They will renew you at three years anyway, so years four and five are not retention insurance, they are a financial product you are selling. 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 twenty nine, and today is the one where the numbers get big. Multi year deals, enterprise agreements, and the question of what a long term is actually worth. And I want to start by removing an assumption that quietly costs buyers a great deal of money. When an account team proposes five years, it usually arrives framed as commitment, or partnership, or occasionally as a favour extended to a valued customer. Here is the arithmetic that governs that room. ServiceNow reports a ninety eight percent renewal rate. They do not need five years to keep your logo. They will keep it at three, and they will keep it at one. So years four and five are not retention insurance that you owe them. They are a financial product that you are selling, and today we work out the price. Three checks, homework, let's go.
Five objectives. First, price duration as currency, because that ninety eight percent means the extra years are something you sell rather than something you concede. Second, set the package rather than the number, because against the best three year offer on the table, five years should buy six to ten incremental discount points plus a waived or heavily capped escalator. Third, lock the price book, because locked quantities with unlocked tier definitions let the vendor hold your unit price flat for five years and still take twenty to forty percent out of you through repackaging. Fourth, cap the meter that is not knowable, because everything else in a long deal is knowable at signature and assist consumption is not. And fifth, build the exit ramp, because a term with no break right converts every later conversation into an appeal, since walking away has stopped being a move that is available to you.
Four framings, and this is the rare case where the vendor has published the exchange rate themselves. Ninety eight percent, the renewal rate, so they will keep your logo at three years and at one. Twenty nine billion dollars, total remaining performance obligation, up roughly twenty two percent, with current RPO at thirteen point two billion, up twenty one and a half percent and around two hundred basis points above their own guidance. Going longer, which is management on the Q2 FY2026 call saying that many deals are going longer and that you should see it in the cRPO and the RPO, and that is a disclosed value driver stated in a filing environment where the words are lawyered. And six to ten points, the incremental discount that five years should buy against the best three year offer, plus the structural rights we will come to. Now the note, because it inverts something people believe. Anyone reasoning from a ninety eight percent renewal rate to a weak buyer position has the logic backwards. High retention means duration is the one thing they still have to buy from you.
Guest analyst clip. I find the renewal rate statistic gets used against buyers far more often than it gets used by them, and it should be the other way around. Somebody sees ninety eight percent and concludes, correctly, that almost nobody leaves, and then concludes, incorrectly, that they therefore have no position. Think about what that number actually tells the vendor. It tells them your renewal is close to certain. It is already in the forecast. Nobody in that account team wakes up worried about whether you will renew, because statistically you will. So retention is not what they are buying from you in that conversation, because they already have it. What they do not have, and what is genuinely uncertain, and what their own investor disclosures say drives the backlog, is duration. Whether you sign for three years or five is a real open question with real financial consequences on their side, and it is entirely in your gift. That is the asset. And what I see customers do, again and again, is give the asset away in exchange for a warm relationship and a couple of discount points, because nobody told them they were holding anything. Before you agree to a longer term, work out what those extra years are worth to them, and then charge for them. It is an unusual position to be in with a vendor of this size. It is worth using.
Retention they already have, duration they do not. That is the asset, and it is entirely in your gift. So let's price it, because the way most buyers get this wrong is subtle and expensive.
What five years should buy against a three year baseline. Discount off published rates above two million ACV, baseline twenty five to thirty five percent, and five years should take you to thirty three to forty five, so six to ten points incremental. The annual escalator, baseline a four to seven percent typical ask, and five years should waive it entirely or hard cap it at zero to three. The price book, which resets at each renewal on a three year deal and should be locked at signature rates for all sixty months on a five. Discount application, which on the baseline is a headline rate across years one to three, and on five years should be stepped, with points specifically earned in years four and five. And the deal desk reserve, held back on the baseline, released as the price of the extra two years. Now the note, which is where most buyers lose the money. They negotiate one headline discount and then agree to apply it across five years instead of three, and call that a win. It is not. That is a rate you already earned, stretched thin. Years four and five each have to purchase incremental economics, priced separately, or you gave them away for free.
