One position, assembled once, and argued from for years. 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. Session ten, and this closes module two. Over the last four sessions we have measured a great deal: the user catalog, the reclassification evidence, the phantom records, and the engine consumption. All of that currently lives in different places, produced by different exercises, understood by different people. Today we put it into one document. The entitlement baseline. What you bought, what you use, the difference between those two things line by line, and the evidence behind every figure. I want to be direct about why this matters more than it sounds. Every negotiation in the second half of this course, all of module eight, the renewals, the conversions, the audit defence, assumes you have this document. Without it, each of those conversations starts with somebody frantically assembling numbers under time pressure. So today: the four parts, the six steps to build it, the five properties that make it survive a challenge, how to read your own gap, and how to keep it alive. Three knowledge checks. Let's start.
Five things by the end. First, name the four parts: entitlement, consumption, the gap and the evidence, and understand why a baseline missing any one of them is not actually a baseline. Second, build it in order, six steps, starting from the order forms rather than from any report, and finishing with somebody's signature on it. Third, make it hold, which is the five properties that let a position survive a challenge two years after the person who built it has left. Fourth, read your own gap: there are four possible positions against entitlement, and each one is telling you to do something different. And fifth, keep it alive, because a baseline is either a living document or a historical one, and the difference between those is about an hour a quarter.
Four things to frame this. One doc: entitlement, consumption, gap and evidence in one place, so that nobody has to reassemble it under time pressure while somebody waits. Two weeks: what it costs to build the first time, assuming the underlying work from sessions five to nine is done, and almost all of that fortnight is collection rather than analysis. Years: how long a good baseline serves. Every measurement, every renewal, every audit and every conversion argues from it. And first: whoever states a position first sets the frame. If SAP tells you your number before you tell them, you are responding rather than proposing, and that is a materially worse place to be even when your numbers are better than theirs. So the framing for the whole session: this is not a compliance artefact. It is the negotiating position, written down before anybody asks for it. Let's play a clip on that.
Guest analyst clip. I want to explain why I care so much about this one document, and it is not really about compliance. It is about who speaks first. Picture two conversations. In the first, SAP arrives with a number. Here is what our records show, here is your consumption, here is the gap. Everything you say after that is a response. You are checking their arithmetic, questioning their definitions, asking for time to verify. Even when you are completely right, you are the party reacting, and the frame belongs to them. In the second conversation, you arrive with a dated, sourced, internally approved position. Here is our entitlement from the order forms, here is our consumption with the method for each figure, here is the gap line by line. Now they are responding to you. And notice what has actually changed: nothing about the software, nothing about the numbers, possibly not even the eventual outcome. What changed is who is explaining themselves. In my experience that is worth more than most of the technical arguments people spend their preparation time on. And here is the part that surprises people. Building it is mostly collection, not analysis. If you have done the measurement work properly, the baseline is roughly two weeks of assembling things that already exist into one place, with a date and a signature on it.
What changed is who is explaining themselves. That is the whole argument for this session in one sentence, and I would put it in front of whoever has to approve the two weeks of effort, because it lands better than anything about compliance. Notice also the point about collection rather than analysis. That matters for who does the work. The hard thinking happened in sessions five to nine. What is left is assembling, sourcing and dating, which is a different and much more delegable kind of task. So let's look at what goes in it.
Four parts, and three of them are close to useless on their own. Entitlement: what you bought, taken from the order forms and amendments. Never from a report, never from memory, and never from a spreadsheet somebody maintains by hand, because all three of those are summaries of the paper and summaries drift. Consumption: what you use, per named user type and per engine, measured or derived, with each figure carrying the method that produced it. The gap: entitlement minus consumption, line by line, with a sign on it. Both directions matter, and we will come back to that, because shortfall is exposure and surplus is money you are already paying. And the evidence: the dated backing for every figure above. Order form references, report outputs, decision notes, derivation methods. Here is the test of whether you have a baseline. Entitlement without consumption is an inventory. Consumption without entitlement is a report. Neither one is a position, and only the gap tells you what to do next.
