True up is the EA mechanism Microsoft uses to convert organic growth into invoiced expansion. It is not a surprise. It is the structure. The leverage is set at signature, not at the anniversary invoice. This page is the practice view on how true up works, why it costs buyers more than it should, and what to negotiate before the meter starts running.
The Enterprise Agreement permits the buyer to deploy new licenses across the year without prior approval and reconciles the deployment annually at the anniversary. The true up is that reconciliation. It is invoiced as net new quantity for the remaining contract term, billed at the locked price level set at signature.
The true up is a self report. The buyer is contractually obligated to count its own deployments and submit the count to Microsoft each anniversary. Microsoft does not measure the buyer. Microsoft accepts the buyer report and reserves the right to verify under the audit clause of the master agreement.
Net new licenses, calculated as the year end count minus the baseline at the prior anniversary, are priced at the locked level. The first year true up is the most visible because it captures all growth since signature. The second year true up captures only year two growth.
Microsoft issues the true up invoice within sixty days of the anniversary. The invoice is binding once accepted. Disputes have to be filed within a defined window and require contemporaneous deployment evidence the buyer should already have on hand.
The new total becomes the floor for the remainder of the term. The buyer cannot return to the prior count mid cycle. This is the asymmetry at the heart of the EA: the count can rise across the term but cannot fall until renewal.
The true up is structurally designed to reward the buyer who deploys deliberately and penalize the buyer who deploys reflexively. We see the same five patterns drive overpayment across most engagements.
License objects assigned to users who left, to test accounts that became permanent, or to service accounts that did not require licensing in the first place. Counts inflate. The true up captures the inflation as new growth.
Defender, Purview, and Teams add ons get attached to the base E3 footprint without a corresponding rationalization of what E3 already includes. The true up bills the add ons as net new quantity even when the use right exists in the base SKU.
An M&A event adds employees mid term. The buyer absorbs the new users into the EA. The true up captures every new seat at the locked price for the remainder of the term, without any negotiation reset.
Users moved from E3 to E5 mid term without step up math negotiated at signature. The true up prices the upgrade at market rather than at the EA discount, eroding the value of the agreement on the most consequential moves.
Missed reporting windows trigger automatic estimates from Microsoft. Microsoft estimates conservatively for itself. Buyers who submit late pay a premium even when their actual count is lower than the estimate.
Across three years, small decisions accumulate. Each individual true up looks reasonable. The third year shows a footprint thirty percent larger than the buyer can defend on consumption. By then the floor is set.
True up exposure is mostly determined before the agreement is signed. Five drafting points control the math across the term. None of them are standard. All of them are negotiable in a credible EA negotiation.
The mid term move from E3 to E5 should be locked as a defined delta at the same level as the base SKU. Microsoft will resist locking it. Buyers who land this protect roughly twenty percent of mid term cost on the most consequential move they are likely to make.
Use rights for products Microsoft has not yet released but is signaling. The Copilot families showed the cost of not having this clause. The current generation of agentic and AI surfaces is the next time this matters.
How acquired headcount enters the agreement should be defined at signature, including transition timing, pricing, and the right to keep an acquired entity out of the agreement for a defined window if its existing licensing makes that economically rational.
The reporting count should match the consumption baseline the buyer can substantiate. A defined evidence standard at signature reduces audit exposure when the count is later questioned.
The standard EA gives the buyer no true down ability during the term. The renewal is the only opportunity to right size. A structured true down clause negotiated at signature can convert the next renewal from a posture conversation into a clean reset, dropping the count to the consumption baseline without penalty. This single clause is the single highest leverage drafting point in the entire EA. Buyers who do not negotiate it pay the floor for the next term.
The practice treats each true up as a discrete engagement, not a procurement formality. The sequence is the same whether the buyer expects a small reconciliation or a significant one.
We start by rebuilding the count from the buyer side. The Microsoft view comes from the Volume Licensing Service Center and the customer reported deployment. The buyer side view comes from Entra ID, the Microsoft 365 admin center, Intune, the Azure consumption tenant, and any third party identity platform the buyer runs. The reconciliation surfaces the gap between assigned licenses and active users, between provisioned objects and authentication events, and between the Microsoft view and the consumption truth.
We then test every add on against the base entitlement. Defender for Endpoint, Defender for Office, Purview, Audit Premium, and Teams calling add ons are routinely deployed alongside base E3 footprints when the use right already exists at the buyer's level. The test is whether the add on delivers a use right the base SKU does not already provide. The answer surprises most procurement teams.
We then sequence the report. Late submission is expensive. Premature submission of an inflated count is more expensive across the term. The practice files the true up at the right count, in the right window, with the supporting evidence the buyer should be able to produce on demand.
Where the buyer has acquisition activity or significant organizational change in the year, we negotiate the true up rather than submit it. Microsoft has discretion. The discretion is exercised in favor of the buyer who is prepared to use it.
Anonymized. Verifiable on reference call. Within the trailing twelve months.
The Microsoft estimate captured every license object provisioned in the year, including dormant test accounts, departed contractor identities, and shadow assignments from acquired entities. The practice rebuilt the count from authentication data, filed the corrected true up with supporting evidence, and negotiated treatment of the acquired entity headcount as a separate event.
They submitted a count we could defend. The Microsoft team did not push back, because the evidence was already in the file.Vice President of Procurement · Top 25 global retailer
Two analyst calls. No pitch. We tell you what we would do, what the leverage actually is, and whether we are the right firm for this engagement.