Executive summary
Right-size the E3 and E5 mix, price Copilot on its own, and time the renewal against the July 2026 list increase, and the Enterprise Agreement uplift comes down. Microsoft sells the EA as a single three-year commitment, and most overpayment comes from a license mix carried forward unexamined, a Copilot count folded into the term, and a renewal timed for Microsoft's convenience rather than yours. This playbook works through how Microsoft prices the EA, what the July 2026 increase changes, how to set the E3 and E5 mix, how the true-up and Software Assurance work, how to handle a software asset management review, and the renewal calendar that caps the uplift.
Read this as a working sequence rather than a reference. The license levers and the contract levers reinforce one another, so a team that measures usage and fixes the mix before the renewal window arrives with a lower, defensible baseline and a stronger case on price. The order is the point: measure, right-size, time, then negotiate.
Our advisors negotiate Microsoft agreements on the buyer side only. Across more than 500 enterprise engagements, the buyers we advise have negotiated over $2.4 billion in software and cloud contracts at an average saving near 38 percent, and our audit defence work averages a 72 percent reduction against the initial claim. The figures below summarize that record and the public mechanics that frame a Microsoft EA.
1. How Microsoft prices the EA, and where the discount sits
The Enterprise Agreement is a three-year commitment to a count of users and products at a negotiated discount off Microsoft's published list, with Software Assurance carried across the term. The published list is the anchor, and the real number depends on the discount you negotiate, the program incentives applied, and how accurately your license mix matches actual use.
A structural change reset the math. Effective November 1, 2025, Microsoft removed the EA price level tiers that previously discounted by seat band, the levels long known as A through D. With the volume tiers gone, a renewal can no longer rely on crossing a seat threshold to reach a lower rate. The negotiated discount and the program incentives now carry the weight that seat count used to, which makes the license mix and the timing the levers that move the number.
The EA also sits inside a wider set of programs, and the renewal is the moment to test whether the agreement is still the right vehicle. Large estates may compare the EA against the Microsoft Customer Agreement and enterprise enrollment paths, while cloud-heavy organisations weigh how Azure commitments sit alongside the EA. The point is not to switch for its own sake, but to arrive at the renewal having priced the alternatives, because an account team negotiates differently against a buyer who has modeled the options than against one who treats the EA renewal as automatic.
Read the proposal the way it was built
The account team is measured on committed revenue and growth, not on your efficiency, so the first renewal proposal tends to carry the existing mix forward, add the new list increase, and fold Copilot into the term. Settle the structural terms first, the license mix, the Copilot treatment, the true-up mechanics, and the price protection, and treat the headline discount as the last move. A mix carried forward without examination is the most common reason an EA renewal lands higher than it needs to.
2. The July 2026 list increase and what it means for timing
The single most important date in a 2026 Microsoft renewal is July 1. Microsoft has published new list pricing effective that date, moving Microsoft 365 E3 into a band of about 36 to 39 dollars per user per month and E5 into a band of about 57 to 60 dollars per user per month. A renewal that closes before that date locks the prior list for the term where Software Assurance applies, which turns timing into a direct saving.
The implication is concrete. An estate renewing in the second half of 2026 faces the higher list as the starting point, so the negotiated discount has to work harder just to hold the prior effective price. An estate that can pull its renewal forward ahead of July 1 captures the prior list across the full three-year term, a structural advantage that no discount percentage replaces.
| SKU | Prior list, per user per month | Published band from July 1 2026 |
|---|---|---|
| Microsoft 365 E3 | about $36.00 | about $36.00 to $39.00 |
| Microsoft 365 E5 | about $57.00 | about $57.00 to $60.00 |
Where a renewal cannot move ahead of the date, the response is to settle the discount and the price protection against the new list explicitly, so the increase does not compound across the term. The increase is published and applies to everyone, which means it is a fact to plan around, not a negotiating defeat.
Insider note. Confirm in writing which list applies to your renewal, the prior or the post-July figure, and whether Software Assurance locks it for the full term. The answer is worth more over three years than several points of headline discount, so it belongs in the term sheet rather than a verbal assurance.
3. Right-sizing the E3 and E5 mix
The license mix is the largest controllable cost in most EAs, and it is usually carried forward from the last renewal without a fresh look. Microsoft 365 E3 covers the core productivity, identity, and management suite. E5 adds advanced security, compliance, voice, and analytics on top, at a list price well above E3, so an estate placed entirely on E5 pays for capabilities that only a subset of users actually require.
