New Commerce made the commitments small, numerous, and easy to lose
The Cloud Solution Provider program sells Microsoft licensing through a partner rather than on direct paper, with no 500-seat floor and no anniversary true-up. Under New Commerce it is not the flexible, commitment-free alternative to the EA that it is sold as: it is a web of small commitments, locked terms, a seven-day cancellation window, and no mid-term reductions, spread across dozens of end dates nobody tracks. Run it with EA-grade governance, or stay on the EA and use the CSP quote as leverage.
Prepared by Redress Compliance · August 9, 2026 · Microsoft advisory. Based on roughly 25 to 35 Microsoft estates advised 2024 to 2025.
Executive summary
The term sets the price, and monthly flexibility costs roughly 20 percent. Under New Commerce, annual and three-year terms lock pricing while a monthly term carries roughly a 20 percent premium, so pay it only on seats that genuinely churn.
CSP holds subscriptions per seat and per term through a partner rather than one organization-wide enrollment, with no 500-seat floor and no anniversary true-up, so seats mix and match per subscription in a way the EA cannot.
The flexibility now lives in the term mix, not in the subscription: upgrades move mid-term but downgrades wait for term end, and term-end dates sprawl unless you co-term deliberately.
The seven-day window is the trap: a New Commerce subscription can be canceled only in the first seven days of its term.
After day seven the term is committed to its end, the partner cannot waive it because the rule is Microsoft's own carried in the Product Terms, and mid-term seat reductions are gone.
Teams that missed the window carried unwanted subscriptions for a full term in roughly one estate in four, so put every new subscription start date into a calendar with a day-five review.
Estates treating CSP as commitment-free carried 10 to 20 percent dead subscription weight within a year, because under New Commerce the estate becomes a portfolio of term-end dates and the calendar, not the contract, is where the money leaks.
A deliberate term mix cut the M365 line 10 to 18 percent, because CSP wins when the mix matches seat behavior. A workforce that is 70 percent stable and 30 percent seasonal prices the stable block on annual or three-year terms and the churn on monthly, which an EA cannot replicate.
Split the estate into stable, variable and seasonal blocks from HR and assignment data, price the stable blocks on annual or three-year terms, put genuinely seasonal seats on monthly and accept the premium knowingly, and co-term the end dates into one or two renewal windows a year.
Partner selection then moves pricing more than Microsoft list does: identical SKU mixes priced 5 to 12 percent apart across partners, so re-quote at least two.
The partner margin is the negotiation surface, and governance is what keeps the saving. Microsoft list is fixed in the channel, so the negotiation is with the partner over margin, onboarding and support bundles, and promo alignment, and competition between partners is your only real lever.
CSP replaces one renewal event with continuous order discipline, so without it term sprawl and assignment drift quietly rebuild the waste the move was meant to remove: run one ordering gate so all new subscriptions route through procurement.
A quarterly utilization pass because unassigned seats on annual terms are the new shelfware, and a term calendar logging every end date, owner and downgrade decision ahead of the window.
Estates over roughly 2,400 stable seats with an Azure commitment usually still price better on EA paper, using the CSP quote as leverage rather than the destination.
The New Commerce term options
| Term | Price behavior | Cancellation | Fits |
|---|---|---|---|
| Monthly | Roughly 20 percent premium | 7-day window each month | Seasonal and contractor seats |
| Annual | Locked for 12 months | 7 days from start | Core workforce |
| Three year | Locked for 36 months | 7 days from start | Stable baseline seats |
New Commerce attached real commitment to CSP: terms lock pricing, cancellation closes after seven days, and mid-term seat reductions are gone.
The rules are Microsoft's own, carried in the Product Terms, so no partner can waive them, and the flexibility now lives in the term mix rather than in the subscription.
The EA prices one enrolled organization with one renewal event.