The price book lock, five points, and the headline is that the unit price is the decoy. The exposure, which is that if you lock quantities and leave tier definitions open, the vendor can hold your unit price perfectly flat for five years and still take twenty to forty percent more by moving capability you already use one tier north. April 2026 sharpened this, because the remap was not tier neutral. ITOM and CSM lost the Foundation tier entirely so both sit at Advanced as a floor, and Change, Problem, Major Incident, Process Mining, Platform Analytics Advanced, and Walk-up Experience all moved up, and DevOps Change Velocity, which used to live inside ITSM Pro, is now Prime only. So name the functions rather than just the SKUs, and if Walk-up Experience or Process Mining is in production today, it gets named as an entitlement you keep at the contracted rate whatever tier it later lands in. Add successor packaging language, mapping any future repackaging at no less than your current entitlement and no greater than your current unit price, entitlement first and price second. And refuse the three year hold on a five year term, because that is a two year repricing window you are being asked to donate. Price any unprotected year at three to five discount points.
Knowledge check one. You are offered a five year term with your unit prices held flat for the full sixty months. What is still exposed? A, nothing, a flat unit price for five years is complete protection. B, the tier definitions, because repackaging can move capability you already use one tier north. C, only the AI consumption line. D, only the annual escalator. Pause here. A price is a rate for a thing. Which half of that is locked?
The answer is B, the tier definitions. A price is a rate applied to a defined thing, and locking the rate while leaving the definition open means the thing itself can move, so a flat unit price attached to a repackaged tier can still cost you twenty to forty percent more, and April 2026 demonstrated exactly that at scale across the installed base. Answer C is a completely real exposure that we come to later in this session, so it is not wrong so much as early. And the clause you want is specific and boring, which is precisely why it works. Named SKUs, named tier equivalents, named per unit rates held flat for the full term, with successor packaging mapped at no less than your current entitlement. Boring clauses are where the money is, and by now that should be a familiar refrain.
Swap rights over a long term, which we met last session and which change character completely at five years. The arithmetic of a long term is unforgiving. The module mix you commit to in 2026 will be wrong by 2029, and the probability that your product mix is still right declines with every extra year you sign, while the commitment does not move at all. What to demand, an annual reallocation window of fifteen to twenty five percent of committed ACV across any SKU in the contracted price book at the same discount percentage. The two conditions to refuse, no requirement that swapped in products carry equal or higher list value, and no net ACV increase trigger, because both of those invert the purpose of the right entirely. And the sentence that protects it, which is that reallocation within the committed pool is not a new subscription, not an add on, and not a contractual change for migration purposes. Session twenty eight introduced this right for any term. Over five years it stops being useful and becomes essential, because your forecast error compounds and your commitment does not.
Knowledge check two. ServiceNow offers four extra discount points for five years, with no price book lock, no swap rights, and no AI top up cap. What is the assessment? A, reasonable, four points is close to the six to ten band. B, duration sold for nothing, because the structural rights are what make a long term survivable. C, good, since a longer term reduces negotiation effort. D, accept it and negotiate the rights next year. Pause here, and ask what you are giving up and what you are getting for it.
The answer is B, duration sold for nothing. The headline points are the visible half of the trade and the rights are the half that decides whether years four and five are tolerable, so a discount with no lock, no swap, and no cap prices your extra duration at approximately nothing while removing your ability to respond to anything that changes. Answer A is close on the number and misses that four points without rights is a worse deal than six points with them. Answer D is the trap inside the trap, and I want to be blunt about it, because after signature there is no next year. The leverage you would have used to get those rights was the signature, and you just spent it. And answer C is the argument procurement teams make for their own convenience, and fewer renewal cycles is the single most expensive convenience available in enterprise software.