The right question is not E3 against E5 across the board, but how many users genuinely need the E5 add-ons, and whether targeted add-on SKUs cover those users for less than a full E5 upgrade of the whole estate. Security and compliance features that justify E5 for a regulated function rarely justify it for every frontline worker, and frontline staff may fit an F-series license at a fraction of the cost.
| User profile | Likely fit | Why |
|---|---|---|
| Frontline and deskless | F1 or F3 | Core apps and management without the full E3 suite |
| Standard knowledge worker | E3 | Full productivity and management, no advanced security add-ons |
| Security, compliance, voice users | E5 or E3 plus add-ons | Targeted add-ons can be cheaper than a full E5 estate |
Build the mix from a measured profile of the user base, not from the prior contract. A renewal that moves a meaningful share of users from E5 to E3 with targeted add-ons, and frontline staff to an F-series license, can cut the per-user run rate substantially before any discount is discussed.
The data to do this already exists in the tenant. Microsoft 365 admin reporting shows which users actively use the E5 security, compliance, and voice workloads and which sit on an E5 license without touching the features that justify it. Pull that usage by workload before the renewal, and the mix stops being a negotiation of opinions and becomes a reconciliation of evidence. A user assigned E5 who has never used an E5-only feature is a documented downgrade candidate, and a list of those users is a stronger position at the table than any general argument about cost.
Renewing a Microsoft EA or rebalancing the E3 and E5 mix? Our advisors model the profile and the term with you.
Microsoft EA Renewal4. Copilot priced separately, not folded into the uplift
Microsoft 365 Copilot is the line most likely to inflate a 2026 renewal, because the account team has every reason to attach it to the EA at full estate scale. Microsoft prices Microsoft 365 Copilot at about 30 dollars per user per month on an annual commitment, an add-on that sits on top of the base Microsoft 365 license. At estate scale that figure is large, so the commitment size is the decision, not the per-user rate.
Price Copilot as its own line with its own measured adoption case. A commitment to Copilot across the whole user base, locked for the EA term, pays for seats that may never adopt the tool. A smaller commitment sized to the users with a clear use case, with room to grow as adoption is proven, captures the capability without funding idle licenses for three years.
The adoption-first sequence
- Identify the roles with a concrete Copilot use case before agreeing any count.
- Commit to that measured population, not to the full estate, and keep growth as an option.
- Separate the Copilot term from the base EA term where possible, so adoption can be reassessed without reopening the whole agreement.
- Tie any larger commitment to a usage milestone rather than a calendar date.
5. The true-up and the enrollment structure
An Enterprise Agreement counts growth once a year through the true-up. You add users and products during the year and reconcile at the anniversary, billed in arrears at the agreed price. The structure is convenient, and it carries a trap: the true-up only counts up. Reductions generally wait for renewal, so a count set too high at signature is paid every year until the term ends.
This asymmetry is why the opening count matters so much. Microsoft benefits from a generous starting position because growth is captured automatically and shrinkage is not. The buyer discipline is to start from a measured count, add only evidenced growth, and treat the true-up as a reconciliation rather than a place to be relaxed about numbers.
| Movement | How the EA treats it | The buyer move |
|---|---|---|
| Users added | Counted at the annual true-up, billed in arrears | Add as needed, the mechanism is designed for this |
| Users removed | Not credited mid-term, waits for renewal | Set the opening count from measured use, not the prior contract |
| Product upgrades | Counted at true-up | Confirm the add-on path before committing the estate |
Insider note. Because the true-up only adjusts upward, the renewal is the one moment to remove licenses that the estate no longer uses. Run a usage reconciliation before renewal so the new baseline reflects real consumption, not the high-water mark the true-up has been holding.
6. Software Assurance and price protection
Software Assurance is the benefit layer carried across the EA term, and its quiet value is price protection. Where Software Assurance applies, the price agreed at signature holds for the term, which is exactly why timing against the July 2026 list increase matters so much. A renewal closed under the prior list with Software Assurance carries that prior list forward even as the published price rises.
Software Assurance also bundles benefits that organisations pay for and then leave unused, from training vouchers to deployment planning and home-use rights. These rarely justify the cost on their own, but where they are used they offset it, so the SA conversation is about which benefits the estate actually consumes, not a blanket accept or decline.
