The three year term, the enterprise wide commitment, the two product tiers, and the true up that decides what growth costs. Three knowledge checks along the way, and 1 clip from a senior cloud advisor.
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 two of forty. Last week we established that the vehicle decides more than the discount. Today we open the first of the three vehicles, the Enterprise Agreement, which for most large organisations is still the container everything else sits inside. Here is the trade in one sentence: an EA sells you three years of price certainty, and it buys that certainty with a commitment that only ever counts upward. Understand that asymmetry and most of what follows, the true up mechanics, the two product tiers, the placement decisions, the reason a shrinking business finds an EA painful, all of it becomes obvious rather than surprising. Today: the anatomy, the two tiers and why they are two negotiations, the true up in detail, what price protection actually holds, and an honest fit test for when this vehicle is the right one. Let's open it up.
Five takeaways. One, the anatomy: master agreement, enrollment, three year term, annual order, and what enterprise wide commitment actually obliges you to do, because that phrase does more work than any other in the document. Two, the two product tiers: Enterprise Products and Additional Products, sitting on separate discount frameworks with separate removal rights, which is one agreement containing two negotiations that most buyers run as one. Three, the true up: how growth gets counted and billed, and the reconciliation that should happen before the number is submitted rather than after. Four, price protection: what the three year hold genuinely gives you, what it does not, and what two price waves and compressed tiers did to the value of it. And five, the fit test, stated honestly, including the estate shapes where the answer is that you should be looking at a different vehicle entirely. This session is mechanics, and mechanics is where the money is.
The anatomy, six elements, and I want you to notice that they are layers rather than one document. The master agreement: the legal frame, use rights, audit, transfer, the general terms, rarely opened in a negotiation and worth opening anyway for the assignment and audit language, which module eight will use when a corporate event arrives. The enrollment: the commercial container, products, price level, term, quantities, and this is where your pricing actually lives, so when someone says the EA, ask which layer they mean. The three year term: the enterprise wide commitment at negotiated pricing, and the protections and caps you write here are inherited by everything that follows, which is why signature day matters so much more than it feels like it does. The annual order: growth reported each anniversary, billed at contracted rates. Then the two product tiers, Enterprise Products, Office, Windows, the CAL suites, carrying the enterprise wide commitment and the compressed discount levels. And Additional Products, everything riding alongside on separate frameworks with, and this is the part worth writing down, separate removal rights. The structural point: an EA is a frame plus an enrollment plus an annual order, each layer offering different things you can change, and a buyer who treats it as one document negotiates only the layer they were handed.
The two tiers, and why this is one agreement containing two negotiations. Enterprise Products: Office, Windows, and the CAL suites, licensed enterprise wide across a qualified user or device count. This tier carries the volume discount levels, the ones that compressed through 2024 and 2025, and it is where the tier mix work from module three does its damage or its good. Additional Products: server products, add ons, the newer subscriptions, all riding alongside on their own discount frameworks, negotiated individually, and crucially carrying removal rights at anniversary that the enterprise wide tier does not give you. Now the sentence that makes this practical: the two tiers behave differently when you want to reduce. A product sitting inside your enterprise wide commitment is stuck for the term. The same product placed as an Additional Product may not be. And placement, in many cases, is a negotiable decision made at signature by someone who was thinking about discount rather than about exit. So the least glamorous question in the whole negotiation, and one of the few that changes what you can do in year two: which tier does each material line sit in, and why. Ask it before signature, get the answer in writing, and place deliberately.
First check. Your EA is signed for four thousand qualified users. A restructure moves six hundred people out of the business in month seven of year one. What happens to your bill? A, it reduces at the next anniversary when you report the lower count. B, it reduces immediately once the accounts are deprovisioned. C, nothing changes: the enterprise wide commitment holds for the term, the true up counts growth upward only, and reductions wait for the renewal. Or D, Microsoft issues a credit for the unused portion. Pause here. What direction does a true up travel? And what is the phrase enterprise wide doing inside the commitment?
The answer is C, and this single asymmetry explains most of the regret I have seen attached to Enterprise Agreements. You commit enterprise wide for three years. The annual true up exists to bill your growth at contracted rates, and it has no downward equivalent, which means headcount reductions, divestitures, cancelled projects, and quiet attrition all sit on your bill until the term ends. Six hundred people leave in month seven and you pay for them for another twenty nine months. A describes a downward true up that simply is not in the standard structure, and it is the single most common misunderstanding I meet in a first conversation. B confuses a technical act, deprovisioning, with a commercial one, because your licence count is contractual rather than observed. And D imagines a refund culture that no volume agreement in this industry has. The consequence for how you buy is precise: an EA rewards accuracy about your floor rather than optimism about your growth, because growth is genuinely cheap to add later at your locked rates, while shrinkage is impossible to remove before the renewal. Size the commitment to the population you are confident will still be there in year three.
