The annuity, the anniversary, the uplift, and what dropping S and S really costs. 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 IBM 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 four, and today is subscription and support, which is the line most IBM customers pay every year without ever deciding to. A perpetual licence gets bought once, and then S and S is paid annually for as long as you own the software, which makes it the recurring half of the estate and the half almost nobody revisits. And here is what makes it worth a full session. Every one of those renewals consolidates onto a single anniversary date in Passport Advantage, which turns that date into the annual negotiation table whether or not anybody sets an agenda for it. In most estates nobody does. Three knowledge checks. Let's begin.
Five objectives. First, read S and S as an annuity rather than as a service line, because that framing is what makes the arithmetic obvious. Second, work the anniversary, since it consolidates every renewal onto one date and that concentration is either your opportunity or their convenience depending entirely on whether you turn up. Third, challenge the uplift using facts your own agreement already gives you, the point level and the consolidation story, neither of which requires you to manufacture anything. Fourth, drop support deliberately and never by accident, which means only on licences carrying a documented no return decision, because reinstatement is priced punitively by design. And fifth, build the anniversary file ninety days out, so the deployed against entitled inventory exists before the renewal quote arrives rather than after it.
So, why does this matter. Four framings. Most, which is the share of estates letting the anniversary pass unworked, accepting the uplift with no negotiation event of any kind taking place. One date, because Passport Advantage consolidates every subscription and support renewal onto a single anniversary, and that concentration cuts both ways. Five to fifteen percent, the consolidation dividend, meaning the better pricing reached by estates routing their purchases through one site number's point level. And by design, which is how lapsed subscriptions reinstate, punitively, and I want to be clear that this is not a penalty aimed at you, it is the mechanism working exactly as intended. Now the note underneath, and it is the finding I would put on the wall. The agreement mechanics quietly priced estates up or down regardless of any negotiated discount, and managing the frame moved more money than the headline discount in most of the estates reviewed.
Guest analyst clip. I want to describe the moment this usually becomes visible, because it is almost always the same moment. Somebody in finance is looking at a multi year view of software spend, and they notice that the IBM number goes up every year even in the years when nothing was bought. And they ask, reasonably, what did we buy. And the answer is nothing, that is support, it renews. Which is true, and it is also the whole problem in one sentence. Because what has actually happened is that a decision made years ago, to buy a perpetual licence, created an obligation that quietly compounds, and no single person is responsible for revisiting it. The purchase had a business case and an approver. The renewal has neither. It has an invoice. And what I find genuinely striking is how often the renewal is larger than any of the individual purchase decisions that were scrutinised at the time. So my advice is not that support is bad value, because for anything running in production it usually is not. My advice is that a line which grows every year deserves the same annual scrutiny a purchase gets once, and in most organisations it receives considerably less.
The purchase had a business case and an approver, and the renewal has an invoice. Hold that, because it explains why this session is about a calendar as much as it is about a price. So let's be precise about what the line actually buys.
Four things. Version upgrades, which is entitlement to new releases of what you already own, and without it you are frozen on the version you hold, permanently. Defect support, the route to a fix when something breaks, and without it you are on self support with no path to an engineering fix. Security fixes, patches for the product you run in production, and this is the reason most estates cannot drop support whatever the arithmetic says. And continuity of entitlement, which is the right to reinstate on ordinary terms rather than punitive ones. Now read that last row carefully, because it is doing different work from the other three. Three of these are product benefits you consume. The fourth is a commercial option you hold. And it is the fourth that decides whether dropping support is a reversible decision or a permanent one, which is the question the whole second half of this session turns on.
Knowledge check one. Your S and S quote arrives six weeks before the anniversary and it is approved because the software is in production. What has just happened? A, a routine administrative renewal with no commercial consequence. B, a year of pricing was set without a negotiation event, which is how most anniversaries pass. C, the uplift was accepted but can be revisited mid year. D, nothing, since support pricing is fixed by IBM policy. Pause here, and ask when the number was actually decided.
The answer is B. Six weeks is inside the window where a quote can only be checked rather than genuinely challenged, because the inventory work that would support a challenge takes considerably longer than six weeks to produce. So the approval is real, the software genuinely is in production, and a year of pricing has just been set with nobody negotiating anything. Answer C is wrong on the mechanics, because the anniversary sets the year and there is no mid year revisit to fall back on. And answer D treats support pricing as policy handed down, when it is a commercial term in your agreement like any other, which is the assumption this entire session exists to dislodge.
The anniversary, five points. Every renewal consolidates onto it, which is the design of Passport Advantage and is genuinely convenient administratively, while concentrating your entire commercial exposure into one event. Which makes it the negotiation table, whether or not anybody treats it as one, and I would put it more strongly than that. An anniversary is a negotiation you are having. The only open question is whether you turned up to it. The file gets built ninety days out rather than six weeks, because the deployed against entitled inventory has to exist before the renewal quote arrives, or else the quote gets trusted rather than checked. It is also the moment other things become negotiable, so Cloud Pak conversion talk lands here where trade in value is real, rather than in IBM's quarter where it is not. And it is the one date you genuinely control, because it is written in your agreement. Put it in a calendar with an owner, because nothing else in this session works without that one step.