Price protection is also the answer to the compounding risk in a multi-year deal. Without a written cap, the prices that apply at each true-up and at renewal can drift with the published list, so a three-year commitment quietly becomes more expensive as it runs. Settle the uplift limits in the agreement itself, naming the maximum increase that can apply at renewal and confirming that the term price holds through the true-ups in between. A cap written into the contract is worth more over three years than a verbal assurance that the relationship will stay friendly.
7. Software asset management reviews and audit defence
A Microsoft software asset management review, or SAM engagement, is a commercial conversation wearing a compliance costume. Microsoft engages through a SAM partner or a formal audit clause to verify deployment against entitlement. The review does not make the partner's tooling the final word, nor its findings an invoice, and the gap between deployment, entitlement, and licensable use is where the defence lives.
The most common over-statements involve server products counted without the correct licensing model, users double-counted across overlapping SKUs, and cloud services attributed without checking the entitlement that already covers them. Each is contestable with evidence, which is why an independent license position, run in parallel and reviewed before anything is shared, is the centre of the defence.
| Control | What it means | The buyer move |
|---|---|---|
| Process | Scope, timeline, and communication channel | Agree scope in writing, route contact through one owner |
| Data | What is collected and reviewed by whom | Run an independent count, review before anything is sent |
| Narrative | The status of the findings | Treat findings as claims to test, not facts to settle |
Insider note. A SAM review often arrives in the window before a renewal, because a compliance gap is a lever on the renewal price. Run your own position early, so a finding becomes a point to negotiate rather than a surprise that sets the tone of the whole renewal.
8. Unified Support and the percentage question
Microsoft Unified Support is priced as a percentage of the products an organisation buys, which means the support bill grows automatically as the estate grows, including as Copilot and E5 raise the underlying spend. This structure replaced the older fixed-incident model, and it is the support line that most often escapes scrutiny in a renewal because it is quoted separately from the licenses.
The percentage basis is the part to negotiate. Because Unified Support scales with product spend, a renewal that adds Copilot and shifts users to E5 raises the support cost twice, once on the licenses and again on the support percentage applied to them. Bring Unified Support into the same conversation as the license mix, and test whether the organisation's real support consumption justifies the tier and the percentage on offer.
9. The negotiation calendar and the lever order
Microsoft negotiations reward a calendar, not a meeting. The levers land when you have a license baseline, a defensible mix, and time before a quarter end concentrates the account team's flexibility. Microsoft runs a fiscal year ending June 30, so the June quarter end carries the most weight, with the other quarter ends adding pressure. The illustrative index below shows where buyer-side preparation changes the outcome. It is an illustrative index with the prepared position set to 100, not a market benchmark.
Relative negotiating position by preparation stage, illustrative index (prepared = 100)
Preparation, not pressure, moves a Microsoft outcome. Illustrative index, not a quote.
Begin at least six months before the anniversary. Establish the license baseline, model the E3 and E5 mix and the Copilot population, decide whether the renewal can move ahead of July 1, and settle structure before the headline discount. Never lead with your own renewal deadline, because the deadline is the account team's strongest lever if you reveal it first.
| Clause | What to verify |
|---|---|
| Applicable list | Prior or post-July 2026 list, locked by Software Assurance for the term |
| License mix | E3, E5, and F-series matched to a measured user profile, not the prior contract |
| Copilot | Sized to the measured population, term separable from the base EA where possible |
| True-up | Opening count set from real use, reconciliation method defined |
| Price protection | Renewal caps and uplift limits written, not assumed |
| Unified Support | Percentage and tier negotiated against measured support use |
Our recommendation: rebuild the E3 and E5 mix from a measured user profile, price Copilot to the population with a real use case, decide the renewal timing against the July 1 2026 list increase, set the opening true-up count from actual use, and settle Software Assurance price protection and Unified Support before you discuss the headline discount. Treat the published list as an anchor and the contract structure as the place the multi-year money is won.
Sources: Microsoft public list pricing for Microsoft 365 E3, E5, and Copilot, the November 1 2025 EA price level change, the July 1 2026 list increase, Software Assurance terms, and Unified Support, as available at the time of review. Outcome ranges are Atonement Licensing advisory figures, indicative and deal-specific, not a quote.
Related reading: Microsoft Licensing hub, Microsoft EA Renewal, Azure MACC Negotiation Guide, and AWS EDP Negotiation Playbook.