The price sheet, the SKU names, the bundles, and the arithmetic to run before any conversation about discount. 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 ten of forty, and this one closes module two. Over the last four sessions we have taken the Enterprise Agreement apart: the enrollment, the annual order and the true up, the metrics, and Software Assurance. Today we put it back together in the form you will actually meet it, which is a quote. Here is why this session exists. Every buyer I work with can describe their agreement in general terms, and a surprisingly small number can read the document that prices it. The quote arrives late, it is long, it is professionally formatted, and it carries an implied instruction to approve rather than to examine. So the reading gets delegated, or skipped, and the negotiation that follows is about a percentage rather than about what is on the sheet. By the end of the next twenty minutes you will be able to open a Microsoft price sheet cold and say what you are buying, what it costs per person per month, and which three lines you intend to change.
Five takeaways. One, the anatomy: a quote is three documents pretending to be one, only one of them governs what you owe, and knowing which is the first defence. Two, the price sheet, column by column: what each field says, what it hides, and the two columns that almost nobody checks. Three, the SKU names: decoding part numbers so you can state what you are buying without asking the account team to interpret it for you. Four, bundle arithmetic: pricing a bundle against its parts, and identifying the component you are paying for because it travels with the ones you wanted. Five, the five calculations: the arithmetic that has to exist before anyone says the word discount, because a discount on the wrong basket is not a saving, it is a better price on something you did not need. Those five are the session, and the homework at the end applies all of them to a quote you already have.
Three documents pretending to be one. The proposal deck: slides about transformation, a savings claim, and a comparison against a baseline you did not choose. That deck binds nobody. It is also, in most organisations, the only part an executive reads, which means its savings number is the one you will be asked to defend at board level. The price sheet: part numbers, quantities, unit prices, price level, coverage dates. This is the quote. Everything you are agreeing to buy is here, and it is normally the least discussed document in the room. Third, the agreement papers: the enrollment, the amendments, and the product terms, referenced rather than attached. Those decide what the price sheet means. What a licence permits, what happens when counts move, what renews and at what. Now notice the pattern. The order of attention in most organisations is the exact reverse of the order of importance. Read the price sheet, then the terms that govern it, and treat the deck as the vendor's opinion of your business.
The price sheet, column by column. Part number: the exact SKU. Check it matches the product you believe you are buying, including edition and add on status. Description: the human readable name, and description and part number can disagree, in which case the part number wins every time. Quantity: seats, cores, or devices, and the questions are which metric and whose count. Compare with your own admin centre figure rather than accepting theirs. Price level: A to D on an EA. Is the level right for your seat count, and is it held for the term. Unit price: per year or per month, and net or pre discount, and those two ambiguities between them explain a lot of confused meetings. Coverage dates: start and end, which is where co termination lives, and where stub periods hide real cost. Extended total: quantity times unit price, and it should reconcile. Two columns get skipped almost every time, price level and coverage dates. A wrong price level quietly overprices every line, and a stub period makes a year on year comparison meaningless without anybody having lied to you.
First check. The proposal deck claims a nineteen percent saving against last year. The price sheet shows total spend rising. Which is more likely to be true? A, the deck, because it is prepared from Microsoft's own data. B, the price sheet, because it is the only one of the two that lists what you are buying, and the deck's saving is measured against a baseline that was chosen rather than observed. C, neither, they measure different things and cannot be compared. D, both, because the saving is against list and the rise is against last year. Pause it. Before you answer, ask yourself what baseline that nineteen percent is measured from, and who selected it.
The answer is B, and D is worth spending a moment on because D is the answer you will be handed in the room. A saving against list price is not a saving. Nobody pays list, and the list figure was set by the party quoting it, so a discount against it measures generosity rather than value. Your cash position is decided by extended totals on this sheet against extended totals you paid last year, on comparable coverage periods. That is arithmetic. It goes up or it goes down. A fails because Microsoft's own data does not convert a chosen baseline into an observed one. C is the interesting wrong answer, because it sounds sophisticated and it is the most expensive option on the slide. It accepts that the two documents cannot be reconciled and therefore abandons the analysis precisely where it becomes useful. They can be reconciled. Rebuild the deck's savings claim from the price sheet, line by line. Either it survives contact with the sheet, or you have just found the first negotiating point of the renewal, twelve months before you needed it.
