M&A, divestiture, multinational structures, education and public sector agreements, and what transfers. Three knowledge checks along the way, and 1 clip from a senior cloud advisor.
The presenter in this session is an AI generated avatar. The curriculum and guidance are real, produced by Redress Compliance analysts from our consulting engagements and market network.
This is a taught session, not a talking head. The instructor works through analyst grade slides, and three times the video stops on a question with four options on screen. Pause, commit to an answer, and the next slide explains which option is right and why each of the others is wrong. Once in the session the frame splits and a senior cloud advisor gives the view from inside real Oracle negotiations, and the instructor picks the clip apart when the slides return.
The full narration of this session, section by section, for reading and reference. Guest analyst clips are marked.
Welcome back, session thirty nine of forty, the last of the content sessions before the capstone. Today is what happens when the company itself changes shape, and there is one fact underneath every case we will look at. Licences are granted to a named legal entity under an agreement. Not to a business, not to a set of people, not to an org chart. To an entity. Which means that when the entity changes, through an acquisition, a divestiture, a restructure, or a transfer, the grant does not automatically follow the people. That sounds like a technicality and it is the whole session, because every expensive surprise in this area is a version of it. And there is a second difficulty that compounds the first: deal timetables are set by lawyers and bankers, so the licensing question is always discovered against a completion date that was fixed before anybody thought to ask about it.
Five takeaways. One, the governing principle: licences are granted to a named legal entity under an agreement, so when the entity changes the grant does not automatically follow. Two, acquisitions arrive with an estate: their agreement, their vehicle, their term dates, and their compliance position, all four becoming yours to manage and none aligning with what you already run. Three, divestitures are harder, because separating a population from an agreement is a licensing project with a legal deadline attached and it is normally discovered after the deal timetable is fixed. Four, corporate events attract attention, being one of the five ways a compliance conversation opens from session thirty one, precisely because entitlement rarely survives them cleanly. Five, some families do not mix: government, education, and non profit SKUs carry their own eligibility and pricing, and mixed estates carry more than one family.
The principle underneath all of them, three consequences of that one fact. Entities matter more than org charts: your internal structure can reorganise freely, while the licensing position follows the legal entities named in the agreement, which is why the first question in any corporate event is which entity holds what, and why nobody in the deal team can answer it. Transfer is permission rather than mechanics: moving entitlement between entities is something the agreement either permits or does not, and where it does the process has conditions and timing attached, so it is never simply an administrative reassignment however administrative it looks in a project plan. And the clock is somebody else's, because deal timetables are set by lawyers and bankers and the licensing work gets discovered against a completion date fixed before anybody asked. The practical implication: your session thirty five register is worth most at exactly the moment a corporate event begins.
Acquiring a company, what arrives with them and what to establish first. Their agreement: which vehicle, which entity, what term, and it runs to its own dates regardless of your plans. Their seat estate: how many, on what tiers, actually used, which is the session thirty two join run on somebody else's tenant. Their compliance position: any known gaps and any open engagements, because inherited exposure is still exposure per session thirty one. Their tenant: one tenant or two, and for how long, since duplicate assignments across tenants are a real and recurring cost. And their partner: who is partner of record and on what terms, which is a relationship you did not select, per session thirty six. The temptation is to consolidate everything onto your agreement immediately, and that is frequently wrong, because a volatile or possibly divestible population suits a flexible profile better.
First check. You acquire a company and want their users on your tenant next month. What has to be true? A, nothing, tenant migration is a technical project. B, the licensing position has to be established first: which entity holds what, whether it transfers, and what their existing agreement obliges them to keep paying for. C, only that you have enough spare licences in your own tenant. D, that their partner agrees to the migration. Pause it, and as you think, ask what happens to their existing agreement at the moment their users stop using it.
The answer is B, and the failure mode is specific and expensive. You migrate the users, you buy licences for them on your tenant, and the acquired entity's committed agreement carries on billing for the seats those people no longer use, because a commitment does not end when the usage does. That is the session thirty five problem arriving inside a corporate event, and it can run for the remainder of a three year term with nobody noticing because the invoice goes to a subsidiary. A treats a licensing question as an infrastructure one, which is exactly how the double payment happens. C solves half the problem and it is the cheaper half. D is a real coordination step that follows the decision rather than determining it. Establish the entity position, the transfer rights, and the residual obligation, then plan the migration around what the agreements actually permit rather than around what the project plan assumed.