Six steps, in this order, the first time through. Step one, collect the paper: every order form and amendment, oldest to newest, into one library. Entitlement lives there and nowhere else, and any gap in the paper is itself a finding. Step two, extract entitlement: one line per licensed item, with quantity, metric and price list version. That is the spine, and every later figure hangs off one of those lines. Step three, measure consumption: users by type and engines by metric, using everything from sessions five to nine. Step four, calculate the gap: line by line, signed, both directions, and do not round small numbers away, because small surpluses are money and small shortfalls become large ones. Step five, attach the evidence: every figure gets a source, a method and a date. Step six, have it approved: a named owner signs it and it becomes the organisation's position rather than one team's analysis. Now, the practical warning. Steps one and two take the longest, and everybody wants to start at step three, because measuring feels like progress and reading old contracts does not.
First knowledge check. Your entitlement figures and SAP's records disagree on one licensed item. What is the correct first move? A, adopt SAP's figure, since it comes from their system of record. B, go back to the order forms and amendments and see what the paper says. C, split the difference for planning and resolve it at the renewal. D, use your own figure, since you built the baseline from your own documents. Pause here and pick one.
The answer is B, go back to the paper. The signed documents are what bind both parties, and an internal system of record on either side is a summary of those documents. A treats a summary as authority, and I have seen those summaries wrong in both directions, including in the customer's favour. D is the interesting one, because it sounds like the confident, self reliant answer. It is actually the same mistake as A with the loyalties reversed: your extract could equally be the one carrying the error, particularly if it was typed from a PDF by somebody working quickly. And C is the worst of the four, because it plans deliberately on a number that is definitely wrong. It is the only option guaranteed to be inaccurate, and it feels reasonable, which is a dangerous combination.
Five properties that let a baseline survive a challenge. Sourced: every figure points at a document or a rerunnable report, and nothing in it exists because somebody remembered it. Dated: the whole document carries an as at date, and so does each figure, because a baseline with no date cannot be compared to anything, including its own previous version. Reproducible: somebody else could rebuild it from the same sources and reach the same numbers, and that is the actual test rather than a nice property. Owned: one named person, approved internally, not the licensing team as a collective noun but a person with a name. And complete: every licensed item appears, including the ones with a comfortable gap, because a selective baseline reads as advocacy rather than analysis and invites exactly the scrutiny you were trying to avoid. Let's play a clip on the third of those, because it is the one people underestimate.
Guest analyst clip. There is one test I apply to a baseline, and it is harsher than it sounds. Could somebody else, who was not involved in building it, rebuild it from the same sources and arrive at the same numbers? Not approximately. The same numbers. Most baselines fail that test, and they fail it for a very human reason. The person who built it knew things. They knew that this engine figure excludes the Brazilian subsidiary because of a decision taken in 2019. They knew that this user count is measured on the first working day of the quarter rather than the last. They knew which of the two spreadsheets called entitlement is the real one. None of that is written down, because when you know something it does not feel like it needs recording. And then that person moves to another role, and what remains is a document full of numbers nobody can reproduce. I would call that a historical record rather than a position, because the first time somebody challenges a figure in it you will not be able to defend it. So the discipline is to write down the thing that feels too obvious to write down. Every figure gets a source, a method, and a date. Yes, it is tedious. It is also the difference between a document that works for five years and one that quietly expires the day its author changes job.
Write down the thing that feels too obvious to write down. That is a good general rule and it is a specific one here, because the obvious things are precisely the ones that carry the exclusions. Every baseline I have seen has two or three quiet decisions embedded in it: a subsidiary that is out of scope, a measurement date that is not the one you would assume, a system that was deliberately excluded. Those are not errors. They are usually correct decisions. They just become indefensible the moment the person who made them is not in the room.
Second knowledge check. You are fifteen percent under entitlement on a large user type. What does the baseline tell you to do? A, nothing, since being under entitlement is the safe position. B, treat it as a live asset: shelfware to reduce, or headroom to trade at renewal. C, reclassify users upward to use the entitlement you already paid for. D, remove it from the baseline, since there is no exposure to report. Pause here before you continue.
