Microsoft's End of EA Volume Discounts: The List-Price Shift Buyer Guide

On November 1, 2025, Microsoft removed the automatic Level A-D volume discounts that rewarded enterprise scale under the EA, MPSA, and OSPA. This guide quantifies why the true renewal impact lands near 20% (not 8%) and shows exactly where you must now manufacture the discount the program no longer grants.

For as long as most licensing professionals have negotiated with Microsoft, seat count carried an automatic reward. Cross 2,400 users and you dropped into a better price band. Cross 6,000 and you dropped again. Those bands, Levels A through D, were baked into the Enterprise Agreement pricing engine. You did not negotiate them. They arrived by right of size. On November 1, 2025, Microsoft removed that reward for all Online Services products. Every organization, whether 250 seats or 250,000, now pays the Level A list price, the same rate published on Microsoft.com.

Microsoft's framing is 'simplification.' The buyer-side reading is different: the vendor eliminated a structural discount worth roughly 6% to 15% for large estates, and it did so in the same fiscal window it raised Microsoft 365 list prices. The two moves compound. If you model them separately, you underestimate the damage. This is the single most important point in this guide, and we will prove the arithmetic below. A Level C or D customer coming off a three-year lock is not looking at the 5% to 8% suggested by the E3 and E5 list price change. In market engagements we consistently model 15% to 20% above the prior effective rate, before Copilot, before Unified Support uplift, before any MCA-E transition cost.

This is a pillar guide. It covers what changed, precisely who is exposed, the timing trap that catches customers mid-term, the compounding math, the current 2026 list prices, and, most importantly, the levers you now have to manufacture discount that the program no longer hands you automatically. If you renew in fiscal 2026 or 2027, read every section. The default Microsoft renewal quote is now built on a materially worse baseline, and only disciplined preparation recovers the margin.

What Microsoft Actually Changed on November 1, 2025

Microsoft announced the change on August 12, 2025, stating it would 'flatten the volume licensing price level discounts (also known as waterfall discounts) for all Online Services products sold in Enterprise Agreement (EA), Microsoft Products and Services Agreement (MPSA), and Online Services Premium Agreement (OSPA) deals.' The effective date was November 1, 2025. In plain terms: the programmatic waterfall that stepped your unit price down as your user count climbed is gone. Everyone lands on Level A, the list price.

The scope is broad on the cloud side and narrow on the on-premises side. The change applies to Microsoft 365, Dynamics 365, Windows 365, Azure, GitHub, and the security, compliance, and identity products. On-premises software pricing (server and client licenses bought perpetually) is unchanged. It does not touch U.S. Government price lists (Federal, State, Local), Education volume licensing, or GCC High customers on AOS-G (the Agreement for Online Services, Government). If you sit in one of those carve-outs, this specific change does not hit you, though the separate July 2026 list-price increase may.

The mechanism matters because the old bands were not marketing. They were a fixed part of the pricing table. Level A was the baseline list price. Level B applied to roughly 250 to 2,399 users at approximately 5% to 8% off. Level C applied to 2,400 to 5,999 users at roughly 8% to 12% off. Level D applied to 6,000-plus users at roughly 10% to 15% off. Those percentages are the discount you just lost, automatically, with no negotiation required to trigger them and, now, no negotiation available to restore them at the program level. We cover the detailed cost by band in our companion piece on what losing Levels A-D actually costs.

Seat count no longer buys a discount. The band that rewarded scale is deleted, and only a negotiated custom discount can now replace it.

The Old Tier Structure, and Exactly What You Lost

To quantify your exposure you need the baseline. The following table reconstructs the pre-November 2025 price-level structure and the approximate automatic discount each band carried. Treat the percentages as the value now removed from your effective rate. An organization sitting at Level D was paying roughly 10% to 15% below list purely on account of size. That entire band collapsed to Level A on November 1.

Read the table against your own contract. If you renewed at Level C, expect an increase of roughly 9% from the discount removal alone. Level D customers should model roughly 12%. Level B customers face roughly 6%. Those figures are the price-level effect in isolation. They are not the full renewal number, because the list-price increase stacks on top. That stacking is where the real damage lives, and it is the reason so many first quotes shock the buyer.