CSP prices each subscription on its own term through a partner, so flexibility replaces the single negotiation table, but you can step a subscription up mid-term while a step down waits for term end, and dozens of subscriptions mean dozens of renewal dates unless you co-term deliberately.
The side-by-side against the EA sits in the CSP versus EA comparison, and Microsoft's steering toward the model in the shift-to-CSP playbook.
When CSP is cheaper, and who should not move
- Split the estate into stable, variable and seasonal seat blocks from HR and assignment data, because CSP wins only when the term mix matches seat behavior, not as a blanket move.
- Price the stable blocks on annual or three-year CSP terms, or leave them on EA paper if the discount holds, because a 70 percent stable, 30 percent seasonal workforce is exactly the shape an EA cannot replicate.
- Put genuinely seasonal seats on monthly terms and accept the roughly 20 percent premium knowingly, because it buys true month-to-month flexibility and is worth paying only where the seats churn.
- Co-term subscription end dates into one or two renewal windows a year, and re-quote at least two partners, because identical SKUs price 5 to 12 percent apart and partner competition is your only real lever on a fixed Microsoft list.
- Estates over roughly 2,400 stable seats with an Azure commitment usually still price better on EA paper, so use the CSP quote as renewal leverage rather than the destination. Even committed EA estates should price CSP, because the quote disciplines the EA renewal.
The Microsoft EA renewal playbook
The renewal moves, the EA framework, the M365 SKU framework, and the buyer-side moves across the full Microsoft estate.
Get the white paper →The governance a CSP estate needs
CSP replaces one renewal event with continuous order discipline, and without it term sprawl and assignment drift quietly rebuild the waste the move was meant to remove.
Three controls hold it: one ordering gate, so all new subscriptions route through procurement and never through partner self-serve portals alone, because an ungoverned portal is how the seven-day window gets missed.
A quarterly utilization pass, because unassigned seats on annual terms are the new shelfware and they do not surface at a single anniversary the way EA true-up shelfware did.
And a term calendar logging every end date, owner and downgrade decision ahead of the window, because under New Commerce the estate becomes a portfolio of term-end dates and the calendar, not the contract, is where the money leaks.
The seven-day trap is the sharpest edge: put every new subscription start date into the calendar with a day-five review, because after day seven the term is committed and the partner cannot waive it, the rule being Microsoft's not theirs.
What is actually negotiable is the partner margin, the onboarding and support bundles, and promo alignment, since Microsoft list is fixed in the channel, so the negotiation is with the partner and competition between partners is the lever.
Run CSP with EA-grade governance, a single ordering gate, a term calendar and a quarterly utilization pass, or stay on the EA and use the CSP quote as leverage. The seat-type detail sits in the M365 license types comparison and the reclaim tooling in the license optimizer.
- Percentile standing for your exact deal size and industry, from real closed transactions
- Scenario simulation before the call: test alternative terms and see the financial impact of each
- A negotiation playbook, talking points, and a two page executive brief on day one
What we saw across CSP transitions, 2024 to 2025
Across roughly 25 to 35 Microsoft estates Morten Andersen advised between 2024 and 2025, CSP moved from fallback to first option for estates under the EA practical floor, and the common advice about it is what generates the waste.
The standard advice frames CSP as the flexible, commitment-free alternative to the EA. We disagree:
Where estates blended annual terms for stable seats with monthly for seasonal staff, against a flat EA renewal quote, from a deliberate term mix.
Estates that missed the seven-day cancellation window and carried unwanted subscriptions for a full term as a result.
New Commerce CSP behaved as a web of small commitments, locked terms, a seven-day cancellation window, and no mid-term reductions, spread across dozens of end dates nobody tracked, and estates treating it as commitment-free carried 10 to 20 percent dead subscription weight within a year.
Partner selection moved pricing more than Microsoft list did: quotes for the identical SKU mix varied 5 to 12 percent across partners, which is why re-quoting at least two is the cheapest lever available.
