What Foundation, Advanced, and Prime actually contain, and what a forced move onto them costs when it lands at renewal. Three knowledge checks along the way, and 3 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. 3 times in the session the frame splits and a senior licensing analyst gives the view from inside real ServiceNow 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 seven. Last time we established that the April twenty twenty six repackaging forces every customer into a remap, and I gave you the headline rule, map populations to the lowest tier the evidence supports. Today we do the detail underneath that rule. What is actually inside Foundation, Advanced, and Prime, feature by feature, so you can hold your own in the conversation where somebody tells you a tier unlocks something. And then the economics, what a forced migration genuinely costs, layer by layer, and which of those layers you control. Here is the sentence that frames the whole session. A like for like migration is not a flat migration. Even if you map your estate perfectly straight across, changing nothing, using nothing new, it still costs more than it did before, and I will show you exactly why and exactly how much. Three checks, homework, and by the end you will have the runway calendar for your own renewal. Let's go.
Five objectives. First, read the tiers properly, what each one contains and what the step between them actually buys, because those are two different questions and the second one is where the money is. Second, price the migration, building the cost stack of a forced move, the bundled AI uplift, the tier fit error, and the consumption meter all arriving at once. Third, spot uniform Prime, which is the single largest avoidable cost in this whole transition, and learn the argument that removes it. Fourth, write the entitlement map, turning last session's population split into a document the contract actually references, so the remap cannot quietly drift afterwards. And fifth, land the three protections, price protection, the written map, and the uplift cap, all secured before the quote arrives rather than after. That last word, before, is going to come up repeatedly today, because in a migration the timing is not a detail, it is most of the position.
Four numbers, and the first one is the one nobody warns you about. Ten to twenty five percent. That is the uplift a like for like map showed across the migrations we benchmarked. Same estate, same people, same modules, mapped straight across, ten to twenty five percent more expensive, purely because the entry price rose to fund the AI that is now bundled into it. You are buying the AI whether you use it or not, so the honest framing is that bundling did not make the AI free, it made it compulsory. Plus one third. That is what uniform Prime standardization added again on top, for capability most users never touched. Notice these stack, so a badly handled migration is that first number times that second one, and now you are looking at a genuinely serious increase. Nine to twelve. Months before renewal that this work should start, and I will defend that number later because it sounds excessive and it is not. And three, the protections that separated the flat migrations from the shocking ones. Before the detail, let's hear how these land in the room.
Guest analyst clip. The conversation I have most often right now goes something like this. The customer calls and says, we are just doing a like for like renewal, nothing is changing, we do not need help. And I ask one question. Have you seen the number yet. Because like for like does not mean flat any more, and that surprises people who have renewed this platform three or four times without drama. What changed is that the floor moved. The entry tier now carries AI capability that used to be a separate line you could decline, and the price of the entry tier went up to pay for it. So a customer who wants exactly what they had last year, no new features, no AI ambitions at all, still gets a bill ten, fifteen, twenty five percent higher, and there is no product change to point at as the reason. That is genuinely hard to explain to a CFO, and I watch IT leaders try. The ones who handle it well do something specific. They separate the two conversations. Here is the structural increase from bundling, which we can cap but not avoid. And here, separately, is the increase from being mapped onto tiers we do not need, which we are not accepting at all. Blur those two together and you lose both arguments. Split them and you usually win the second one outright.
Separate the two conversations. The structural increase you cap, the mapping increase you refuse. That is the cleanest tactical advice in this session, and it works because it concedes the argument you cannot win in order to win the one you can. If you fight the bundling as unfair, you spend your credibility on a decision made in San Diego that your account team cannot reverse. If you concede it gracefully and then hold the line on tier fit, you are arguing about your own estate, where the evidence is yours. Now, what is actually in these tiers.
Inside the three tiers. Core workflow first, Foundation carries the full service management core, incident, request, change, problem, the catalog. That is not a stripped tier, and I stress that because the name invites you to hear it as the cheap one. Advanced adds process depth and automation capability. Prime adds the agentic layer. Now Assist, row two, bundled in all three, with the allocation growing as you climb. Row three is the row I want you to memorize. The AI platform, Workflow Data Fabric and AI Control Tower, ships in Foundation too. All of it. Prime adds custom AI skill building on top, but the platform itself is not a step up, it is a floor. Row four, autonomous agents, Prime and only Prime, the one genuine capability gate on the ladder. And row five, who belongs, most fulfiller populations belong in Foundation, Advanced is bought when measured usage exhausts Foundation, and Prime is bought for the population that will actually run agents. Read rows two, three, and four together and the ladder becomes clear. You are not climbing for the AI platform, you already have it. You are climbing for autonomy and for allocation. Those are the only two things the step up actually sells you.