Now the uplift, and the good news is that your own agreement hands you two arguments. The point level, because everything you buy accrues points toward your pricing level, so consolidating purchases under one site number improved pricing by five to fifteen percent across the estates reviewed. The consolidation story, because if you have been routing purchases through several enrolments then pulling them together is a fact about your own buying behaviour that improves your position. Both of those are already true today. Neither requires a competitive threat, a migration plan, or anything you have to construct, which makes them unusually clean arguments. The uplift is a term rather than a rate card, negotiated like anything else, and the estates that treat it as fixed are precisely the estates that pay it compounding year after year. And then the last point, which matters as much as the rate. An uplift applies to whatever support you are carrying, including support on software that nobody has run for two years.
Guest analyst clip. The point level is the piece of Passport Advantage that customers most often do not know they are sitting on, and it is worth understanding because it is entirely mechanical. Everything you purchase under the agreement accrues points, and those points set your pricing level, and the level applies to what you buy next. Straightforward enough. Where it goes wrong is organisational rather than commercial. A large company acquires another company, or a division buys independently, or a region sets up its own enrolment because that was easier at the time, and now you have three site numbers. Each one accrues points separately. So an organisation that is, in commercial reality, one large customer, presents to the pricing engine as three medium sized ones, and gets priced accordingly. Nobody did anything wrong at any point in that story. And the fix is not a negotiation, it is an administrative consolidation that happens to be worth somewhere between five and fifteen percent. What I would say to anyone who has grown by acquisition is go and count your site numbers before your next anniversary, because that number is quite often greater than one and almost nobody has checked.
Go and count your site numbers, because an organisation that is one large customer commercially can present as three medium ones. That is an administrative fix worth five to fifteen percent, which is an unusual ratio of effort to return.
Knowledge check two. You have bought through three separate enrolments for years. What does consolidating them under one site number do? A, nothing, discounts are negotiated per purchase. B, accrues points toward one pricing level, worth five to fifteen percent in the estates reviewed. C, only simplifies administration. D, triggers a new agreement and resets your terms. Pause here, and ask what the agreement does with everything you buy.
The answer is B. Points accrue across everything bought under the enrolment and they set your pricing level, so fragmented buying quietly prices you as several smaller customers rather than as the one large customer you actually are. Answer C names a completely real benefit and misses the commercial one entirely, and I flag it because that is the usual reason consolidation gets deprioritised. It goes on somebody's list as an administrative tidy up, competing with work that has a number attached to it, and it loses. Put the five to fifteen percent next to it and it stops losing.
Dropping support, and this is the decision that is only cheap if it turns out to be permanent. Five points. The saving is immediate and real, because support on software you no longer run is pure waste and it is carried in most estates simply because nobody was ever asked to look. The cost is optionality, since you are giving up the right to reinstate on ordinary terms, and that right is worth considerably more than it looks on the day you drop it. So the test is a documented no return decision, meaning you drop support only where somebody has written down that this software is not coming back, with their name against it. Archive the entitlement records either way, whatever you decide, because entitlement you cannot evidence is entitlement you will eventually be asked to buy a second time. And the security question usually settles it, because for anything sitting in production the patch route is the real constraint, whatever the arithmetic says about the support line.
Guest analyst clip. The documented no return decision sounds like bureaucracy and it is the single control that separates a good support drop from an expensive one, so let me explain what it actually does. When somebody proposes dropping support to save money, the analysis is usually correct. The software is not being used. The saving is real. What is missing is a view about the future, and specifically about whether this product could plausibly come back, through a project restarting, an acquisition bringing it in, or a regulatory requirement resurfacing. And the person proposing the saving is very rarely the person who would know. So the discipline is simply to get the answer written down by somebody who owns the application, not the licence. That is it. And the reason it matters so much here rather than with other vendors is the reinstatement pricing, which is deliberately unattractive, so the cost of being wrong is not the support you saved, it is considerably more than the support you saved. I have seen organisations drop support to save a modest sum and reinstate eighteen months later for a multiple of it, and in every one of those cases the information that would have prevented it existed somewhere in the building.
Get the answer written down by somebody who owns the application rather than the licence, because the person proposing the saving is rarely the person who would know. Now, why being wrong is so expensive.
Reinstatement, priced punitively and deliberately so. Five points. It is not a penalty aimed at you, it is priced to make lapsing unattractive, which is the mechanism working as designed rather than a negotiation you lost. Which means the drop is close to one way, so treat every support drop as irreversible when you model it, and if the case only works while assuming cheap reinstatement then the case does not work. The trap is the accidental lapse, because a renewal missed through administration rather than through decision costs you exactly the same as one dropped on purpose, and it happens a great deal more often. So the calendar is a commercial control, and the anniversary date with a named owner is what separates a decision from an accident, and it costs nothing at all to put in place. And partial drops need the same discipline, because dropping support on part of a product family is still a decision per licence, and each one needs its own documented answer.
