Microsoft EA Renewal: Evaluate the Structure, Not Just the Price
Microsoft's July 1, 2026 list increase raises E3 8 percent and frontline plans up to 43 percent, so the agreement you sign in 2026 locks your structure and your rate for three years.
Prepared by Redress Compliance · June 2026 · Representative 9,200 seat Microsoft estate (benchmark scenario, not a quote)
Executive Summary
A Microsoft renewal is sold as a price talk. It is a structure decision. The agreement vehicle, the edition mix, and the carried shelfware move more money than the headline discount, and they set your cost for the full three year term.
The clock matters in 2026. On July 1, 2026 Microsoft raises list prices: Microsoft 365 E3 from $36 to $39, E5 from $57 to $60, and the frontline plans F1 and F3 by 25 to 43 percent. A renewal closed before that date, with price protection in the paper, holds the lower rate for three years.
The biggest single lever is the suite tier. Account teams push a blanket E5 uplift across the whole base. On a representative 9,200 seat estate, a usage based mix of E5, E3, F3, and F1 lists at $4.41M a year against $6.70M for blanket E5 on the licensed count, a structural gap of $2.29M before any negotiated discount.
Two new SKUs reshape the 2026 evaluation. Microsoft 365 E7, the Frontier Suite, went generally available on May 1, 2026 at $99 per user, and Microsoft 365 Copilot stays at $30 per user. Both belong in the evaluation as targeted allocations, not estate wide defaults.
What a renewal evaluation actually decides
A renewal decides three things at once: the agreement vehicle, the product mix, and the price. The first two move more money than the third, yet most buyers spend their energy on the discount alone.
The structure you renew into is sticky. An Enterprise Agreement runs three years, and the seat counts and editions you set at signature define the floor you pay against until the next renewal. A discount on the wrong structure still overpays.
| Decision | Where the money is | Buyer move |
|---|---|---|
| Agreement vehicle | Commitment level, price protection, and exit terms differ by EA, MCA Enterprise, and CSP. | Evaluate all three against your size and Azure growth, not the incumbent EA by default. |
| Product mix | The blanket E5 uplift and duplicate add ons sit here, the largest line on most renewals. | Split the suite tier by who uses the features. Cut standalone SKUs the suite already covers. |
| Carried shelfware | Seats bought at the last renewal and never deployed still bill every anniversary. | Reconcile the active count to deployment data before you accept a single seat. |
| Headline discount | Visible and negotiable, but applied on top of whatever structure you accept. | Settle the structure first. Negotiate the discount last, against a usage baseline. |
How to inventory the Microsoft estate before renewal
A defensible baseline counts active usage by product, not the licenses on the last order. Deployment and sign in data set the number, and that number is your floor for the whole negotiation.
Pull three sources before the account team sends a quote: the Microsoft 365 admin center active usage reports, Entra sign in logs for true active users, and the prior enrollment order to find shelfware. The gap between licensed and active is the first line of savings.
The three data pulls that build the baseline
- Active usage by workload: who actually opens Teams, the Office apps, and the E5 security and voice features in the last 90 days.
- Identity reconciliation: disabled, duplicate, and service accounts stripped out of the billed count.
- Shelfware reconciliation: seats on the prior order with no matching active user, sized for removal at renewal.
On the worked 9,200 seat estate, the prior order carried 9,800 seats of blanket E5. The 600 seat shelfware block bills near $0.41M a year at list, and a usage based mix removes another $1.88M. The bridge below shows both.
Benchmark ranges: Redress Compliance advisory engagement file, 2024 to 2025. Confirmed against your estate during delivery.
How do you right size editions across E3, E5, E7, F1, and F3?
Right sizing matches each user to the lowest edition that covers their real work. The mistake is uniformity: one suite tier applied to everyone because it is simpler to quote, not because it fits.
Five editions cover most enterprises. Microsoft's published Microsoft 365 enterprise pricing sets the 2026 list rates below, and each rises on July 1, 2026.
| Edition | Who it fits | List $/user/mo (2026) | Seats | Annual list |
|---|---|---|---|---|
| Microsoft 365 E5 | Security, compliance, and voice power users | $57 | 4,800 | $3.28M |
| Microsoft 365 E3 | Standard knowledge workers | $36 | 2,200 | $0.95M |
| Microsoft 365 F3 | Frontline with a device and light apps | $8 | 1,700 | $0.16M |
| Microsoft 365 F1 | Frontline, no Office desktop apps | $2.25 | 500 | $0.01M |
| Usage based mix | Right sized to real use | Blended | 9,200 | $4.41M |
Microsoft 365 E7, the Frontier Suite, sits above E5 at $99 per user. It bundles E5, Copilot, the Entra Suite, and Agent 365, and Microsoft prices it about 15 percent below the roughly $117 those four cost separately. Treat it as a targeted SKU for agent heavy teams, not an estate wide upgrade.
How do you evaluate the Copilot, Security, and Compliance stacks?
These three stacks carry the most overlap and the most upsell. Evaluate each on real adoption, because they are where blanket uplifts and duplicate add ons hide.