The answer is B. A surplus is not neutral. You paid for it, and you are paying maintenance on it every year whether anybody uses it or not. It is also the most useful thing you can bring to a renewal, because unused entitlement is a genuine concession you are able to offer. A confuses safe with free, which is the most expensive confusion in this topic. C is worth naming explicitly because I have seen it proposed seriously: never grow consumption to justify a purchase you regret. That converts a recoverable overbuy into a permanent one. And D is the selective baseline problem from the previous slide arriving one slide later, and it also destroys the only record that the surplus exists at all.
So, four possible positions and what each one tells you. Comfortably under: that is shelfware, and you are paying maintenance on capacity nobody uses, so reduce it at renewal or trade it for something you do need. Close to entitlement: the healthy state, and the one that needs the most watching, so track the crossing date quarterly and buy on your own timing. Just over: a real exposure and usually a cheap one to close if you act early, so verify it is real and then apply the levers before you buy anything. And substantially over: that is a commercial conversation, and it is happening whether you start it or somebody else does, so verify and then choose your moment and your framing deliberately. Notice that only one of those four is a compliance problem. The other three are commercial information, and all four of them are completely invisible without the baseline. Let's play a clip on the half of this that nobody reads.
Guest analyst clip. Everybody looks at the shortfalls. Almost nobody looks at the surpluses, and I think that is the single biggest waste in software asset management. Here is the psychology. A shortfall is frightening. It has a name, compliance risk, and somebody might be blamed for it, so it gets attention immediately. A surplus feels like prudence. We bought a bit extra, we are safely covered, nothing to worry about. Except a surplus is not free and it never was. You paid for it, and you are paying maintenance on it every single year, and in most estates the total value sitting in unused entitlement is larger than the total value of the shortfalls. It is just quieter. Now here is the part that matters commercially, and it is the reason I would look at the surplus half first. Unused entitlement is the most useful thing you can bring to a renewal. It is a genuine concession you are able to offer. You can hand some of it back in exchange for something you actually want, or you can trade it against the volume you are short on elsewhere. Neither of those conversations is available to you if the surplus is not written down, because you cannot trade an asset you have not counted. So when you read your baseline, read both columns with the same attention. One of them is your exposure. The other one is your currency.
One of them is your exposure, the other one is your currency. Hold on to that, because it reframes what this document is for. And that leads to the four things the baseline does that nothing else does. It answers the measurement: when the annual request arrives the work is already done, so you are reviewing a position rather than discovering one under a deadline. It prices the renewal: you cannot negotiate what you have not counted, and every ask and every concession gets measured against these lines. It survives an audit: a dated, sourced, approved position changes the character of the conversation from investigation to verification, which is a different meeting with different people in it. And it costs the conversion: moving to S/4HANA or RISE reprices every line in this document, and you simply cannot model that move without it. Four different audiences, one document. That is why it earns the effort, and it is also why it should not live in one person's spreadsheet.
Five ways a good baseline stops being true. It was never dated, so nobody can tell whether it describes today or the year it was built, and in the absence of a date people assume stale. A purchase was not added: new entitlement arrives, the order form gets filed, and the baseline still shows the old quantity, which means you are understating what you own. The price list version changed: same metric name, different wording, so the definitions your figures were built against no longer apply and nobody was told. The owner left, which is the commonest cause by a wide margin, and we will come to that. And it became a project artefact: built for one renewal, filed with the project, never touched again, and rediscovered three years later by somebody who quite reasonably does not trust it. Notice that four of those five are maintenance failures rather than errors. The document was right when it was written, which is exactly what makes the decay hard to see.
Last knowledge check. Your baseline is fourteen months old and there has been one acquisition since. How much of it do you trust? A, all of it, since the underlying contracts have not changed. B, the entitlement, once you add the acquired contracts. The consumption needs re-measuring. C, none of it: rebuild from scratch, because an acquisition changes everything. D, the consumption, since it was measured. The entitlement is the uncertain part. Pause here, and think about which of the four parts actually ages.
The answer is B. The four parts age at completely different rates, and that is the practical point of this check. Entitlement changes only when somebody signs something, so it is stable, and updating it is cheap. Consumption moves continuously, so at fourteen months it is a historical record rather than a current one. A ignores the acquisition, which is precisely the event that adds entitlement nobody in your team has read yet. C throws away the expensive, durable half of the document in order to avoid rechecking the cheap, volatile half, which is the wrong way round and also costs you two weeks. And D has it exactly backwards, treating the stable part as uncertain and the volatile part as settled.