So what does each step buy, stripped of the pitch? Foundation to Advanced buys process depth plus a bigger assist allocation. The justification you should require is either measured exhaustion of the Foundation pool, real numbers from your own instance, or named features your teams tried to use and could not. Not a roadmap. Advanced to Prime buys autonomy and custom skills, and this is a capability gate rather than a volume gate, which is precisely why it attaches to a population and not to an estate. The platform is not the step, we have covered that, and I will keep repeating it because it is the most common thing sold to you twice. Headroom is part of the step, and this one is subtle, a meaningful chunk of what the higher tier costs is simply a larger pool, so if your problem is volume rather than capability, buying pool on the lower tier is sometimes the cheaper path to identical arithmetic. Ask for both quotes and compare. And finally, evidence runs both directions. The feature grid that justifies climbing is the same document that justifies falling, and it is your usage record, not their proposal. First check.
Knowledge check one. The account team justifies moving your whole estate to Advanced because it unlocks the AI Control Tower and the Workflow Data Fabric. What is wrong with that argument? A, nothing, those are genuine Advanced features. B, both ship in every tier including Foundation, so they are not a reason to climb. C, they are Prime only, so Advanced would not help either. Or D, they are separate SKUs and must be bought outside the tier. Pause here, and think about which row of that contents table answers it.
The answer is B. Now Assist, the Workflow Data Fabric, and AI Control Tower were bundled into all three tiers in April, and that bundling is exactly what lifted the entry price you are already paying. So selling them as an unlock is selling you something you own twice over. Answer C confuses them with autonomous agents and custom skills, which genuinely are Prime only, and answer D describes the world before April, which is gone. Now, the way you handle this in the room matters. Do not accuse anyone of anything. The correction is polite and specific. Those ship in Foundation, so show me the capability I cannot reach today. That sentence does two things at once, it closes the false argument and it invites them to make a real one, and sometimes there is a real one. You want the real one.
Now the economics, the migration cost stack, five layers, and I want you to watch the right hand column because that is the strategic content of this slide. Layer one, the bundled AI uplift, ten to twenty five percent on a like for like map. Do you control it? Partly. The uplift itself is real and structural, but a cap and price protection bound how far it travels. Layer two, tier fit error, roughly a third again when the estate gets standardized on Prime for simplicity. Do you control it? Entirely. That is your population map. Layer three, the consumption meter, the variable half of the bill arriving for the first time with the migration. You control it, pool size, unit rate, and rollover are all negotiable at signature, and session eight is dedicated to it. Layer four, mapping drift, which is whatever the remap decides your old entitlements were worth if nobody wrote it down. Controlled, by writing it down. And layer five, timing, the leverage you lose when the mapping happens inside the quote instead of before it. Controlled, by starting early. Count the column. One layer is imposed on you and four are decided by preparation. That is why the same migration lands flat for one buyer and expensive for another, and it is not because one of them negotiated harder in the room.
Knowledge check two. Your migration quote comes in thirty percent above current spend. Which line do you attack first? A, the bundled AI uplift, since it is the biggest structural change. B, the tier fit, meaning how much of the estate got mapped to Prime. C, the per user rate, through a deeper discount. Or D, the support and customer success line. Pause here, and ask which layer of that stack you fully control.
The answer is B, tier fit. It is the layer you control completely and the one that moves the most money, because uniform Prime standardization added roughly a third on top of the structural uplift, for capability most of those users will never touch. Answer A is the instinct, go after the biggest number, but the bundled uplift is a decision made above your account team's pay grade, and arguing that it is unfair spends credibility for nothing. You bound it with a cap, you do not win it. Answer C is the classic error, treating a mapping problem as a pricing problem, and it is worse than it looks, because a deeper discount leaves the wrong tier sitting in your contract for the whole term, uplifting annually. And D is rounding error. So the rule for the whole session. Fix the shape of the deal before the rate of the deal. Shape first, always.