The true up, five points, and this is the operational heart of an EA. The mechanism: at each anniversary you report added qualified users and added products, and they bill at the rates you negotiated at signature for the remainder of the term. The count is yours: you report it, which makes accuracy your responsibility and, more usefully, your opportunity, because an unreconciled count is almost always an over count. The reconciliation: assigned licences against real people, leavers removed, service accounts and test tenants identified and explained, done before the number is submitted rather than discovered afterwards. The timing trap: a peak that happens to land near your anniversary can commit you for the rest of the term, so seasonal populations, project teams, and contractor waves deserve to be timed deliberately rather than accidentally. And the calendar: this work starts sixty to ninety days before the anniversary, because done in the final fortnight it is a submission, and done early it is a negotiation input. Our guest analyst has watched the timing trap cost real money, so let's hear it.
Guest analyst The most expensive fortnight I have ever seen a client spend was a true up nobody planned. A logistics business, Enterprise Agreement, about six thousand qualified users at signature. Their anniversary fell in early October. Their peak season, the ramp for the Christmas period, started in late September, and that year they onboarded roughly eleven hundred seasonal workers, every one of them provisioned with a full licence because that was the standard build. The true up was submitted in the second week of October by someone in IT operations who, quite reasonably, reported what the admin centre showed. Seven thousand one hundred users. Those eleven hundred seasonal people left in January. The commitment did not. They paid for eleven hundred phantom users for the remaining twenty six months of the term, and at their contracted rate that was a little over two million. Now, what makes this worth telling is how cheaply it could have gone differently. Two things would each have solved it. Reconciling the count against actual employment status before submitting, which would have flagged that eleven hundred of those users were fixed term. Or simply provisioning seasonal staff on a lighter licence, which their own policy already allowed and nobody had enforced. The lesson I give every client now: your anniversary date is a commercial event, not an administrative one, and if it sits anywhere near your busy season, that is not bad luck, it is a number you can negotiate at renewal. Move the anniversary. It costs nothing to ask.
Eleven hundred seasonal workers reported at an October anniversary, gone by January, paid for until the end of the term. Two million, and either a reconciliation or a lighter build would have prevented it. And the closing point is the one to take away: your anniversary date is a commercial event, and if it sits on your busy season, that is negotiable at renewal. Second check.
Check two. Preparing your true up, you find three hundred and forty assigned E3 licences whose users have not signed in for over ninety days: leavers never removed, plus two test tenants. The correct action: A, report them anyway, assigned licences are licences. B, reclaim and deprovision before the count is submitted, so the true up reports the real qualified population rather than an administrative artefact. C, report them and ask Microsoft for a credit next year. Or D, exclude them silently and keep no record. Pause here. Once a number is reported at the anniversary, what does it become for the rest of the term?
The answer is B. A reported count does not merely bill this year, it becomes the baseline for the remainder of the term, and since check one established there is no downward true up, those three hundred and forty licences get paid for repeatedly rather than once. Reclaiming before submission is ordinary hygiene with an outsized commercial payoff, and I want to be clear that it is entirely legitimate: you are reporting your qualified population accurately, which is exactly what the agreement asks of you. A pays for administrative debt as though it were headcount, and administrative debt compounds quietly in every estate I have ever reviewed. C leans on the credit mechanism that does not exist. And D reaches the right number by the wrong route, because exclusions need a documented basis, both so the count is defensible if anyone reviews it later, and so that next year's team knows what was done and why rather than rediscovering the same three hundred and forty accounts from scratch. Session thirty two builds that evidence discipline properly. For now: reclaim, document, then report.
Price protection, which is the EA's central promise, and it deserves a precise reading. What is held: per unit pricing for the products on your enrollment, for the term, which means growth added at true up bills at your signature rates rather than at whatever list price has become since. In a rising market that is genuinely valuable, and the market has been rising. What is not held: the renewal. Protection ends with the term, and the renewal reprices against the market as it stands at that moment, which is precisely why the renewal programme in module six starts twelve months out instead of when the quote arrives. And what changed: two price waves inside twenty four months compounding roughly eleven to nineteen percent against the prior cycle, plus volume tier discounts on the enterprise products compressing through 2024 and 2025 for another four to nine percent. Read those together carefully, because the conclusion is not the obvious one. The protection held. It did exactly what it promised. What got more expensive was renewing it. So the buyer's reading is this: a three year hold is worth more in a rising market, not less, and the moment where the real money moves is not your signature, it is your renewal.
