Everything in this course, run against a single organisation. Three knowledge checks along the way, and 4 clips from a senior licensing analyst.
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. 4 times in the session the frame splits and a senior licensing analyst gives the view from inside real SAP 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, and welcome to the last session. Twenty nine sessions of method, and today we run all of it against a single organisation, from the first export to the signed agreement. The estate is a composite, drawn from real reviews, and it is deliberately unremarkable, because most estates are. What I want you to see is the sequence: what gets measured, what the measurement finds, what survives contact with the business, what gets tabled, and what actually moves. Including the parts that were wrong, because a first pass always contains some. Three knowledge checks, then your first ninety days. Let's begin.
So, what this session does. It runs one estate end to end, a composite organisation, from first measurement to signed agreement. It applies the method in order: measure, find, position, negotiate, operate, each step drawing on earlier sessions. It shows the findings honestly, including the ones that were wrong. It states the outcome plainly, so what moved, what did not, and which parts mattered most. And it leaves you a first ninety days: five things to do, in order, starting the week this course ends.
The estate, then, and it is a shape you will recognise. A manufacturer of moderate size, eight years into Salesforce, renewing in nine months. Two clouds, Sales and Service, several thousand users, grown by acquisition and by hiring. A data layer, with Data Cloud bought two years ago and an Agentforce pilot running in Service. The portfolio: MuleSoft, Tableau, Slack and Marketing Cloud, each bought separately at different times. And no single view, meaning four renewal dates, nine metric types, and nobody who could describe all of it. Let me be clear about which of those is the actual problem.
Guest analyst clip.
No single view is the condition being treated. Everything that follows is a consequence of nobody holding the whole picture, and notice that it is an organisational problem rather than a Salesforce one.
Month one: measure. Four exports and the questions they answer. Users by type, from the org rather than the order form, answering who holds what and who has actually used it. Consumption by meter, so credits, conversations, sends, API and storage, answering what the slope is and when each one runs out. Capacity, meaning environments and reserved cores, answering reserved against used and what is actually exercised. And the contract set, every order form and referenced document, answering quantities, dates, uplift language and definitions. Four exports and a week of reading. Nothing on that list requires a tool, a consultant, or a single conversation with the vendor, and that is worth sitting with, because it is where the whole position comes from.
First knowledge check. Where should the user count come from? A, the order form, which is what you are billed on. B, the org, so you can compare entitlement against reality. C, the account team's report. D, the HR headcount system. Pause here and pick an answer before you continue.
B. The whole exercise is the comparison, so you need both sides of it, and only the org tells you what is assigned and what has actually been used. A is one half, and it is the half you already know. C is useful as a cross check and it is their measurement, so starting from it concedes the baseline. And D is how organisations end up licensing headcount rather than users, which was the first failure named back in session eighteen and it recurs everywhere.
So, what the measurement found. Five findings, in the order they surfaced. Dormant users, around a tenth of assigned licences with no sign in for a quarter, mostly leavers. Tier drift on Tableau, with three times as many Creators as people who had actually prepared a data source. A consumption slope, with Data Cloud tracking to exhaust its commitment two months before term end. Idle capacity, two non production environments running production redundancy and exercised rarely. And no uplift cap anywhere in the agreement, because nobody had ever asked, so nobody had ever refused. None of those required special insight. They required somebody to look. Now, what happened when the findings met the business.
Guest analyst clip.
So, month four, what survived. The dormant users held, uncontested, because leavers have nobody to defend them, and that is the free reduction. Tableau partly held, with managers correcting roughly a quarter of the list, which is what made the rest credible. And the consumption finding changed shape entirely: not a reduction at all, but a schedule change and headroom bought early rather than late.
One environment was defended and correctly kept, because it absorbed the quarterly regression run, and the other one went. And the uplift cap became the priority, because once you model it across the term it is worth more than everything above it combined.
Second knowledge check. Managers rejected a quarter of the Tableau downgrade list. What does that mean? A, the measurement was unreliable and should be redone. B, the process worked, and the surviving three quarters are now defensible. C, the reduction should be dropped. D, the managers were protecting their teams. Pause here before you continue.
B. Corrections are the entire point of the manager review, because a list that survives challenge from the people who know the work is exactly what you cannot be argued out of later. A treats normal exceptions as failure, when parental leave and quarterly working patterns are real and the data cannot see them. C throws away three quarters of a good finding because a quarter of it was wrong. And D is sometimes true and it is not the useful reading, because the corrections you actually receive tend to be specific and checkable rather than defensive.