The AI meter, which is the one line that is genuinely not knowable at signature, and that asymmetry is the whole problem. Everything else in a five year deal you can model. Assist consumption you cannot, and an uncapped top up rate compounds quietly inside a contract you have already agreed not to leave. It is also their priority ask, because ServiceNow crossed a billion dollars in AI ACV in Q2 2026, with agentic deployments up ninefold in nine months, tracking above their own one point five billion target, and that is the quota your account team is measured on. So trade explicitly. Their ask is AI attach and yours is a ceiling, and you trade one for the other deliberately rather than letting them collect both, which is the default outcome. Three demands hold up under pressure. A fixed per unit top up price for the full term with an annual true down so unused pool does not silently become next year's baseline. Pool sizing tied to actual fulfiller count with reallocation across business units, because a pool stranded in one department is shelfware with a meter attached. And strike the legacy AI line items that still surface on renewal drafts alongside the bundled entitlements. Expect a discounted rate for twenty four months and a promise to revisit. Revisit means reprice.
Guest analyst clip. The thing that worries me most about long ServiceNow terms right now is not the price, it is the AI meter, and specifically the combination of an uncapped consumption line and a commitment you cannot exit. Take those two separately and each is manageable. An uncapped meter on a one year deal is a problem you can fix next year. A five year commitment with everything else fixed is a decision you can model. Put them together and you have created something quite specific, which is an open ended cost inside a closed door, and that is the worst structure in commercial software. Because when the consumption runs hot in year three, and adoption is genuinely succeeding so you cannot simply switch it off, what exactly is your move. You cannot leave. You cannot reduce, unless you negotiated the right. And the rate is whatever the rate is. So my strong advice on any term beyond three years is that the AI top up rate has to be fixed for the full term, not for the first two years, and I would rather see a client sign three years with a capped meter than five years with an uncapped one, even at a better headline discount. That is not a popular position with a deal team chasing a number. It is the one I would defend in front of a board in year four.
An open ended cost inside a closed door, which is the worst structure available. Three years with a capped meter beats five years with an uncapped one, even at a better headline discount. Which brings us to the door itself.
The exit ramp, and why no five year term should be unconditional. A long commitment with no break right does something very specific to you. It converts every subsequent conversation, expansion pricing, packaging migration, professional services scope, AI top ups, into a rate discussion where walking away is not an available move. You are not negotiating anymore, you are appealing, and those are different activities with different outcomes. Four provisions do the work. A termination for convenience or committed ACV step down at the end of year three, twenty to thirty percent rather than full termination, on ninety to a hundred and twenty days notice, and the partial right is far easier for their legal team to approve than a full one. An M&A adjustment allowing pro rata reduction when headcount leaves the enterprise, which matters more than most buyers assume across a sixty month window. An SLA and service credit schedule that survives the whole term rather than lapsing with the original order form. And a most favored terms clause pulling forward better packaging economics published later, which is your only real protection against a second commercial reset like April 2026. Expect a governance concession instead of a contractual one. A scheduled meeting is not an exit ramp.
Knowledge check three. Your account team responds to a year three step down request with an annual executive business review. What is that? A, a fair substitute, executive attention has real value. B, goodwill rather than a right, and it does not survive a change of account owner. C, better than a step down, because it is recurring. D, grounds to walk away from the negotiation. Pause here, and ask which of those two you could actually enforce in eighteen months.
The answer is B, goodwill rather than a right. A scheduled meeting is not an exit ramp, because it creates a forum for a conversation you could have had anyway, and it depends entirely on individuals who will rotate through your account several times across a sixty month term. Answer A is fair that executive attention has genuine value, and it is simply not the thing you asked for, and accepting a substitute for a structural right is how buyers end up with a warm relationship and no options. Answer D overreacts to what is a completely normal counter, and treating it as an insult tells them more than you want to reveal. Draft those four exit clauses before you name a term length, because every one of them is far cheaper to obtain while the fifth year is still yours to sell.