Divesting one, five things that make it harder than an acquisition. The population has to be separated: identifying exactly which users, which licences, and which workloads leave with the business is a data exercise nobody has done, run against a completion date. Your commitment does not shrink: an EA commitment made for the whole organisation does not reduce because part of the organisation left, which is the session twenty one asymmetry in its most painful form. They need something on day one, because the divested business needs working email and identity from completion, which usually means a transitional arrangement negotiated under time pressure. Transitional service agreements have a licensing tail, since somebody has to be licensed for services you provide during transition. And the remaining estate is now the wrong size, because your seat count, commitment, and vehicle were all sized for an organisation that no longer exists in that shape.
Guest analyst The corporate event I use as the cautionary example was a divestiture at an industrial group, and the licensing team found out about it from a press release. Which is more common than you would hope. The transaction was well run by every other measure, and completion was about ten weeks out when somebody in IT asked the obvious question, which was what happens to the four thousand or so people who are leaving with the divested business. And it turned out three separate things had not been considered. The first was that those four thousand seats sat inside an EA commitment that ran for another two years and did not reduce because a division had been sold. The second was that the divested business needed working email from day one, and the transitional services agreement, which had already been drafted and largely agreed by the lawyers, committed the seller to providing IT services for eighteen months without anybody having established who was licensed for that. And the third, which was the one that cost the most time, was that the divested entity had never been separately named in the agreement, so the question of what it was entitled to on its own was genuinely unclear. None of that was anybody's fault individually. The deal team ran a transaction, and licensing is not on a standard diligence checklist. What that group changed afterwards was one sentence in their corporate development process: licensing gets a seat at diligence. That is it. And the next divestiture, two years later, was handled in about three weeks of work rather than five months of argument.
Found out from a press release, and fixed permanently by one sentence in the diligence checklist. Second check.
Check two. You divest a division representing twenty percent of your seats, mid term on an EA. What is the position? A, the commitment reduces automatically with the divestiture. B, it does not reduce on its own, so you open a mid term conversation on the basis that the organisation the commitment was sized against has structurally changed. C, nothing can be done until the renewal. D, the divested entity becomes liable for their share. Pause it, and ask yourself first what your agreement actually says about reduction, and then, separately, what a reasonable counterparty might still be willing to do.
The answer is B. A is the assumption most deal teams make and it is wrong, because a commitment is a promise about spend rather than a description of your headcount, and it does not adjust because part of the business left. But C is equally wrong in the other direction and it is the more expensive belief, because it produces three years of paying for a fifth of an estate you no longer have. A divestiture is actually one of the few genuinely strong cases for reopening mid term: the entity that made the commitment has structurally changed, the vendor keeps you as a customer either way, and the divested business is usually a brand new customer for them, so there is something in it for both sides. D describes something that can be negotiated between the parties in the transaction and is not automatic, so if it matters it belongs in the sale agreement rather than in an assumption. Open early, with numbers, and expect a restructure rather than a release.
Multinational structures, three complications and one thing that helps. Which entity signs: a group agreement signed by a parent and used by subsidiaries needs those subsidiaries covered by it in writing, and estates discover the gap when a subsidiary is audited or divested, which is the worst possible moment. Currency, price lists, and local terms: rates differ by region and price lists move independently, so a global comparison run in one currency at one moment hides that, and a benchmark from one region is not evidence about another. Data residency shapes the estate, because where data must live can force separate tenants or specific service plans for reasons that have nothing to do with cost. And the thing that helps: a group with genuine aggregation is negotiating with real scale, per session twenty nine. The trap is a group that behaves like a federation, where each country negotiates separately while the vendor sees the whole relationship.
