A migration accepted as a formality is a renewal negotiated by the vendor alone, and the mapped quote carries a 10 to 25 percent uplift before anybody uses anything new. 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 thirty three, and today is the negotiation that every ServiceNow customer now faces whether they asked for it or not. The tier migration. Legacy SKUs reached end of sale on the first of July 2026, which means every renewal from here migrates onto Foundation, Advanced, and Prime. And it will be presented to you as a mapping exercise. Administrative. A formality. Somebody will say we have mapped your current entitlements across and here is the quote. I want to give you the sentence that should be in your head when that happens, because it is the whole session. A migration accepted as a formality is a renewal negotiated by the vendor alone. Nothing improper takes place. The mapping just gets done by the party whose interests it serves, because the other party did not turn up. Three checks, homework, let's go.
Five objectives. First, treat it as a contract event rather than a product update, which makes it a full renegotiation. Second, separate the two costs, because a like for like map showed a ten to twenty five percent uplift purely from the AI now bundled into the entry tier, before anybody touches any new capability at all. Third, refuse uniform Prime, because standardizing the whole estate on the top tier added roughly a third again, for capability most of your users will never open. Fourth, demand the entitlement map, a written mapping of every legacy entitlement to its new tier equivalent, so that nothing is quietly upgraded a tier during the move. And fifth, price the meter that arrives with it, because migration introduces the consumption model for the first time, and buyers who model only the tier price miss the entire variable half of their new bill.
Four framings. The first of July 2026, legacy SKU end of sale, which is why every renewal from here migrates whether or not you asked for it. Ten to twenty five percent, the uplift a like for like map showed, purely from the AI baked into the entry tier, before any new capability was used by anyone. A third again, which is what uniform Prime standardization added on top of that, for capability most users never touched, and that is the single largest avoidable line in the whole move. And both at once, because the tier change and the consumption meter arrive together, so a model that prices only the tier misses the variable component entirely. The note underneath is the one to hold. A migration accepted as a formality is a renewal negotiated by the vendor alone, and I want to stress that nothing improper happens here. The mapping is simply done by the party whose interests it serves, because the other party treated it as paperwork.
Guest analyst clip. The word formality does an enormous amount of work in these conversations and I have come to treat it as a signal rather than a description. When somebody tells me a step is a formality, what they usually mean is that it is routine for them, and that is often perfectly true. The migration is routine for the account team. They have done it dozens of times this year. It genuinely is, from where they sit, a mapping exercise with a well understood process. The problem is that routine for them and low stakes for you are two completely different statements, and the word formality quietly merges them. I reviewed a migration where the customer had accepted the mapped quote in about a week, because it was described as a formality and they had other priorities that quarter, which was reasonable. The increase was somewhere around thirty percent, and when we decomposed it afterwards, roughly half of that was a uniform Prime assumption nobody had ever discussed out loud. Nobody hid anything. The quote said Prime. It was just never presented as a decision, and so it was never made as one. So my rule is simple. Any time a step is described as a formality, that is the moment to ask what decisions are being made inside it, because formality means nobody expects you to look.
Formality means nobody expects you to look, so that is precisely the moment to ask which decisions are being made inside it. Let's decompose the number, because it is four costs and only one of them is on the quote.
Four costs. The bundled AI uplift, which comes from the entry price rising to fund Now Assist being bundled into the base, and that one is not avoidable, but it must be named rather than absorbed silently into a total. Uniform Prime, from the whole estate being mapped to the top tier for simplicity, and that is avoidable and it is the single largest avoidable cost in the move. Mapping drift, individual entitlements landing a tier above their legacy equivalent, avoidable with a written entitlement map. And unmodelled consumption, the assist meter arriving alongside the packaging and getting priced at your first true up, avoidable by modelling it before signature rather than after. Read that right hand column as a set, because it is the encouraging part of this session. Only the first is genuinely unavoidable, and even that one is negotiable in how it is phased and capped. The other three are entirely a function of how carefully the move is negotiated, which means they are yours to fix.
So, uniform Prime, and I want to be fair to the argument. It is proposed for simplicity, and one tier across the estate genuinely is easier to administer, quote, and explain, so that is a real argument rather than a trick. And it is the largest avoidable cost, roughly a third again on top of the bundled uplift, for capability most of your users will never open. The gate that justifies it is real too, because building net new custom AI agents is gated to Prime alone, and that is a genuine reason for some population and not for the entire estate. Which turns the whole thing into a question about population rather than principle. How many people build agents? That number is your Prime population, and everybody else has a cheaper correct answer. And that makes blended the default ask, a tier model priced per population rather than one badge applied to everybody because it is tidy. Notice that you are not arguing that Prime is wrong. You are arguing about how many people need it, which is a much easier argument to win.
