the step is worth about five points while the commitment increase needed to reach it is often much larger in percentage terms. If reaching the next threshold requires committing well beyond your consumption, the unused portion is still owed, and the extra five points is erased by spend you never used.
What is the real cost of a shortfall?
It is paid in cash, not in a reduced discount. The effective rate is the discounted price plus the value of any commitment you fail to consume, so a commitment set to reach a tier that leaves 10 to 25 percent unused can produce a worse effective rate than a smaller commitment at a lower tier.
How should the commitment be sized?
To the conservative demand case, then use flexibility terms to capture upside. Model three scenarios before agreeing anything, and require the conservative case to still make the agreement worthwhile. A commitment you cannot miss is safer than a tier you cannot fill.
Does Marketplace spend help reach a tier?
It helps fill a commitment with spend you would make anyway rather than with speculative growth, which is a genuine advantage. The contribution toward commit is capped at 25 percent, so it eases the arithmetic without rescuing a commitment that was oversized to begin with. Confirm which transactions qualify in your own private terms.
Which flexibility terms matter most at renewal?
Rollover of unused commitment between years, a ramp schedule for growing workloads, a defined shortfall grace or cure period, and review points on major change. For volatile demand rollover usually matters most, because it converts a hard penalty into a timing problem.
When should an EDP renewal start?
Six to nine months before expiry. Early modeling is what prevents a rushed commitment at the wrong tier under deadline pressure, and it leaves time to build the three demand scenarios and map eligible Marketplace routing before the first conversation.
Do Savings Plans replace an EDP?
No. Savings Plans cover compute and layer underneath an EDP rather than substituting for it, and they add their own commitment lock. Model both together, because stacking two commitments against one uncertain demand forecast is how estates end up owing on two instruments at once.
Negotiating AWS 7: AWS's Playbook, and the Counters
Five moves AWS runs in almost every account: the forecast that is not yours, migration funding that buys your future baseline, the first offer, the scarcity story, and the renewal that reprices your own discipline.