Month nine, the negotiation, and what was tabled in what order. The measurement first: assigned against active, by type, with the method stated, and it was not contested. The reduction, named, meaning specific licences agreed internally with managers' corrections already applied. The structural asks: uplift cap, reduction band, definitions on the order form, co-termination. The growth they wanted, which was the Agentforce expansion, real, and the thing to trade with. And all of it timed to their year end, because the renewal date had been moved eighteen months earlier, deliberately. Let me say what that ordering does.
Guest analyst clip.
Measurement first, then the reduction, then the structure, then the thing they want. The order matters, because each step establishes the credibility the next one draws on.
So the outcome, honestly. The uplift cap was agreed, and it was the single largest item, and it cost almost nothing to ask for. The reduction was applied: dormant users in full, Tableau tiers in part, roughly as prepared. The reduction band was refused and deferred to the following renewal, because not everything lands the first time and pretending otherwise would be dishonest. Definitions went on the order form, which cost nothing and quietly removed a whole category of future argument. And the discount barely moved, which mattered least, and which is the point this entire course has been making from about session two onward.
What happens next, because a renewal is not an ending. The calendar starts immediately, with quarterly numbers from month one rather than from month twenty seven. The refused ask is logged, so the reduction band goes into the file as the opening item next time. The note is written that week, while the reasoning is fresh and the concessions still make sense. Consumption is watched, because the Data Cloud slope did not stop, it was simply funded properly, and it will need managing all term. And it compounds, because the second renewal starts from a position rather than from nothing, which makes it easier than the first and the third easier still.
Last knowledge check of the course. Which single item delivered the most value in this estate? A, the dormant user reduction. B, the uplift cap. C, the Tableau tier correction. D, the improved discount. Pause here and pick an answer before you continue.
B. The cap applies to the whole agreement every year for the rest of the relationship and it compounds, while the reductions are one time corrections to a quantity, valuable and finite. A and C are real money and they are the visible work, and they are also what made the room take everything else seriously, so please do not hear this as saying they did not matter. And D moved least and would have been reported most enthusiastically, which is precisely the habit worth breaking. Let me pull the whole method together.
Guest analyst clip.
Five things, in order, forever. Measure what you have, from your own systems, quarterly, in a format that does not change, because the format staying identical is what makes a trend visible. Know what you signed: quantities, dates, definitions and the documents referenced, read once a year. Watch what moves on its own, because consumption and outcome metrics rise without anybody deciding, so give them an owner. Decide before they quote, meaning the term sheet, the ranking and the walk away, agreed with your sponsor. And trade for what lasts: caps, bands, definitions and dates, because the discount is the smallest thing on the table and it is the only thing anybody celebrates.
Three sentences to finish. Salesforce is not complicated because the products are hard, it is complicated because nine different metrics behave in four different ways and almost nobody in your organisation holds the whole picture, which is the condition this course has been treating throughout. Everything useful starts with measuring your own estate, because a number you produced from your own systems is the only thing in the room that cannot be dismissed, and every position worth holding is built on top of it. And the discount is the smallest item on the table, so spend your preparation on the terms that compound, the rights that let you act on what you measure, and an operating rhythm that makes the next renewal duller than this one.
Your first ninety days, five things, in this order. Week one, the four exports: users by type, consumption by meter, capacity, and every contract document. Week two, find the notice date and put it in a system with an owner and an alarm nine months before it, because that single act changes more than anything else on this list. Month one, build the portfolio page: every line, its metric, its family, its direction and its owner, and the blanks are your work programme. Month two, take the free reduction, meaning leavers and dormant accounts, because nobody defends them and it funds your credibility for everything harder. And month three, find the uplift clause, and if there is not one, you have just identified your most valuable renewal ask.
Where to go from here, all on redresscompliance dot com. The Salesforce licensing knowledge hub collects every guide referenced across these thirty sessions. The Salesforce licensing pillar hub is the structured reference by topic, for looking things up. Salesforce licensing tools has free calculators for the arithmetic in this course. Salesforce licence optimization covers the reduction work end to end when you are ready to run it. And Salesforce licensing experts is there if you want independent, client side help and another pair of eyes.
And that is the course. Thirty sessions, from the edition ladder to the term sheet. If you take one thing from all of it, make it this: measure your own estate, because everything else in this course is built on a number you produced yourself, and almost nobody does it. Thank you for staying with it through thirty sessions. Go and pull the four exports this week, and good luck with your renewal.