There is a second, uglier layer for the largest estates. Many Level D customers negotiated additional discount on top of the automatic waterfall, sometimes 30% to 40% off Level D pricing. Microsoft is now well positioned to reset that negotiation from scratch. A 25,000-seat E3 customer who held 40% off Level D pricing is exposed to a quote that removes both the 40% negotiated discount and the 12% Level D volume discount. If you carried a deep custom discount stacked on a deep automatic band, you are the single most exposed profile in the entire installed base.

Why Your Increase Is Near 20%, Not 8%

This is the central angle of the guide, so we will show the arithmetic rather than assert it. Microsoft announced the Microsoft 365 and Office 365 list-price increase on December 4, 2025, effective July 1, 2026. E3 rises 8.3% (from $36 to $39 per user per month). E5 rises 5.3% (from $57 to $60). Read alone, those numbers suggest a mid-single-digit renewal impact. That reading is wrong for any large estate, because it ignores the discount you also just lost.

Work a concrete case. An organization at Level C for M365 E3 was paying roughly $32.45 per user per month (list of $36 minus about a 10% Level C discount). After both changes, that same seat prices at $39.00: the new list price, with no volume discount applied. That is a jump from $32.45 to $39.00, an increase of 20.2%. The headline E3 list increase was 8.3%, but the customer feels 20.2% because the removed discount and the higher list compound against each other. This is the entire thesis of this guide, unpacked in full in the compounding math breakdown.

Scaled to a real estate, the numbers are severe. A 10,000-user organization on M365 E3 moving from $32.45 to $39.00 per user per month faces an increase of approximately $785,400 per year. A 25,000-user organization on M365 E5 that previously enjoyed Level D volume discounts faces an effective annual increase of roughly $3 million. These are not worst-case scenarios. They are the arithmetic of the two changes applied to a standard renewal.

Now add the layers Microsoft does not put in the headline. Even 3% Copilot adoption pushes the combined figure from about 20% to roughly 23% against the pre-November 2025 baseline, before any Unified Support uplift, MACC requirement, or MCA-E transition cost. Unified Support deserves a specific warning: because its fee is calculated as a percentage of your product spend, a higher product spend automatically inflates your support bill with no corresponding increase in value delivered. The compounding does not stop at the subscription line.

An 8.3% list increase and a lost 12% band do not add to 20%. They compound to it. Model both, or you will under-forecast the renewal by double digits.

The Timing Trap: Why Renewal Date Decides Your Exposure

The change is renewal-triggered, and the trigger mechanics catch customers who assume a mid-term contract shields them. Here is the rule. For EA contracts that started on or before October 1, 2025, the no-programmatic-discount rule applies only when, after October 31, 2025, you add an online service you had not previously purchased. For all EA contracts that started on or after November 1, 2025, the no-programmatic-discount rule applies to every purchase.

Translate that into daily behavior. If your EA term extends past November 2025, you keep your current pricing on your existing subscriptions until renewal. Good. But any new cloud service you add after November 1, 2025, that was not already in your agreement prices at Level A list immediately. That means a routine mid-term addition (a new security SKU, a Dynamics module, a Copilot pilot) drops onto your bill at full list, not at your negotiated band. The trap is procedural: your procurement team executes a small add-on believing they are inside the old pricing, and they are not.

The buyer move is a moratorium on discretionary net-new adds until you have modeled the renewal properly. Do not let a $40,000 pilot anchor a new SKU at list price when you could fold it into a negotiated renewal three months later. If you can defer new-service additions to your renewal, do it, and negotiate them in as part of the whole. If you cannot defer, consider provisioning the pilot through CSP, where the pricing dynamics differ and the commitment is monthly. We compare the two paths in detail in our guide to CSP versus EA for Microsoft 365.

Current 2026 List Prices You Will Now Pay in Full

Because the discount bands are gone, the published list price is now your starting point, not your ceiling minus an automatic band. Know the numbers cold. The Microsoft 365 increase lands July 1, 2026: E3 moves from $36 to $39 per user per month, E5 from $57 to $60. Office 365 E1 held at $10 per user per month through the July 2026 changes. An EEA-driven no-Teams E3 SKU is available at $33.75, which is a live lever for cost-conscious estates that run Teams under a separate arrangement or not at all.