Education, public sector, and other families, all with their own rules. Education: separate SKUs, separate pricing, eligibility rules, and the thing to watch is which parts of a mixed organisation actually qualify. Government and GCC: dedicated environments and separate plans, where feature parity is not automatic and needs checking before anybody assumes it. Non profit: discounted and donated offers with conditions attached, where eligibility gets reassessed and status can change. Mixed estates, carrying more than one family in one organisation, where lines get quoted in the wrong family, which is the session ten price sheet check. And charity or trading arms, different entities with different eligibility, which is the entity question from the start of this session arriving again. The recurring error across all of these is quoting in the wrong family, and in a mixed organisation that is a structural risk rather than a clerical slip, because an eligibility question answered wrongly can unwind a lot of purchasing.
Last check. A merger completes in eight weeks and nobody has looked at licensing. What do you do first? A, plan the tenant consolidation, since that is the longest project. B, establish both entity positions: what each holds, what transfers, what each remains obliged to pay, and any open compliance exposure. C, ask both account teams for a combined quote. D, wait until completion, since nothing can change before then. Pause it. Eight weeks is short, so ask yourself which single piece of work everything else depends on.
The answer is B, and everything else genuinely does depend on it. You cannot plan a consolidation without knowing what transfers. You cannot evaluate a combined quote without knowing what each side already owes. And you cannot size the risk without knowing whether either party has an open compliance engagement, which corporate events attract per session thirty one. A is the longest project, and starting it first means designing against constraints you have not established. C invites the vendor to define your position before you have established it yourself, which is the session fourteen error appearing in a new setting. D is simply wrong on the facts, because the pre completion period is when you have the most freedom to shape the outcome and the most access to the other side's information through the diligence process. Start with two register rows, one per entity, and build from there, and you will answer most of the deal team's questions in the first week.
The corporate event method, three things to do the week you hear, because the licensing work always starts late and this is how it starts less late. One, get into due diligence: ask to be included, or at minimum to submit questions to it, because entitlement, agreements, compliance status, and tenant structure are all standard diligence items and licensing is routinely left off the list. Two, build both register rows: your entity and theirs in the session thirty five format, vehicle, entity, term dates, change window, products, owner, which is two rows that answer most of what the deal team will ask you. Three, name the residual obligations: what each side keeps paying for regardless of what the people do, because that number is what stops a migration plan producing a year of double payment and nobody else in the process will produce it. And make sure corporate development knows to call you.
Session thirty nine, three sentences. One: licences attach to a named legal entity under an agreement rather than to a business or a set of people, so when the entity changes the grant does not automatically follow, and that single fact explains every case in this session. Two: acquisitions arrive with an agreement, an estate, a compliance position, a tenant, and a partner, none of which align with yours, while divestitures are harder because your commitment does not shrink when part of the organisation leaves. Three: corporate events attract compliance attention precisely because entitlement rarely survives them cleanly, so get into due diligence, build both register rows, and name the residual obligations before anybody plans a migration. That is the last content session. Next time we put all forty sessions on one table at once.
Homework, about an hour, and this week you prepare for the call you have not had yet. One, find corporate development: who runs transactions in your organisation, and do they know to involve licensing, and introduce yourself before you need to rather than afterwards. Two, list your entities: which legal entities are covered by your agreements, in writing, and most estates find at least one that is ambiguous when they actually check. Three, check the subsidiary coverage: are all the entities using your agreement actually named as covered by it, because that is the gap that surfaces at an audit or a divestiture. Four, draft the diligence questions: the ten things you would want to know about a target's licensing position, ready before any deal starts. Five, identify the families: does your organisation touch education, government, or non profit SKUs anywhere, because mixed estates carry more than one.
Five reads before the capstone, all free on redress compliance dot com. First, Microsoft 365 GCC and government licensing, on the dedicated environments and where feature parity genuinely differs. Second, the higher education software licensing guide, for eligibility, separate SKU families, and mixed institutions. Third, the CIO checklist for licensing in M and A due diligence, which is written for another vendor and whose diligence discipline transfers directly, and it is the best short piece I know on getting licensing into the process. Fourth, navigating Microsoft's shift to CSP and NCE, because that is what an acquired estate is most likely to be sitting on. And fifth, common Microsoft audit findings, for why corporate events attract compliance attention in the first place. Next session is the capstone: an EA renewal, a Copilot expansion, an Azure commitment, and an MCA-E migration proposal, all on the table at once. See you there.