Four counters, and they arrive in a predictable order, which is genuinely useful because it turns each one from a surprise into a checkpoint. One, the end of sale argument, where you are told that because legacy SKUs went end of sale a long term is now the only way to protect what you have. That is half true and entirely self serving, because what actually moves you onto current packaging is a triggering event you control, which is session twenty eight. Two, the early renewal offer, dressed as a favour, and you treat the pull forward as a separate transaction with its own price, which session thirty takes in full. Three, points without rights, three to five points above your current rate contingent on five years with no lock, no swap, and no cap, which is duration sold for nothing and which you can now name on sight. And four, the escalation excuse, where the rights require executive escalation and cannot be committed this quarter. Escalation is a scheduling problem rather than a policy problem, and it resolves remarkably quickly against a quarter end clock.
Guest analyst clip. There is one request I make in every long term negotiation and it does more work than anything else I do, which is to ask them to itemise. Not to justify, not to explain, just to itemise. If you are asking me for two extra years, price them. Show me what year four bought and what year five bought, as separate lines, against the three year offer that was on the table before this conversation started. And what happens next is genuinely diagnostic. If the account team can do it, you are dealing with somebody who has thought about the deal structurally and you will probably have a productive negotiation, and the numbers may well be reasonable. If they cannot, and quite often they cannot, then you have just learned that those two years were free in their model, which means they are free in yours too unless you charge for them. I have watched that request change the temperature of a room, because it moves the conversation from adjectives to arithmetic. Partnership, commitment, strategic alignment, all of those words survive perfectly well until somebody asks for the line items. And it is a completely fair request. You are being asked to commit two additional years of spend. Asking what they are worth is not aggressive. It is the minimum any board would expect you to have asked.
Ask them to itemise, and if they cannot, those years were free in their model and they will be free in yours unless you charge for them. Session thirty closes module six with the renewal calendar itself, the twelve month runway run backwards from the date, and the approval ladder that decides which numbers are even available to you.
Three sentences. A ninety eight percent renewal rate means they keep your logo at three years, so years four and five are a financial product you are selling into a disclosed backlog metric rather than retention insurance that you owe them. Price the package rather than the number, six to ten incremental discount points, a waived or zero to three percent escalator, and a price book locked for the full term, because locked quantities with unlocked tier definitions still cost you twenty to forty percent. And a long term needs the rights that make it survivable, reallocation of fifteen to twenty five percent of committed ACV a year, an AI top up rate fixed for all sixty months, and an exit ramp at year three, because a scheduled meeting is not a break clause.
Homework, about an hour, five items. Quantify the concession, which means multiplying your annual spend by the extra years being asked for, saying that number out loud as the incremental backlog you are prepared to sign, and requiring that it be priced line by line. Check what your price book actually locks, rates only or rates and tier definitions, and if it is only rates then you have just found the twenty to forty percent gap. Name your production functions, the capabilities you genuinely run today, by name, so that they can be written in as entitlements rather than assumed inside a tier that may move. Model the AI line across the term, taking your assist burn from session twenty three and extending it at the top up rate for the full length, because that is your uncapped exposure and you want it as a number. And draft the four exit clauses, step down, M&A adjustment, surviving SLAs, and most favored terms, and draft them before you name a term length rather than after.
Five guides. Multi-year term length as currency is today's session in full, the disclosures, the pricing table, the price book lock, the swap rights, the AI cap, and the exit ramp, and if you are facing a long term proposal it is the single most useful thing on this list. The ninety eight percent renewal rate and walk-away credibility explains why high retention strengthens rather than weakens your position, which is counterintuitive enough to be worth reading properly. The AI ACV quota leverage trade covers how to sell them the AI logo while buying the pool cheaply, which is the explicit trade this session recommends. The early renewal pull-forward trap is the second counter in today's list, priced in full, and it is the bridge into session thirty. And tier migration and what it costs maps the April 2026 remap tier by tier, which is the exposure the price book lock exists to close. Next time, the renewal calendar, and module six closes. See you there.