Two practical uses for this table. First, it is your benchmark: any renewal quote should be measured against these numbers, and any 'discount' Microsoft offers should be quantified as a percentage off these figures, not off some inflated starting point the deal desk invents. Benchmarking discipline against the new list baseline is covered in depth in our guide to benchmarking your Microsoft deal against the list-price baseline. Second, the no-Teams E3 SKU and the flat E1 price are reminders that SKU selection is now a discount lever in its own right. If seat count no longer buys you a band, the mix of what those seats consume is where you claw margin back.

Where You Must Now Manufacture Discount

The blunt reality: seat count will no longer automatically drive lower pricing. Large enterprises can still negotiate custom discounts, but those now depend on strategic factors Microsoft cares about (product adoption, platform commitment, overall spend, multi-year direction), not on the raw volume that used to trigger a band. You have to construct leverage that the program used to hand you for free. This is the discipline that separates a 20% increase from a flat renewal. We cover the full toolkit in how to manufacture Microsoft discount now that volume levels are gone.

Lever 1: A credible CSP alternative on the table

Model both EA and CSP quotes 9 to 12 months before EA expiry. A credible CSP alternative sitting on the table is itself the single strongest lever on the EA discount, because it converts a captive renewal into a competitive one. Historically the objection to CSP was the loss of the EA three-year price lock. That objection is now weaker: CSP offers three-year subscription terms for Microsoft 365 E3 and E5 (with or without Teams), Teams Enterprise, E5 Security, and E5 Compliance. You can hold a multi-year price on the CSP side while retaining monthly flexibility elsewhere. Take both quotes into the renewal as genuinely competing options, not as a bluff you cannot execute.

Lever 2: The MCA-E migration threat

Microsoft is steering customers toward the Microsoft Customer Agreement for Enterprise (MCA-E). That creates a lever. In market engagements, customers who credibly threaten the MCA-E migration at EA renewal typically capture 3 to 7 percent additional discount versus the previous EA term. The word 'credibly' is load-bearing: Microsoft's deal desk knows the difference between a customer who has done the transition analysis and one who is posturing. Do the analysis, understand the operational cost, and be prepared to move. The threat is only worth 3 to 7 points if you could actually execute it.

Lever 3: A deployment baseline scored against entitlement

This is the largest single source of renewal leverage, and most organizations leave it on the table. A deployment baseline (what you actually use) scored against your entitlement (what you pay for) exposes the gap between committed and consumed seats. In our engagements, disciplined buyers who bring a rigorous baseline held or cut the renewal band by 12 to 25 percent against the first Microsoft quote. The reason this works is structural, and it connects to the trap in the next section: the default renewal rolls forward every seat you committed three years ago, whether or not you still use it. Prove the overcount and you shrink the number Microsoft is trying to grow. Our companion piece on the EA true-up explains how the committed count inflates over a term.

Lever 4: Timing to Microsoft's fiscal calendar

Microsoft's fiscal year ends June 30, and each quarter close is a pressure point. Sales representatives face intense revenue targets and are measurably more flexible and generous with discount as those deadlines approach. Align your decision point, not just your renewal date, with the end of a Microsoft quarter, and ideally with the June 30 year-end. You cannot always move the renewal, but you can control when you signal your intent to sign, and that timing has quantifiable value. The full playbook sits in our EA negotiation tactics for 2026.

A credible CSP quote plus a defensible deployment baseline is worth 12 to 25 points against the first offer. That is not a discount Microsoft grants. It is one you build.

The True-Down Analysis: Removing Seats the Renewal Assumes You Still Need

The default renewal quote rolls the prior baseline forward. It assumes you still need every seat you committed three years ago. This assumption is almost always wrong, and it is wrong in Microsoft's favor. Here is why. Mid-term true-ups add seats but they never remove them. Every acquisition, every project ramp, every temporary headcount surge added seats to the committed count. None of those events removed seats when the need passed. By renewal, the committed count is often well above active use, and the renewal quote prices all of it forward at the new, higher, undiscounted rate.