The written entitlement map, four parts. Populations, named, each group by function and headcount, service desk, major incident team, approvers, requesters, the people who will run agents. Named groups, not departments, because departments contain mixed populations and that ambiguity gets priced against you. Tier per population, with the feature evidence attached, and note the tone here, Foundation is a positive answer, an evidenced conclusion, not a fallback you are embarrassed about. Old to new lineage, which legacy SKU each population held and what it converts to, and this is the row that stops your prior entitlements being quietly revalued downward during the remap, which is the mapping drift from the previous slide. And pool per population, the assist allocation each group draws, modeled from measured burn, so the meter gets mapped as deliberately as the seats do. Then the crucial part. The map goes into the order form, or is referenced by it, dated and versioned, exactly like the definition freeze from session four. Same discipline, same reason. An unwritten map is not a map, it is a proposal, and proposals get redrawn by whoever is holding the pen at renewal.
Guest analyst clip. I want to tell you about two customers who did the same migration in the same quarter, because the contrast is almost a controlled experiment. Both were mid sized estates, both around four thousand users, both moving off legacy tiers at renewal. The first treated the migration as a procurement event. They waited for the quote, they got a uniform proposal, they negotiated hard on the rate, and they got a respectable discount off a number that was structurally wrong. They are now three years into paying Prime economics for a population that files tickets. The second customer treated it as an evidence exercise. Nine months out they built a map, four populations, tier per population, the legacy lineage for each, and the assist burn modeled per group. When the account team's proposal arrived, they did not negotiate it. They replaced it. They sent their map back and said, price this. And here is the detail people miss, that reframing was not confrontational at all, it was helpful. The account team had built their proposal on no information, because nobody had given them any. Handed real data, they built a better quote, and the deal closed faster than the first customer's did. The difference in annual cost between those two outcomes was larger than either customer's entire negotiated discount. One of them argued about price. The other decided what they were buying.
One argued about price, the other decided what they were buying. And notice the second point in there, the reframing was helpful rather than confrontational, because the account team had genuinely built their proposal on nothing. That is the usual case. Nobody at ServiceNow knows which of your users approve versus resolve, they cannot know, and in the absence of your evidence they will default to the shape that suits them. Supply the evidence and the default disappears. Now, the terms that make the map stick.
Three protections. Price protection across the transition, meaning the remap prices against your current effective rates, per population. Without that sentence the migration is not a migration at all, it is a fresh pricing exercise conducted on their terms with your switching costs visible to everyone. The entitlement map in writing, referenced by the order form, dated, populations and tiers named, which converts your analysis from a proposal into an obligation. And the uplift cap surviving the move, which is session four's cap applied across the transition rather than reset by it. I want to dwell on that one for a second because it is the quietest trap in this whole module. You negotiated a cap, you are proud of it, and then the SKUs change and the new order form starts a new term with new language, and the cap you won two years ago does not carry. Nobody lies to you, nobody hides it, it simply does not come along unless you insist. Check that specifically. And beside those three, the meter terms, pool size, overage rate, rollover, which we build properly next session. The migration is the one moment all of them are on the table together, which is inconvenient and also the opportunity. Third check.
Knowledge check three. Your renewal is four months away, and the migration quote just landed on your desk. What does that timing tell you about your position? A, it is fine, four months is plenty for a mapping exercise. B, you are late, the mapping should have started nine to twelve months out, so you are now negotiating against their map. C, it does not matter, the tiers are fixed so there is nothing to map. Or D, you should delay the renewal until the mapping is finished. Pause, and ask yourself who has already done the mapping by the time a quote exists at all.
The answer is B, you are late. And the reasoning is simply that a quote cannot exist without a map, so if the quote is in front of you, somebody mapped your estate, and it was not you. That is recoverable, but understand what changed. Your job is no longer to propose a map, it is to displace one that already exists and that people have begun to plan around. That is harder, and it is slower, and you are doing it against a clock. Answer A underestimates the work, feature grids and burn data take weeks to assemble properly and you have other responsibilities. Answer C is exactly the passive reading the packaging invites, three tiers, nothing to decide, and it is wrong, the population split is the decision. And D is rarely available, the renewal date is contractual. So the honest summary is this. Late is recoverable, absent is not. If you are four months out, start the map today and accept that you are catching up.