Knowledge check one. Your migration quote maps every fulfiller to Prime. Forty people build custom AI agents. What is the counter? A, accept, Prime is the current top tier and simplifies administration. B, a blended model, Prime for the forty who build agents and the correct lower tier for everybody else. C, refuse Prime entirely and stay on Advanced. D, negotiate a discount on the Prime rate across the estate. Pause here, and ask what actually requires Prime, and for how many people.
The answer is B, the blended model. Building net new custom AI agents is gated to Prime, which is a real requirement for those forty people and no requirement whatsoever for the rest, so a blended model pays for the capability exactly where it is used. Answer C is the overcorrection, and it loses a genuine need, and the business will reverse it within a quarter and you will have spent credibility for nothing. Answer D is the reflex this course keeps interrupting, because a discount on the wrong tier for the wrong population is a discount on a mistake, and remember the scale here, uniform Prime added roughly a third again over the bundled uplift. You cannot discount your way out of that. You have to not buy it.
Now the written entitlement map, and I would argue this single document does more work in a migration than any discount you negotiate alongside it. What it says, every legacy entitlement you hold today mapped line by line to its named new tier equivalent, at a named rate. What it prevents, quiet upgrades, because without it individual capabilities drift a tier upward during the mapping and the drift only surfaces on an invoice months later. Name functions rather than just SKUs, so if Walk-up Experience, Process Mining, or DevOps Change Velocity are in production today, they get named as entitlements you keep whatever tier they now sit in. And when to ask, which is before you accept the migration quote, because protection agreed after signing is worth very little and the map is the cheapest thing on the entire list to obtain. Session twenty nine asked for this as part of a five year price book lock. Here it is the minimum for any migration at any term length, because the packaging change happens to you regardless of how long your term is.
Knowledge check two. Your account team says the migration is a formality and the mapped quote reflects your current entitlements. What do you ask for? A, nothing, a like for like map should be flat. B, the map in writing, line by line, plus the arithmetic showing where the increase comes from. C, a bigger discount on the total. D, a delay until the packaging settles. Pause here, and ask that if it is genuinely like for like, why the number is higher.
The answer is B, the map plus the arithmetic. A like for like map is not flat, because the entry price rose to fund the bundled AI, so a ten to twenty five percent uplift is entirely expected and it should be stated openly rather than absorbed into a total. Answer A is the assumption that leaves the increase unexamined, and it is a reasonable assumption which is exactly why it is dangerous. Answer C negotiates the total without knowing its composition, which is session twenty six's decomposition problem showing up in a new setting, and you will negotiate hardest on the component that was never going to move. And answer D is simply not available to you, because legacy SKUs went end of sale on the first of July 2026, so the migration is happening at this renewal whether the packaging feels settled or not.
Three protections, all agreed before you accept the quote, plus two notes on timing. The written entitlement map, documented line by line with functions named, so nothing is quietly upgraded and you have a reference point that survives the account team rotating. An uplift cap across the term, year over year, so that the migration is a one time step rather than the first move of a climb you then have to negotiate again in twelve months. And the consumption terms, pin the pool, pin the overage rate, add rollover, because the meter arrives with the packaging which is exactly why it belongs in this negotiation rather than the next one. Then do it nine to twelve months out rather than at the quote stage, which is module six's calendar applied to this specific event. Because after is worth little, and every one of these three is cheap while your signature is unspent and close to worthless the day after you have given it.
Guest analyst clip. I want to make the case for asking for the entitlement map even when you are fairly confident the mapping is fine, because people skip it on the grounds that it feels slightly distrustful and their account team has been good to them. Here is why I ask anyway, every time, and it has nothing to do with trust. The people doing your mapping today will not be the people you deal with in two years. Account teams rotate, deal desks reorganise, and the person who understood why your Walk-up Experience entitlement was treated a particular way will have moved on. What survives is the document. So the map is not primarily a check on the current team, who are usually doing their best with a large and genuinely complicated remap. It is a record for the version of this relationship that exists in year three, when somebody asks why you are entitled to something and nobody currently employed remembers. I have seen customers lose entitlements they had absolutely been promised, not through bad faith, but because the promise lived in a conversation and the conversation lived in somebody's head and that person went to work somewhere else. Ask for the map. Frame it as wanting a clean record, which is true, and I have never seen anybody refuse it when it is put that way.
The map is a record for the version of this relationship that exists in year three, when the promise lives in somebody's head and that person has left. Frame it as wanting a clean record, which is true. Now let me put the whole thing together as a model.
The blended tier model, which is the practical output of this session, and the pleasing thing is that it is built almost entirely from work you have already done in this course. Start from the activity data, because session twenty five's four pass review already tells you who does what, which makes tier assignment a classification exercise on real usage rather than a badge decision. Assign the gated population, whoever genuinely needs Prime gated capability, counted, named, and priced at Prime, and it is usually a small fraction of the estate. Assign everybody else correctly, to the tier their measured use justifies, which is session sixteen's buyer test applied at the exact moment the packaging is changing anyway. And then cap and pin, uplift capped, pool and overage rate pinned, rollover added, all inside the same conversation because they all arrive together. Now the note, which is the opportunity hiding inside all of this. The packaging change is the one moment when re-tiering is administratively normal rather than an awkward request, which makes it the cheapest chance you will get to fix a tier that has been wrong for years.