Run a true-down analysis before you engage. Identify inactive accounts, departed employees still holding licenses, seats assigned to service accounts, and users over-licensed to E5 who only consume E3 functionality. The last category is often the largest recoverable line. Our guide to Microsoft 365 add-ons shows how E5 bundles frequently duplicate standalone add-ons already in the estate, another source of removable spend. Every seat you can prove is inactive is a seat you do not renew, and in the new pricing world where you have lost the volume band, right-sizing the count is one of the few unambiguous wins fully within your control.

Timing matters here too. You cannot true down mid-term under a standard EA, only at renewal (or at anniversary for certain constructs). That makes the pre-renewal window the one moment you can shed the accumulated overcommitment. Miss it and you re-commit the inflated number for another three years at the higher rate. Start the baseline exercise 9 to 12 months out, in parallel with the CSP and MCA-E modeling, so you enter the negotiation with the true-down already quantified.

MPSA and OSPA: The Same Hit, Different Wrapper

The change applies to MPSA and OSPA as well as EA. If you have held online services under MPSA, the volume price levels are gone there too, and the calculus of whether MPSA remains the right vehicle shifts. MPSA's historical appeal was flexibility across affiliates and transactions without a single committed EA. With the volume discount removed, the pricing advantage of holding a large MPSA estate narrows toward the EA and CSP alternatives. Whether MPSA is still worth holding depends on your transactional pattern, affiliate structure, and mix of on-premises versus cloud, since on-premises pricing under MPSA is unaffected. We work through the decision in MPSA online services after the discount removal.

The practical takeaway for MPSA holders: do not assume your agreement type protects you. Re-run the same three models (EA renewal, CSP, MCA-E) against your MPSA position, because the level playing field the change created means your optimal vehicle may have moved. On-premises-heavy estates have more reason to hold; cloud-heavy MPSA estates should test whether consolidation into a negotiated EA or a competitive CSP arrangement produces a better effective rate now that the automatic band is gone from all three programs alike.

On-Premises as a Partial Hedge

The change does not touch on-premises software pricing. That is not an invitation to reverse a cloud strategy, but it is a reminder that the licensing decision is not binary. For specific workloads, particularly stable, predictable ones where you do not need the constant feature velocity of the cloud subscription, perpetual on-premises licensing plus Software Assurance may now compare more favorably against a subscription that lost its volume discount and gained a list-price increase in the same year. We examine the trade-offs in online services versus on-premises licensing after the 2026 change.

The hedge is narrow and workload-specific. Most estates will remain predominantly cloud, and Microsoft's product direction pushes the value there. But the discount removal changed the relative economics enough that a disciplined buyer should at least test the on-premises comparison for eligible workloads rather than defaulting to the subscription because it is what the last renewal did. Where the numbers favor on-premises, that is a workload you have effectively insulated from both the discount removal and the annual list-price creep.

The Azure MACC Trap in the Same Renewal

Large renewals increasingly carry an Azure Microsoft Azure Consumption Commitment (MACC). The discount removal does not change Azure list mechanics directly, but the MACC interacts badly with an inflated overall spend. Two specific mechanics deserve caution. First, any consumption above your commit prices at the full Azure rate card, so an aggressive commit does not protect overage. Second, and more dangerous, a shortfall at term end means the unused commit converts to a back-billed invoice, or rolls forward only by written exception you negotiated in advance.

In practice, Microsoft sales teams size the MACC to a growth story you may not deliver. Combined with the higher subscription spend now flowing from the discount removal, an oversized MACC becomes a second line of unrecoverable cost. Size the commit to a defensible consumption forecast, not to the number that unlocks a headline concession, and negotiate rollover or shortfall-relief language in writing before you sign. Do not let the deal desk trade you a subscription discount for an Azure commitment you cannot consume, because the net effect can be a worse cash position than the concession appeared to deliver.

Your Renewal Playbook, in Sequence

Pull the levers together into an ordered plan. The sequence matters, because each step feeds the next and the whole exercise must complete before Microsoft's first quote anchors the negotiation. Start 9 to 12 months before your EA expiry. Preparation late in the window forfeits the CSP and MCA-E leverage entirely.