So, an hour a quarter and it never has to be rebuilt. Refresh consumption quarterly: the user counts and the engine register you are already running, dropped in, gaps recalculated. Update entitlement on signature: any new order form updates the baseline the week it is signed, not at the next review, because that is the failure we just looked at. Re-read definitions annually, when your price list version changes, checking every metric your figures depend on. Re-approve yearly: the owner signs it again each year, and the real purpose of that is not the signature, it is that it forces somebody to actually read the document. And hand it over properly, which is the one I want to spend a moment on. Let's hear why.
Guest analyst clip. The most common way a baseline dies is not neglect. It is a handover. Somebody built it, understood it completely, kept it current for three years, and then moved to another role. Their successor inherits a document, opens it, and finds a hundred and forty lines of numbers they did not produce and cannot yet defend. So what happens next is entirely predictable. They keep it as reference and quietly stop trusting it. Six months later somebody asks a question about a figure, the new owner cannot answer, and the decision gets made to rebuild it properly at some point. That rebuild takes another two weeks, and three years of accumulated context is simply gone. So here is what I would do, and it takes half a day. When the owner changes, the successor rebuilds one section from source. Not the whole thing. One section. They go to the order forms, extract the entitlement lines themselves, rerun the consumption method, and check that they land on the numbers already in the document. If they do, they now trust the whole thing, because they have proved to themselves it is reproducible. If they do not, you have just found a real problem at the cheapest possible moment. Half a day, at the one point in a baseline's life where it is most likely to be abandoned.
They keep it as reference and quietly stop trusting it. That is such an accurate description of what happens, and notice that nobody does anything wrong in that story. The successor is being sensible. They will not defend numbers they cannot reproduce, and they should not. The half day rebuild fixes it because it converts inherited numbers into verified ones, and it does so at the exact moment when the cost of doing it is lowest and the cost of not doing it is highest.
Three sentences. Entitlement, consumption, the gap and the evidence, in one dated document, because any three of those without the fourth is not a position. Build it from the signed paper upward rather than from a report downward, and have a named person approve it, which is what stops it being a draft. And it ages unevenly: entitlement changes only on signature while consumption is historical within a year, so refresh the half that moves and leave the half that does not. That closes module two. You now have both axes measured, the optimization work done on each, and one document that states where you stand. Module three opens next session with the topic that has cost SAP customers more in headlines than any other, and which does not appear anywhere in the numbers you have just assembled: indirect access.
Homework before session eleven, about an hour. One, find the order forms. Every one, oldest to newest. If you cannot assemble a complete set, that gap is the most important finding in this session and it outranks everything else on the list. Two, extract ten lines: ten licensed items with quantity, metric and price list version, which is enough to prove the format works before you commit to doing all of them. Three, fill in what you have: consumption for those ten lines from the last four sessions of work, and leave blanks where you genuinely have nothing rather than estimating. Four, sign the gap: plus or minus for each line, and look at the surpluses as hard as the shortfalls, because that is the half nobody looks at. And five, name the owner. Who signs this document? If the honest answer is nobody, you have just found the reason the baseline does not exist yet.
Five guides, all on redresscompliance dot com. The SAP licensing guide is the whole map in written form, which is the context this baseline sits inside. Establishing an internal compliance program covers who owns the baseline, who approves it, and how the quarterly refresh gets governed, which is the half of this session that is organisational rather than analytical. The licence optimization guide is what to do with the gaps once the baseline makes all of them visible at the same time. The audit readiness strategy guide treats the baseline as the core artefact of readiness and shows how the annual rehearsal from session five uses it. And the audit survival guide covers what a dated, sourced, approved position actually changes about the conversation when somebody does come asking.
That is session ten, and that is module two complete. Ten sessions in, you have the map, both axes, the paper, the measurement, the optimization work and the position. Next time we open module three with indirect access: the history, the cases that made it famous, and why it is the exposure that does not show up in any user count you have just built. See you then.