The buyer-side move is to run CSP with EA-grade governance, a single ordering gate, a term calendar, and a quarterly utilization pass, or stay on the EA and use the CSP quote as leverage.
Under New Commerce the calendar, not the contract, is where the money leaks, so the discipline that matters is not the negotiation but the continuous order gate that stops a seasonal seat renewing on an annual term and an unassigned seat surviving a quarter.
The EA comparison for larger estates sits in the CSP versus EA comparison, and the renewal framing in the EA renewal playbook.
Your first five moves
- Split the seat estate into stable, variable and seasonal blocks from HR and assignment data, because CSP only wins where the term mix matches seat behavior.
- Quote two or more CSP partners on the identical SKU and term mix, because identical SKUs price 5 to 12 percent apart and partner competition is your only real lever.
- Build the term calendar with owners and downgrade decisions per end date, because under New Commerce the calendar, not the contract, is where the money leaks.
- Set the day-five review for every new subscription start, because after day seven the term is committed and no partner can waive it.
- Run the quarterly unassigned-seat sweep on all annual terms, because idle seats on annual terms are the new shelfware. The Microsoft practice runs the governance with you.
Frequently asked questions
What is Microsoft CSP licensing?
The Cloud Solution Provider program sells Microsoft 365, Dynamics and Azure subscriptions through a partner who owns the billing, first-line support and margin, while Microsoft sets list pricing and the New Commerce term rules.
You hold subscriptions per seat and per term rather than one organization-wide enrollment, with no 500-seat floor and no anniversary true-up, so seats mix and match per subscription in a way the Enterprise Agreement cannot replicate.
How is CSP different from a Microsoft Enterprise Agreement?
The EA prices one enrolled organization with locked pricing and one annual true-up across a three-year enrollment, concentrating leverage in a single renewal event.
CSP prices each subscription on its own term through a partner with no seat floor, trading that single negotiation table for term flexibility.
Under New Commerce that flexibility lives in the term mix, not the subscription: terms lock pricing, cancellation closes after seven days, and mid-term reductions are gone.
What does the monthly CSP term premium cost?
Roughly 20 percent over the annual term price under New Commerce. It buys genuine month-to-month flexibility and is worth paying only on seats that genuinely churn, such as seasonal and contractor staff.
The buyer-side pattern is to price stable seats on annual or three-year terms where pricing locks, and pay the monthly premium knowingly on the block of seats that actually vary, rather than putting the whole estate on either extreme.
Can you cancel a Microsoft CSP subscription mid-term?
Only within the first seven days of the term. After that the subscription is committed to term end, and the partner cannot waive it because the rule is Microsoft's own, carried in the Product Terms. You can step a subscription up mid-term but not reduce or cancel it.
That is why every new subscription start date needs a calendar entry with a day-five review, since teams that missed the window carried unwanted subscriptions for a full term in roughly one estate in four.
Is CSP cheaper than an Enterprise Agreement?
For estates under roughly 2,400 seats or with significant seasonal staff, often yes, by 10 to 18 percent in our engagement file when the term mix is deliberate, because pricing stable blocks on annual or three-year terms and churn on monthly is a shape the EA cannot match.
Large stable estates with Azure commitments usually still price better on EA paper, so they should price CSP as renewal leverage rather than as the destination.
What governance does a CSP estate need?
EA-grade discipline, because CSP replaces one renewal event with continuous order management and without it term sprawl and assignment drift rebuild the waste.
Run one ordering gate so all new subscriptions route through procurement rather than partner self-serve portals, a quarterly utilization pass because unassigned seats on annual terms are the new shelfware, and a term calendar logging every end date.
Owner and downgrade decision ahead of the window, because under New Commerce the calendar, not the contract, is where the money leaks.
The Microsoft EA Preparation Playbook: The Work That Wins the Renewal
Five workstreams in order: the license position, the usage file, the demand forecast, the benchmark and alternatives files, and the ask list drafted before Microsoft drafts theirs, with the executives aligned before the first meeting.