A short but important detour, because there are two frames and they behave completely differently. Passport Advantage gives you points, a pricing level, a global site enrolment, and a single anniversary date, and it is built for an organisation that buys continuously. Express drops the points and drops the single anniversary, replacing them with per order buying where each item renews twelve months from its own purchase date. Who each fits, then. Express fits an occasional purchaser genuinely well, and it quietly overprices anyone buying continuously, because nothing accrues toward a level. And the renewal shape differs completely, because per item renewal dates mean there is no anniversary to work, which means the negotiation event this entire session is built around simply does not exist for you. So go and check which one you are on. It is on your paper, most people have never looked, and the answer changes what every other lesson in this module is worth.
Knowledge check three. A business case to drop support assumes you can reinstate cheaply in eighteen months if the product comes back. What is wrong? A, nothing, reinstatement is a standard process. B, reinstatement is priced punitively by design, so a case resting on cheap reinstatement does not hold. C, only that eighteen months is too long. D, reinstatement is impossible once support lapses. Pause here, and ask what the mechanism is actually designed to discourage.
The answer is B. The pricing exists precisely to make lapsing unattractive, so a model that assumes an easy return has quietly assumed away the mechanism it is supposed to be accounting for. Answer D overstates it in the other direction, because reinstatement is genuinely available and that is rather the point, it is simply expensive. So the rule I would give you is to model every support drop as irreversible, and if it only works with a cheap return then do not drop it. That single test would have prevented most of the reinstatement conversations I have watched, and it costs nothing to apply.
Guest analyst clip. If I had to reduce this whole session to one recommendation, it would be to put three dates in a calendar, and I recognise how underwhelming that sounds for a session about a large annual number. Here is why it is the answer. Everything expensive in subscription and support happens because of timing rather than because of price. The uplift is accepted because the quote arrived too late to challenge. The lapse happens because nobody was watching the date. The consolidation never happens because there was no moment that forced the question. None of those are pricing failures, they are calendar failures, and they are the kind of failure that recurs annually until somebody fixes the calendar. So the three dates are the anniversary itself, ninety days before it, which is when the file starts, and the notice or cancellation date if your paper has one. Put a name against each. What I find is that organisations who do that one thing start having a genuine annual negotiation within a cycle, not because anybody got better at negotiating, but because for the first time they are in the room while the decision is still open.
So here is the file, five items, ninety days before the anniversary. Deployed against entitled, built before the renewal quote arrives so that the quote gets checked rather than trusted, and this is the item everything else depends on. The shelfware decision, with support stripped only where a no return decision is documented, and entitlement records archived either way. The uplift challenge, using the point level and the consolidation story, which are the two facts the agreement itself provides you. The conversion conversation, with Cloud Pak trade in value discussed at the anniversary where it is real rather than in IBM's quarter where it is not. And release language, demanded on any renewal being priced under audit shadow, because a renewal signed under that shadow ought to close it rather than leave it open. Five items. None of them requires a tool you do not have, and all five need somebody to start ninety days out.
Three sentences. Subscription and support is the recurring half of an IBM estate, it consolidates onto a single Passport Advantage anniversary, and most estates let that anniversary pass unworked, which locks in another year of uplift without any negotiation event taking place at all. The two arguments against the uplift are already sitting in your own agreement, the point level and the consolidation story, and consolidating fragmented purchasing under one site number was worth five to fifteen percent in the estates reviewed. And dropping support saves real money while spending optionality, because reinstatement is priced punitively by design, so support comes off only where a no return decision is documented and the entitlement records are archived either way.
Homework, about an hour. Find your anniversary date, which is on your agreement, and put it in a shared calendar with an owner plus a reminder ninety days ahead of it. Check the frame, Passport Advantage or Express, and if you are on Express while buying continuously then that is a finding worth raising on its own merits. Count your enrolments, meaning how many site numbers your organisation actually buys through, because more than one is a consolidation conversation worth five to fifteen percent. List support on unused software, anything you are paying to support that nobody has run for a year, and do not drop anything yet, just write the list. And read last year's uplift, what it actually rose by and whether anybody challenged it, and I will predict now that the answer to the second half is no.
Five guides. The Passport Advantage guide has the agreement mechanics in full, the points and pricing levels, the single anniversary, the Express comparison, and the five item agenda we just walked through, so it is the one to read this week. Passport Advantage negotiation covers how the anniversary is actually worked and what moves when you arrive with an inventory rather than with an objection. IBM subscription licensing, how it works, sets the subscription model against perpetual plus support and shows when each is genuinely the cheaper shape. The IBM subscription licensing guide covers the mechanics underneath the annuity including what lapse and reinstatement do to a position.
And IBM third party support in 2026 is the honest assessment of the alternative to the support line, including what you give up and what that is worth, which is worth reading before anybody in your organisation proposes it rather than after. Next session we build the entitlement baseline, the Passport Advantage reports, bundled and restricted use entitlements, and the things the portal does not tell you. See you there.