Knowledge check three. When is the best moment to correct a tier that has been wrong for three years? A, at the next standard renewal, as a separate request. B, during the migration, when re-tiering is administratively normal and everything is being remapped anyway. C, immediately, mid term. D, never, tier changes are too disruptive. Pause here, and ask when requesting a different tier requires no explanation at all.
The answer is B, during the migration. Every entitlement is being mapped to a new equivalent regardless, so proposing a different destination tier for a population is an ordinary part of that exercise rather than a special request that has to be justified on its own merits. And that difference in framing matters enormously, because the same change presented as a downgrade invites a defence of the status quo and a conversation about why you bought the higher tier in the first place. Answer A makes it exactly that standalone downgrade conversation, which is harder to win. Answer C is mid term, where session twenty eight showed that additions and changes price worst of all. The packaging change is the cheapest re-tiering opportunity you are going to get, and it only comes around when the vendor decides it does.
Four moves for a migration quote, and treating migration as a full renegotiation is the whole method. One, demand the map, written, line by line, functions named, the cheapest item on this list and the one that prevents quiet upgrades. Two, blend the tiers, Prime only for the gated population, because uniform Prime is roughly a third again for capability most people never open. Three, cap the uplift, year over year across the term, so the migration is a one time step rather than the first move of a climb. Four, pin the meter, pool size, overage rate, rollover, because the consumption model arrives with the packaging so it is negotiated here or it is not negotiated at all. And then the timing, start nine to twelve months out rather than at the quote stage, because by the quote stage the mapping has already been done for you and you are reacting to somebody else's arithmetic.
Guest analyst clip. I want to end on the optimistic reading of the migration, because this session has necessarily been about defence and there is genuinely an opportunity in here that most customers miss. Think about what a repackaging actually is from your side. Every single entitlement you hold is being opened up, examined, and reassigned, by both parties, at the same time, as a normal piece of work. When else does that happen? Never, in my experience. Normally if you want to change a tier for a population you have to make a case, justify it, explain why the original decision was wrong, and somebody has to own having got it wrong. During a migration none of that applies, because everything is moving anyway and nobody is defending a previous decision. So the customers who came out of this best were not the ones who fought hardest on the price. They were the ones who treated it as a chance to fix things that had been quietly wrong for years. The population sitting on a tier they never used. The module that came in a bundle and never deployed. The pool that was sized for a plan that changed. All of that can be corrected in one conversation where correcting things is the agenda. If you have a migration in front of you, that is worth a week of somebody's time, and it is a week you will not get offered again.
A migration is the one moment when correcting things is the agenda rather than an awkward request, so it is worth a week of somebody's time. Session thirty four takes the contract terms themselves, the eight clauses that outlast the discount, which is where every protection from this session gets its final wording.
Three sentences. Legacy SKUs reached end of sale on the first of July 2026, so every renewal migrates, and a migration accepted as a formality is a renewal negotiated by the vendor alone. A like for like map showed a ten to twenty five percent uplift purely from the AI bundled into the entry tier, and uniform Prime standardization added roughly a third again for capability most users never opened. And the protections are a written entitlement map with functions named, an uplift cap across the term, and pinned consumption terms, all agreed before the quote is accepted and planned nine to twelve months out rather than at the quote stage.
Homework, about an hour, five items. Find your migration status, whether you have migrated already or it is coming at your next renewal, and if it has already happened, find out whether there was ever a written map, because that answer tells you a lot. Count the gated population, how many people genuinely need to build custom AI agents, and that number is your Prime population and it is usually much smaller than the quote assumes. Decompose your migration quote into bundled uplift, tier standardization, mapping drift, and consumption, and if the quote is a single number then that is itself the finding and it is the first thing to raise. List your production functions, the named capabilities you actually run today, so they can be written into the map rather than assumed inside a tier that may move. And model the meter, taking your assist burn from session twenty three and pricing it at the migration's overage rate, because that is the variable half that nobody quotes you.
Five guides. Tier migration and what it costs has the bundled uplift, the uniform Prime problem, and the price protection terms, and it is today's session in written form so it is the one to have open when a migration quote arrives. Foundation, Advanced, and Prime compared sets out what each tier actually contains after the remap, which is the reference you need to argue a destination tier line by line rather than in general. The 2026 pricing tiers pillar carries the full packaging map including which capabilities moved up and which lost a tier entirely. Now Assist consumption and overage in 2026 covers the meter that arrives with the packaging and the four terms that cap the variable half of your migrated bill. And the pro versus enterprise decision is the buyer test on tier fit, applied at the one moment when re-tiering is administratively normal. Next time, the eight clauses. See you there.