Microsoft 365 Copilot
Copilot is a $30 per user add on on E3 or E5, confirmed on Microsoft's Copilot page. It is not a renewal default. Allocate it to teams with measured usage, and avoid committing the whole base on a three year term before adoption data exists.
The Security stack
E5 already includes Defender, Entra ID P2, and the security suite. The recurring waste is standalone SKUs that duplicate what E5 covers, bought by a different team and never reconciled at renewal. Map every standalone security SKU to its E5 equivalent before you renew either.
The Compliance stack
E5 Compliance covers Purview information protection, eDiscovery, and insider risk. Where only a subset of users needs it, the E5 Compliance add on on an E3 base often beats moving the whole group to full E5.
Per user versus per device: where should you switch?
Most Microsoft 365 editions are licensed per named user, which fits when each person uses one or more devices. The math inverts when many workers share one device, as on a manufacturing floor, a retail counter, or a hospital ward.
For a shared shift device, per user licensing multiplies: every worker who signs in needs their own F3 at $8. A per device license for that shared workstation is a single flat cost no matter how many shifts pass through it. The crossover sits around four workers per device.
| Scenario | Right model | Why |
|---|---|---|
| One worker, several devices | Per user | One user license covers all that person's devices. |
| Two or three workers per device | Per user, F3 or F1 | Still below the per device flat cost at this ratio. |
| Four or more workers per shared device | Per device | One device cost beats four or more named user seats. |
Illustration uses F3 at $8 per user and an illustrative $30 per device bundle. Benchmark ranges: Redress Compliance advisory engagement file, 2024 to 2025.
Which agreement vehicle should you renew into?
The vehicle is the most overlooked lever, because the incumbent EA renews by inertia. Three vehicles are live in 2026, and each carries different commitment, price protection, and flexibility.
| Vehicle | Floor and commitment | Price protection | Best fit |
|---|---|---|---|
| Enterprise Agreement | 500 seat minimum, three year term, consolidated anniversary true up. | Rate locked for the term once signed. | Stable estates that value the locked discount and Software Assurance. |
| MCA Enterprise | $500K minimum annual Microsoft spend, the modern EA replacement. | Locks the catalog rate only where negotiated, else list increases flow through. | Material Azure spend where MACC integration matters. |
| CSP | No long term minimum, monthly or annual terms through a partner. | None for the term, list moves pass straight through. | Smaller or volatile estates that value flexibility over a locked rate. |
Three contract mechanics decide the vehicle choice, and each one moves real money.
- The EA 500 seat floor: falling below it pushes you toward MCA Enterprise or CSP whether you want the move or not.
- Price protection is not automatic on MCA Enterprise: without it negotiated into the paper, the July 2026 increase reaches you at the next order.
- The EA anniversary order deadline: a hard date each year to add seats at the locked rate. Miss it and the addition prices at list.
Where the common advice on Microsoft renewals is wrong
The standard reseller advice is to renew the existing Enterprise Agreement as is and put the energy into the discount. We disagree.
In the renewals we evaluated in 2024 to 2025, renewing the structure unexamined locked in a blanket E5 base and carried shelfware that a vehicle and mix review would have cut. The discount, even a good one, applied on top of a structure already too large.
The buyer side move is to evaluate the vehicle, split the suite tier by real use, and remove shelfware before price is discussed. A renewal won on discount alone costs more than one rebuilt on the right vehicle and a usage based mix.
Opening quotes ran above a usage baseline.
Across roughly 40 to 60 Microsoft EA renewal evaluations in 2024 to 2025, opening renewal quotes ran 18 to 28 percent above a baseline built on real deployment data.
Structural gap at list on the worked estate.
On the representative 9,200 seat estate, moving from blanket E5 to a usage based mix cut the annual list cost by 34 percent, before any negotiated discount.
Benchmark ranges: Redress Compliance advisory engagement file, 2024 to 2025.
The 9 month renewal motion that compresses cost
Start twelve months out, lead with usage data, and keep price for last. The motion below compresses 18 to 28 percent off the opening quote because the structure is settled before the discount conversation begins.
Inventory and reconcile
Pull active usage, reconcile identities, and find shelfware. Build the usage based baseline by product. The data, not the prior quote, sets the number you negotiate against.
Right size and choose the vehicle
Split E5, E3, F3, and F1 by real use. Allocate Copilot and E7 to measured teams only. Evaluate EA, MCA Enterprise, and CSP against your size and Azure growth.
Negotiate and lock
Negotiate the discount against the baseline, secure price protection in writing, and close before the July 1 increase reaches the term. Confirm the anniversary order deadline.
Recommendation
Make the structure the basis of the renewal, not the discount headline. The vehicle and the edition mix move more money than any discount, and they lock for three years. Settle them first, on a usage baseline, and the discount conversation starts from a smaller, defensible number.
- Lead with usage data. A baseline built on active deployment, not the prior order, removes shelfware and right sizes the suite tier before Microsoft quotes.
- Lock the rate before July 1, 2026. Close with price protection in the paper so the list increase does not reach your term, and confirm the anniversary order deadline for future seats.
Redress Compliance runs this renewal evaluation on your side of the table only: baseline, right size, choose the vehicle, then price. We are glad to tie a meaningful part of the fee to delivered value.