Decoding the SKU names, five habits. Base or add on: almost every line is one or the other, and an add on carries a prerequisite base licence, so an add on line implies a base line somewhere. Edition and tier: the letter and number carry the tier, and descriptions get shortened in decks and in email, which is how an E3 conversation becomes an E5 order without anybody deciding it. Step up SKUs: their own part numbers, moving an existing licence up a tier for the remainder of the term, priced differently from a fresh licence, which is why the same upgrade appears at two very different numbers depending on how it is quoted. Government, education, and non profit: separate SKU families, separate prices, separate eligibility, and mixed estates carry more than one family. And the suffix that carries the term: annual, monthly, and multi year commitments are different SKUs of the same product, and the flexibility difference between them is worth real money that never appears in the description.
Guest analyst The quote I think about when this topic comes up was, on the face of it, a win. European manufacturer, mid sized estate, and the account team had come back with a rate improvement of about four points over the previous term. The procurement lead was pleased, and honestly he had earned it, because he had pushed hard and the improvement was real. He asked us to sanity check before signature. So we did what we always do, which is retype the price sheet into our own spreadsheet and add one column: cost per user per month. And the four point improvement was there, exactly as described. But cost per user per month had gone up by nine percent. Nobody had misled anybody. Two things had happened underneath the percentage. List prices had moved, so the same discount was being applied to a bigger number. And the mix had shifted, because a step up had been folded into the renewal and several thousand seats were now sitting a tier higher than they had been. Both were visible on the sheet. Neither was visible in the percentage. What I remember is his reaction, which was not annoyance at the vendor. It was annoyance at himself for negotiating the ratio for four months instead of the number. We reopened it on cash per seat, corrected the mix against their own persona data, and the deal that eventually signed was worth about six percent less in real money than the one he had been about to celebrate.
A four point better rate and a nine percent worse cost per seat, both true at the same time. Negotiate the number, not the ratio. Second check.
Check two. A quote lists four thousand seats of a security add on and three thousand one hundred seats of the base plan that add on requires. What is your first move? A, approve it, add ons are commonly overbought and the extra will be used. B, ask which nine hundred users hold the add on without the prerequisite base, because either nine hundred lines are unusable, or nine hundred base licences exist elsewhere and the estate is being counted in two places. C, reduce the add on to three thousand one hundred and move on. D, raise the base to four thousand so the numbers match. Pause it, and notice as you think that C and D both fix the appearance of the problem without ever finding out what caused it.
The answer is B, and the reason is that the gap of nine hundred is the most useful number on the entire sheet. Both tidy answers throw it away. Follow it instead. If those nine hundred users hold their base plan under a different vehicle, a CSP subscription or a separate enrollment, you have just found a fragmented estate and a co termination problem worth considerably more than this line item. If they hold no base plan at all, you are being quoted nine hundred add ons that cannot be assigned to anybody, which is a straightforward correction and a straightforward saving. D is the expensive reflex: raising the base to match buys nine hundred base plans in order to justify a number that may itself be wrong, and it is remarkable how often that is the path of least resistance in a busy quarter. C is cheaper and still wrong, because it closes the question. And A is how estates accumulate add on shelfware that survives three consecutive renewals, on the assumption that growth will absorb it. Growth rarely absorbs anything. Growth adds.
Bundle arithmetic, and I want to be fair here, because bundles are not automatically bad. Unpriced bundles are. Three steps. Price the parts: list every component and price each one standalone at your own discount. If the bundle beats that total, the bundle is genuinely cheaper and you should take it. If it does not, you are paying a premium for packaging, and the conversation becomes which components you actually want. Find the passenger: most bundles carry one component nobody asked for, riding along with the ones you did want. Name it, value it, put it on the table. A passenger you can name is a negotiating item. A passenger you cannot name is just cost with a story attached. Check the duplication: bundled components frequently duplicate something you already license elsewhere, and duplication only becomes savings if you decommission the incumbent, where decommission means a signed plan with a date on it rather than an intention expressed in a meeting. The test: a bundle is worth its premium only if you would have bought the passenger separately.
The five calculations, and none of these need a tool. Unit cost per user per month: extended total divided by users divided by months, and it is the only figure that compares cleanly across quotes, across vehicles, and across years. Effective rate against list, per line rather than blended, which is your real discount as opposed to the number in the deck. Year on year per seat, on the same coverage length, which tells you whether the deal improves your position or simply grows with you. Cost of the mix change: reprice this year's seat mix at last year's rates, and the difference between that and this year's total separates price movement from volume movement. And shelfware adjusted cost: divide by assigned and active users rather than licensed users, which gives you what you pay per person who actually uses the thing. That fourth one changes meetings. It turns an argument about whether the deal is good into two smaller and far more answerable questions, and an analyst can run it in an afternoon.
