ServiceNow Now Assist pricing: a tier uplift on everyone, and a consumption pool on top.

Now Assist charges you twice: a tier uplift on every licensed user of a workflow product, and a consumption pool for what adopters actually use. Size and negotiate each one separately, before the AI disappears into a platform bundle.

How is ServiceNow Now Assist priced?

Now Assist is priced through two separate charges that run at the same time. The first is a tier uplift of 30 to 60 percent on the underlying license, charged on every licensed user of the workflow product. The second is a consumption pool that meters what the users who adopt AI actually do with it.

Most AI add ons pick one mechanism: a per seat subscription or a consumption meter. Now Assist uses both, and the two are independent. Getting the size of one right does nothing for the other, which is why buyers who negotiate only the headline uplift still overpay.

The tier uplift lands on the whole licensed population

Before April 2026, ServiceNow sold Now Assist through the Pro Plus and Enterprise Plus editions of ITSM, CSM and HR Service Delivery. In the quotes we reviewed, the enhanced tier replaced the base tier for the whole workflow product. Every licensed fulfiller carries the premium, whether or not they ever open a Now Assist panel.

The consumption pool meters what adopters do

Each enhanced tier includes an allowance of AI actions, and extra capacity is sold in packs. ServiceNow counts consumption in assists on its rate card, and different skills consume different numbers of assists. A case summary and an agent that resolves a case end to end do not draw the same amount from the pool.

What changed with Foundation, Advanced and Prime in 2026?

In April 2026 ServiceNow cut its tiers from five to three: Foundation, Advanced and Prime. Now Assist is bundled into every tier and metered in assists against a per tier allocation, and the legacy SKUs ended sale in July 2026. Only Prime allows you to build new custom AI skills and agents.

The two meters survive the change. The tier price now carries the AI cost for every licensed user, while the allocation and top up packs meter consumption. An active subscription keeps its terms until you renew or change it. The tier mapping is compared in our Foundation, Advanced and Prime guide.

What does the enhanced tier cover in each workflow product?

Coverage is set per workflow product. An enhanced tier on service management does not extend Now Assist to customer or HR workflows, so a company running three workflow products faces three separate uplift decisions. The table shows what we saw quoted on the legacy Plus tiers.

The allowances are shown in actions, as they appeared on the order forms we reviewed. Ask the account team to restate each allowance in assists against the current rate card. Skills consume assists at different rates, so an allowance that looks generous for summaries can cover far less once automation skills go live.

How do the two enhanced tiers differ?

They differ in capability as well as allowance, which makes the choice a scoping decision as much as a price decision. The first enhanced tier adds generative capability: summarization, drafted replies and knowledge article generation. It suits workflow products where helping a person read and write faster is the main use.

The higher tier adds agent and process automation: components that close cases and trigger workflows themselves. It carries roughly double the action allowance. Drafting help in one product and real case deflection in another are two requirements, and buying the higher tier for both overpays on one of them. Edition pricing context is in our ServiceNow pricing guide.

What does the uplift cost when only a third of users adopt AI?

It costs almost three times what the same premium would cost if you paid only for adopters. The example below is hypothetical, and the $100 base rate is illustrative because ServiceNow publishes no list prices. Say you license 1,000 ITSM fulfillers, accept a 40 percent uplift, and 35 percent of them use Now Assist in the first year.

Across both meters, the gap between the quote and what the adopters needed comes to $342,000 to $362,000 in year one. The uplift gap alone is $312,000, roughly six to ten times the pool gap. That is why the argument about who carries the premium should come before any argument about pack size.

How the numbers move with the size of the service desk

Keep the same illustrative rate, the same 40 percent uplift and the same 35 percent adoption, and the uplift gap scales in a straight line with the fulfiller count:

• 300 fulfillers. The uplift costs $144,000 a year against $50,400 priced on 105 adopters, a gap of about $94,000. That is often too small for a long fight, but it is reason enough to phase.
• 1,000 fulfillers. The $312,000 gap from the table, large enough to justify its own line in the negotiation.
• 5,000 fulfillers. The uplift costs $2,400,000 a year against $840,000 on 1,750 adopters, a gap of $1,560,000 for one workflow product. Add CSM and HR on the same terms and the gap repeats per product.

Why do the two meters have to be negotiated separately?

They answer different questions, so they respond to different arguments. The uplift is a per user premium that scales with how many people are licensed for the workflow product and ignores whether any of them invoke AI. The pool is a consumption prepayment that scales with how much the adopters do.

If you negotiate hard on one and accept the other, half the problem stays open. The open half is usually the larger one, because the uplift applies to everyone while the pool applies only to the active minority.

Adoption is the variable that links them

In our reviews adoption ran at 30 to 40 percent, which makes both purchases wrong in the same direction at once. You pay the premium for a population that does not use the feature, and you prepay consumption for users who have not arrived. The two errors add up.

Paying the uplift on everyone while prepaying a pool built from someone else's forecast means paying twice for the same slow adoption.

Why we advise against buying AI for every workflow product in one deal

The usual advice is to put ITSM, CSM and HR on the enhanced tier together, because a bigger order earns a bigger discount. We think it costs more than it saves. Coverage does not travel between products, adoption stayed around a third of users, and adding the second product later carries no commercial penalty.

Start with the workflow product where the use case is strongest. Size the pool from measured ticket volume in that product, then add the next product under a written price hold on the same uplift percentage. The renewal sequence around this is set out in our ServiceNow renewal guide.