Which brings us to the runway, twelve months to signature, and now I can defend that number. Twelve to nine months out, evidence only, feature usage per team, assist burn per environment, the module one renewal file refreshed. Nothing commercial happens in this phase, no vendor conversation, and that is deliberate, you are assembling facts before anyone can anchor you. Nine to six, draft the entitlement map and price both versions, your split against their likely uniform remap, and the gap between those two numbers is your case, written down, before anyone quotes you anything. Six to three, table the map early, and I mean before their quote frames the conversation, plus open the meter terms alongside the tier terms so they are negotiated as one package. Three months to signature, land the three protections and the pool clauses, and verify the cap survived. Look at that timeline again and notice something. Every single step is work you can do without the vendor in the room. There is no gate, no permission, no dependency on their timeline. Which is exactly why this is available to any buyer willing to start early, and why so few do.
Guest analyst clip. People push back on the twelve month runway constantly, so let me tell you where the time actually goes, because it is not twelve months of meetings. It is about fifteen days of real work spread across twelve months, and the spread is the point. Pulling feature usage per team takes a week, but it takes a week that has to happen after somebody agrees which teams you are measuring, and getting that agreement inside a large IT organisation takes another few weeks of calendar, not effort. Modeling assist burn per environment needs a quarter of actual data, and you cannot compress a quarter. Getting your CIO comfortable with a Foundation heavy map requires a conversation, then a second conversation after they have thought about it, and if you are asking leadership to accept the lower tier for four thousand people while the vendor is telling them AI is the future, that conversation needs runway of its own. And then the account team needs time to take your map to their deal desk, because a non standard structure has to be approved internally on their side too, and deal desks are slow in the last month of a quarter. So when I say nine to twelve months, I am not describing effort. I am describing the calendar that other people's decisions occupy. Start at four months and every one of those steps happens under pressure, which is exactly when buyers accept a uniform map because it is simply the fastest thing to sign.
Fifteen days of work spread across twelve months of other people's calendars. That is the honest defence of the runway, and it reframes the ask. You are not asking your team for a year of effort, you are asking them to start early enough that the decisions in the middle have room to breathe. And note the last observation, under pressure buyers accept the uniform map because it is the fastest thing to sign. Time pressure does not just weaken your negotiating, it changes what you are willing to buy. Let's recap.
The migration, in three sentences. The AI platform layer ships in all three tiers, so the ladder sells autonomy and allocation, and any pitch that climbs you for Now Assist, Data Fabric, or Control Tower is selling what you already own. A like for like map still costs ten to twenty five percent more because the entry price rose to fund bundled AI, and uniform Prime standardization adds roughly a third again for capability nobody uses. And only that bundled uplift is imposed on you, because tier fit, the meter, mapping drift, and timing are all decided by preparation, which is the reason the same migration lands flat for one buyer and expensive for another. Next session we take the meter apart properly. What draws on the assist pool and how fast, why seat based forecasts understate it by construction every single time, and the four contract terms that cap the variable half of your bill.
Homework, about an hour. One, grade your populations, take last session's draft map and attach evidence to each group, the features they actually used last quarter against the tier grid. Two, price the two maps, your population split against a uniform remap at the tier above, and write both totals down with the gap between them, because that gap is the number you will quote to your own leadership, not to ServiceNow. Three, find the reset, check whether your uplift cap language survives a product migration or resets when the SKUs change, and I will warn you now, most reset. Four, date the runway, count backwards from your renewal and mark twelve, nine, six, and three months in the calendar. And five, ask one question, send the account team a written request for the tier feature grid and the pool allocation per tier. Written, because their answer becomes your evidence, and because a request that simple is very hard to decline.
Further reading, five guides. The tier migration guide is today's cost stack in written form, with the price protection language and the planning rule. Foundation, Advanced, and Prime compared gives you the tier contents at full resolution for your homework evidence. The pricing tiers pillar puts the repackaging in context. The twenty twenty six pricing playbook is the whole thing as a document you can hand to your team, which is useful when the person who needs convincing is your own CIO rather than the vendor. And the edition downgrade case study shows the ladder argued downward on feature evidence, which is structurally the same argument your entitlement map makes. That is session seven. Build the map, date the runway, and I will see you in session eight for the meter.