When the EA is right, stated honestly, because a course that only sells you one answer is not worth your time. Large stable seat base: strong fit, and the number from session one applies, above roughly two thousand four hundred stable users the EA rate typically wins by six to fourteen percent. Predictable growth: strong, because true ups bill new people at your locked rates, which makes growth genuinely cheap inside an EA. Volatile or seasonal population: weak, and you have just heard two million dollars of why. Shrinking headcount: weak, because you pay for departed people until the renewal regardless of anything you do. Heavy Microsoft standardisation: strong, since an enterprise wide commitment matches how that estate actually runs. And mixed or migrating estates: it depends, and the right answer is often a deliberate split, the stable base on an EA and the volatile population somewhere with monthly flexibility, which session five builds into a proper framework. Notice what the fit test is actually about: shape, not size. A six thousand seat company with a large contractor population can be a worse EA candidate than a three thousand seat company with a flat one. Size gets you the discount tier. Shape decides whether the vehicle fits.
Last check. You are signing a new EA. Which single structural decision made now will most change your options in year two? A, the headline discount percentage on Enterprise Products. B, which tier each material product sits in, plus the reduction and swap rights written at signature, because those decide what you can shed when the estate changes. C, the choice of licensing solution provider. Or D, the payment schedule, annual versus upfront. Pause here. In year two, the business changes. What determines whether the agreement can change with it?
The answer is B. The discount sets what you pay for what you committed to; placement and reduction rights decide whether that commitment can move when the business does. A product inside the enterprise wide tier is fixed for the term, full stop. The same product placed as an Additional Product may carry anniversary removal rights, and swap language, where you can negotiate it, turns a wrong guess into an adjustment instead of three years of shelfware. A matters, genuinely, and it is also recoverable at the next renewal, which structural placement is not. C affects your service quality and contains a margin worth negotiating, but it does not change your structural options. D is a treasury question with a real cash value and no effect whatsoever on flexibility. And here is the rule that will recur in every single module of this course, so let me state it plainly: the terms that decide what you can undo are worth more than the terms that decide what you pay. Payment recurs annually and can be renegotiated. Flexibility, once surrendered at signature, is gone for the term.
The EA file, which is what every EA holder should be able to produce in about ten minutes, and most cannot. The paper: master agreement, every enrollment with its term dates, the current price sheet, the amendments, and the answer to which tier each material product sits in, written down rather than held in somebody's memory or, worse, in a departed colleague's memory. The counts: qualified users and devices as reported at each true up, set against assigned licences and active users per tier today, because the gap between what you reported and what is actually used is your reclamation opportunity and it is usually larger than anyone expects. And the calendar: anniversary dates, the reconciliation starting sixty to ninety days before each one, the term end, and the renewal programme opening twelve months out, every entry with a named owner. If that file does not exist in your organisation, understand what that means practically: at the next renewal, the only complete picture of your estate in the room belongs to the vendor, and every number in the conversation will be theirs. Next session, the Microsoft Customer Agreement, the evergreen alternative Microsoft is steering enterprises toward, and the protections that quietly do not come with it.
Session two, three sentences. One: an EA is a frame plus an enrollment plus an annual order, carrying two product tiers on separate discount frameworks with separate removal rights, and blending them into one conversation gives away structure that Microsoft's own paperwork keeps carefully apart. Two: the true up counts upward only, so the number you report becomes a floor for the rest of the term, which makes the reconciliation before submission one of the highest return hours available anywhere in the agreement. Three: the three year price hold is genuinely valuable in a rising market and it ends at the renewal, which is where the money actually moves, and why module six starts twelve months early. Next week: the Microsoft Customer Agreement. See you there.
Homework, about an hour, and this week you reconcile one true up. One, pull the reported counts: what you declared at the last anniversary, per product and per tier, and if nobody kept a copy, request it from your licensing solution provider, which is a reasonable ask and tells you something about the relationship if it is difficult. Two, pull today's assignments: assigned licences per tier from the admin centre, with sign in activity for the last thirty to ninety days beside them. Three, find the artefacts: leavers still assigned, test tenants, service accounts, duplicates, then count them and price them at your contracted rate, because a number with a currency symbol in front of it gets attention that a number alone does not. Four, check the placement of your three largest non core products: enterprise wide tier or Additional Products, and note which ones you would want to shed if you could. And five, diary the anniversary minus ninety days with an owner, because that single calendar entry is what converts next year's true up from a submission into a decision. One hour, and you will very likely find more than the hour cost.
Five reads before next session, all free on redress compliance dot com. First, the Enterprise Agreement pillar, the structure and the benchmark data in reference depth, including the price wave and tier compression numbers we used today. Second, the EA true up guide, the counting mechanics step by step with the reconciliation calendar laid out. Third, Microsoft licensing true ups and how to avoid costly mistakes, which catalogues the specific errors that turn a routine true up into a permanent cost, our logistics client's among them. Fourth, EA discount negotiation levers, on what genuinely moves on the enterprise products tier. And fifth, the Microsoft licence reclamation guide, the hygiene that pays for itself before every anniversary. That is session two. The frame, the two tiers, the true up that only counts up, and a price hold that matters most in the market we are actually in. Next week, the evergreen alternative. See you there.