How should you size the tier and the assist pool before you sign?

Size each meter against what it measures: the uplift against the people who will use AI, and the pool against the work they will do. One adoption estimate should drive both numbers, and it should come from a workflow product you already run.

For the pool, rebuild the forecast from your own data instead of adjusting the vendor's version:

1. List the skills you will switch on. Name each one, for example incident summarization or resolution notes, and leave out anything planned for a later phase.
2. Count the work each skill touches. Use the monthly average of ticket volume over the last 12 months for the records that skill acts on.
3. Apply your adoption estimate. Multiply by the share of fulfillers you expect to use that skill in year one.
4. Convert to assists. Multiply by the assists per use for that skill on the current rate card, and have the account team confirm the figure in writing.
5. Compare with the quote. The gap between your annual total and the quoted pack is the number to negotiate.

How to check your own numbers in the instance

Everything you need to argue both meters already sits in your ServiceNow instance or subscription portal. Pull these before the first pricing call:

• Ticket volume. Report on the incident, sn_customerservice_case or sn_hr_core_case table, grouped by month for the last 12 months. That is the workload the pool should match.
• Real adopters. Now Assist Admin, then Performance, opens Now Assist Analytics. The Usage and Adoption dashboard shows how many users invoke skills, which is the number the uplift should be priced on.
• Consumption by skill and instance. The Now Assist usage view in Subscription Management compares the purchased allocation with assists consumed, by skill and by instance. It needs the usage_admin, sn_sub_man.admin or admin role, updates nightly, and hides usage for any instance where data sharing is restricted.
• Licensed population. The fulfiller count per product in Subscription Management is the number the uplift multiplies. Check who holds a fulfiller role before you accept it, using our fulfiller and requester guide.

If Now Assist is not live yet, use the workflow product you already run as the baseline. Refuse a pool sized for products you have not enabled.

What have we seen in recent Now Assist pricing reviews?

The consumption forecast is the weakest part of most quotes. Across roughly 20 to 30 Now Assist pricing reviews we ran in 2024 and 2025, the forecast ServiceNow supplied was rarely tied to real ticket volume. Pools were sized against an adoption curve instead of a workload, the same overcommitment we see across consumption priced AI.

Three findings recurred:

• Packs sized too large. Quoted credit packs ran 25 to 50 percent above first year actual usage.
• Uplift on the full population. The enhanced tier premium sat on the underlying license of everyone licensed for the workflow product, adopters and non adopters alike.
• AI locked into the platform SKU. The AI features were bundled into platform SKUs in a way that blocked clean unbundling at the next renewal.

The third finding decides the renewal. Once the AI sits inside a platform SKU, the renewal starts from a bundle that cannot be taken apart, so the separation has to happen while the first order is still open and ServiceNow still wants the deal. Related guidance is collected in our ServiceNow knowledge hub.

What will the ServiceNow account team say, and how should you answer?

Expect these lines in most Now Assist conversations. Each has a reply that keeps the two meters apart:

• "AI is part of the tier, so every user gets it." Then price the premium on the users who will adopt it. Quote the uplift for our expected adopters and spread that amount across the seats.
• "The forecast is based on similar customers." Rebuild it from our incident volume for the last 12 months and the skills we plan to enable, with the assists per skill from the current rate card.
• "Buy all three workflow products now for the better discount." We will buy one product now and want a written hold on the same uplift percentage for the others for the full term.
• "The Plus SKUs have ended sale, so you need Prime." Show us the mapping line by line. We will pick Foundation, Advanced or Prime by the processes we run and compare total cost including the assist allocation. Our note on tier migration costs covers the comparison.
• "The AI cannot be priced separately." Then the order form needs the tier price, the AI allowance and the top up rate as separate lines before we sign.

What contract terms should you ask for on Now Assist?

Ask for terms that keep each meter visible and adjustable for the life of the contract. These are the ones worth spending negotiating time on:

1. Separate line items. Tier price, AI allowance and top up packs priced on their own lines. This is the only point where unbundling is clean.
2. Allowance stated in assists. Name the rate card version in the order form, so a later rate card revision cannot shrink what you bought.
3. Top up price fixed for the term. Overage is where an undersized pool gets expensive; see our top up unit price benchmark and the note on consumption overage.
4. Price hold for later products. The same uplift percentage for CSM or HR when you add them, so phasing costs nothing.
5. Right to reduce the pool. A reduction at each anniversary if measured burn stays below the allowance.
6. Instance scope. Written confirmation of which instances draw from the pool, since Subscription Management reports assists per instance and test activity should not consume production capacity.

What to do next

1. 12 months out: split the meters. Put the tier uplift and the assist pool on separate lines in your budget, because one is a per user premium on the whole population and the other a prepayment for the active minority.
2. 9 months out: measure. Pull 12 months of ticket volume per workflow product and current adoption from Now Assist Analytics.
3. 6 months out: choose tier by product. Match each workflow product to the tier its use case needs, generative help or automation, and price the uplift on expected adopters.
4. 4 months out: phase. Commit to the workflow product with the strongest use case first and secure a price hold for the others.
5. 3 months out: size the pool. Set it from measured ticket volume and refuse capacity for products not yet enabled.
6. At signature: unbundle. Confirm the AI components are priced separately from the platform SKU. The ServiceNow practice can run the sizing with you.

Frequently asked questions