• Build the deployment baseline scored against entitlement. This is worth 12 to 25 points against the first quote and it must be complete before you engage.
• Run the true-down. Identify inactive, departed, service-account, and over-licensed seats. Quantify the seats you will not renew.
• Model the EA renewal at the new Level A list prices. Do not accept Microsoft's rolled-forward baseline as the starting number.
• Model a credible CSP alternative, using the new three-year CSP terms where you need price lock. This is your strongest single lever.
• Complete the MCA-E transition analysis so the migration threat is credible, not posturing, and worth its 3 to 7 points.
• Impose a moratorium on discretionary net-new online-service adds until the renewal, so you do not anchor SKUs at list mid-term.
• Size any Azure MACC to a defensible forecast and negotiate rollover and shortfall relief in writing before signing.
• Time your signal of intent to a Microsoft quarter close, ideally the June 30 fiscal year-end, to capture peak deal-desk flexibility.

Two supporting resources tie the sequence together. Our 2026 Microsoft price increase preparation guide lays out the full timeline and the buyer-side moves in order, and our impact analysis of the removed volume discounts quantifies the cost by seat band so you can size your own exposure before the first meeting.

The Bottom Line for 2026 Renewals

Microsoft removed a structural discount worth 6% to 15% for large estates and raised list prices in the same fiscal year. For a Level C or D customer coming off a lock, the two changes compound to a 15% to 20% increase against the prior effective rate, reaching 23% or more once modest Copilot adoption and Unified Support uplift are added. The 8.3% E3 headline is not your number. If you plan around it, you will be short by double digits at renewal.

The discount that used to arrive automatically must now be manufactured. A defensible deployment baseline, a credible CSP alternative, a genuine MCA-E migration analysis, a disciplined true-down, and timing to Microsoft's fiscal calendar are the tools that recover the margin. Buyers who bring all five hold or cut the band by 12 to 25 percent against the first quote. Buyers who accept the default renewal absorb the full 20-plus percent. The gap between those two outcomes is entirely a function of preparation, and preparation must start 9 to 12 months out. Begin now.

Related reading: Microsoft Price Level Discount Removal: What Losing Levels A-D Actually Costs.

Frequently asked questions

When did Microsoft remove the EA volume discounts?

Microsoft announced the change on August 12, 2025, and it took effect on November 1, 2025. It flattens the Level A-D waterfall discounts for all Online Services under the EA, MPSA, and OSPA. On-premises pricing, U.S. Government price lists, Education volume licensing, and GCC High AOS-G are not affected.

Why is my Microsoft increase closer to 20% and not 8%?

The 8.3% figure is only the E3 list-price increase effective July 1, 2026. It ignores the volume discount you also lost. A Level C customer moving from an effective $32.45 to the new $39.00 list price sees a 20.2% jump because the removed discount and the higher list compound against each other, not add.

Does the change affect my EA if my term runs past November 2025?

Your existing subscriptions keep their current pricing until renewal. However, any new online service you add after November 1, 2025, that was not already in your agreement prices immediately at Level A list. Defer discretionary net-new adds to your renewal, or provision pilots through CSP, to avoid anchoring SKUs at full list mid-term.

How do I get a discount now that volume levels are gone?

Seat count no longer triggers an automatic band, so you must manufacture leverage. The strongest levers are a defensible deployment baseline scored against entitlement (worth 12 to 25 points against the first quote), a credible CSP alternative on the table, a genuine MCA-E migration threat (3 to 7 points), a pre-renewal true-down, and timing your signal to Microsoft's June 30 fiscal year-end.

Is CSP now a viable alternative to the EA price lock?

Yes, more than before. The old objection was CSP's lack of a multi-year lock, but CSP now offers three-year subscription terms for Microsoft 365 E3 and E5, Teams Enterprise, and E5 Security and Compliance. Modeling a credible CSP quote 9 to 12 months before EA expiry gives you a competing option that is itself the strongest lever on the EA discount.

Are MPSA and on-premises licensing affected?

MPSA online services lost the volume discount just like the EA, so re-run your vehicle analysis. On-premises software pricing is unchanged, which makes perpetual licensing plus Software Assurance a narrow, workload-specific hedge for stable estates that do not need constant cloud feature velocity. Test the comparison rather than defaulting to the subscription.