Last check. Your quote shows a deeper discount than last term and a cost per user per month that is eleven percent higher. What happened? A, nothing, a deeper discount always means a better deal. B, the discount is measured against a list price that rose, or against a richer mix, so the percentage improved while the cash per seat went up, and only the per seat calculation settles it. C, the quote contains an error and should be sent back. D, the extra cost is the price of the deeper discount. Pause it. A discount is a ratio, and ratios move when either number moves.
The answer is B, and this is the single most common way a good looking Microsoft renewal ends up costing more, with nobody misleading anybody at any point. List prices move, and the twenty twenty six waves moved them. Mix moves too, because a step up from a lower tier to a higher one raises cost per seat even at an improved percentage. Both can be true simultaneously, and both are visible in about twenty minutes using the third and fourth calculations from the previous slide. A is the belief that turns discount percentage into the negotiating objective, which is precisely the objective a vendor would choose for you if given the choice. C is premature, because the sheet is probably arithmetically correct and it is the interpretation that is wrong. D treats the two figures as a trade, and they are not a trade, they are two independent movements that happen to arrive in the same document. Negotiate the cash per seat, and let the percentage land wherever the arithmetic puts it.
The quote reading method, three steps, none of which require the account team. One, rebuild the sheet. Not a copy, a rebuild: retype it into your own spreadsheet with your own formulas for extended totals and per seat cost. Half of what you need to know gets discovered in the act of rebuilding, because that is the moment lines that do not reconcile stop reconciling in front of you. Two, price the alternatives. The same basket under the other vehicle, and the same basket with the mix corrected to your actual personas. Two comparison columns beside the quote, which is the difference between negotiating a price and choosing between priced options. Three, write the one page. Per seat cost, year on year movement, the split between price and volume, and the three lines you intend to change. On a large estate the whole method takes an analyst two to three days, and measured against a three year commitment that is the highest return work anybody in procurement does all quarter.
Session ten, three sentences. One: a quote is three documents with very different weight, and the price sheet is the one that decides what you owe, so the deck's savings claim is an argument to be tested against the sheet rather than a finding to be reported upward. Two: read part numbers rather than descriptions, because base, add on, step up, and term are all carried in the SKU, and mismatches between related lines are the most useful information on the sheet. Three: discount is a ratio and cash per seat is a number, so run the five calculations first and let the percentage land where the arithmetic puts it. That closes module two. You now have the Enterprise Agreement end to end: how it is structured, how it counts, how it meters, what maintenance buys, and how it is priced on paper.
Homework, about an hour, and this week you rebuild a quote you already have. One, find the sheet. Not the deck, the price sheet from your most recent order or renewal, and if nobody in procurement can produce it within a day then that is finding number one and it is worth raising on its own. Two, rebuild it. Retype it with your own formulas, confirm the extended totals reconcile, and note every line that does not. Three, add the per seat column: cost per user per month for every product line, then sort it. I promise the order will not be the order you expected. Four, split price from volume. Reprice this year's quantities at last year's unit prices, and the difference between that and this year's total is price movement. Everything else is you. Five, name three lines you intend to change at the next renewal, with the reason and the number, and keep that page. It is your opening position, written twelve months early.
Five reads before next session, all free on redress compliance dot com. First, how to evaluate a Microsoft renewal proposal, which walks cost, terms, and bundled services line by line. Second, benchmarking Microsoft EA discounts, on what a competitive rate actually looks like and why the percentage is the wrong target to aim at. Third, the CIO level playbook on evaluating renewal proposals across EA, MCA, and CSP, which is step two of the method: the same basket priced three ways. Fourth, Microsoft 365 add ons and duplicate cost, for where bundle components quietly duplicate something you already license. And fifth, common Microsoft licensing mistakes, most of which are visible on a price sheet long before they become audit findings. Next session opens module three and the part of the estate everyone has an opinion about: the Microsoft 365 map. E1, E3, E5, the frontline F tiers, and Business. What each one includes, who each one is genuinely for, and the upgrade paths between them. See you there.