Yahoo Finance

William Mcdermott

Thank you, Darren. Good afternoon, and thank you very much for joining today's call. There are two big themes I'll cover. The first is ServiceNow's outstanding business fundamentals. We'll discuss another quarterly result that beat expectations driving our remaining performance obligation to nearly $23 billion and a platinum balance sheet with $10 billion in cash and investments on hand.
We are one of one, a 20%-plus grower at scale with margin accretion, printing free cash operating at the Rule of 54. We blew through our original $10 billion dream, and now we're on track for our midterm target of $15 billion plus by 2026, with a long-term vision to be the defining enterprise software company of the 21st century.
The second theme is ServiceNow is the leadership position in the seismic shift to a genetic AI. Stunning 150% quarter-over-quarter deal growth in our key Pro Plus AI offerings is an exciting sign of what's ahead, especially with the new innovations we announced today. It doesn't matter to ServiceNow, who builds the models with the precipitous drop in LLM compute costs, there is much more capital allocation available for the business impact layer. Our position at the center of data, AI agents, workflow orchestration and enterprise governance is the nexus of AI's massive value creation opportunity.
Let's talk about the business fundamentals first. You'll recall that during the course of 2024, we consistently beat expectations and raised our guidance twice. In Q4, our biggest quarter of the year, we exceeded expectations on top of that beat and raise track record. Subscription revenue growth was 21% and CRPO growth was 22%. That's double IDC's 2024 growth rate for the IT sector.
Operating margin was 29.5%. Each of those results was 50 basis points above the high end of our guidance. So two was our full year 2024 free cash flow margin of 31.5%. In terms of the shape of the business, ServiceNow continues its enterprise platform expansion. We had 19 deals greater than $5 million in net new ACV.
We landed our largest new logo deal ever. We had 2 additional customers cross the $100 million in total ACV, doubling our total in one quarter. It's clear that our AI platform message is standing and it's scaling.
On the product areas, 76 of our tech workflow deals were over $1 million, including one over 15 million, ITSM, ITOM and ITAM and Security and Risk were all in at least 15 of the top 20 deals IT asset management was actually in all of our top 20. Customer and industry workflows more on this later, continued its disruption of the category with 30 deals over $1 million. Employee and creative workflows, both had very good quarters with at least 17 deals, each over $1 million. We're very proud of ServiceNow's performance in 2024. No enterprise software company has ever executed at this scale the way we have.
You'll hear from Gina on the 2025 guide, in a few minutes. To say, we expect another great year is a given. This is consistent with how we have guided the business in recent years when you'll recall, we consistently overachieved our stated goals. Even on this bigger revenue base, be assured, we will save the innovative fast growth highly profitable company we are today.
Let's spend some time on AI. I gave you the 150% quarter-over-quarter result. This is the number of customer deals for the AI-fueled Pro+ level in our ITSM, CSM and HRSD solutions. Let's go a level deeper. IDC forecasts worldwide revenue for AI platforms software will grow to USD 153 billion in 2028.
The with a 40.6% CAGR during that period. For 2025 alone, they forecast 46% growth in AI software spending. With this secular strength, AI is fueling a top to bottom reordering of the enterprise technology landscape. You've seen the impact of this in energy, infrastructure and hardware. You'll now see it even more in software.
Every CEO I meet wants to simplify their business to unlock the heaviness of legacy architectures. They want to innovate with data workflows and agentic AI.
ServiceNow is winning with 1,000 customers on the Agent AI journey already. Here's why. First, ServiceNow has the AI agents, the workflows, the workflow data fabric, Raptor DB and the integration hub in one platform at scale. The majority in enterprise software is struggling with their well-known integration challenges and in some cases, exaggerating the readiness of incremental progress. Second, ServiceNow has thousands of AI agents, leveraging 20 years of automation, data and outcomes.
This is knowledge gleaned from trillions of workflows we run, including billions of automation knowledge, sources and tools. We've been building an genic architecture. For the 5-plus years I've been the CEO and lockstep with Jensen Wang and our great friends at NVIDIA. Third, since enterprises will experiment with more than one vendor's agents, ServiceNow's platform handles orchestration and governance. Nobody wants a new cloud mess where every department is out buying agents and nobody understands what the agents are doing.
Our AI agent orchestrator eliminates sprawl, managing teams of AI agents to tackle the most complex enterprise processes. Even as we reach new buyers, like Chief Data and AI officers ServiceNow's role as the CIO's control plane is the ultimate move in enterprise AI. If you haven't seen Idris Elba's latest brand work for ServiceNow, you can check it out on servicenow.com or on the ServiceNow YouTube channel. The theme is every corner of your office. Our differentiation goes east to west and north to south.
And here's what this all means for our business. We are only getting started. The opportunity is in its early days. The unique agility of our business model sets us up for a quantum leap and future business growth. We have predicted and protected that are seat-based subscription staying as is a foundation you can feel secure in.
We have also included a massive upgrade path to Pros Raptor DB and Workflow Data Fabric. So seat-based subscription still there, rock solid, not going anywhere. Then we have these massive upgrade paths to the new innovation. Customers still like the predictability of this approach, and they're committed to long-term transformation on our platform. Now we are also enabling elements of consumption-based pricing as AI agents become a potent value driver for the enterprise.
While we could have launched an additional SKU and offered AI agents as an add-on to drive more immediate revenue growth, our strategy prioritizes accelerating adoption. So by foregoing upfront incremental new subscriptions, we are enabling faster penetration into our customer base and monetizing the hockey stick of usage over time. As the agents become increasingly productive they will drive the consumption pricing meter. And that, of course, will be in addition to the seat-based licensing foundation, you'll get both. The consumption of assists with agentic AI will be a multiple factor higher than that of a standard now assist query.
This approach ensures customers can access our AI capabilities as quickly and as seamlessly as possible while simplifying the sales process at the same time. This is the most transformative technology in decades. We are solidifying our leadership position as the AI platform for business transformation. The bedrock strength of this business model is built on driving the success of our customers. Some examples include Petrobras works with ServiceNow connecting data across more than 6,000 applications and catalog items, processing over 250,000 requests monthly to improve decision-making and governance.
A large DoD agency works with ServiceNow, getting immediate visibility into critical data and assets previously trapped in manual processes and outdated legacy systems. New Zealand Parliamentary service is using ServiceNow to reduce employee request response times from weeks to 2 days. And our largest nonprofit deal to date, Scouting America will leverage the ServiceNow platform to digitally transform its legacy HR and customer service processes with AI. Grupo Bimbo will revolutionize its factory operations using ServiceNow to get real-time insights and boost production efficiency. Toyota, Rolls-Royce, the federal service desk NTT Data, they all work with ServiceNow to put AI to work for people.
As always, there's so much more to give you a complete picture of ServiceNow's progress. We told you about the launch of Workflow Data Fabric, which is significantly expanding ServiceNow's addressable market. The increasing traction on a genic AI is surging awareness of workflow data fabric and our RaptorDB Pro offering, with new customers like NTT signing this quarter. We're further augmenting Data Fabric capabilities with the new Oracle and Google Cloud integrations announced earlier today.
In terms of our key strategic industries, the public sector is poised for a major modernization effort this year and beyond. The United States government alone spends $125 billion a year on IT, much of which is unauditable not cataloged and not tracked against actual usage. That's why we're so enthusiastic on the new Department of Government Efficiency mandate from President Trump. If you look at our track record across all federal government customers. ServiceNow is saving millions and millions in costs through efficiency improvements and millions and millions of hours by automating mind numbing work.
There's so much more we will do to help.
In terms of our partner ecosystem, there is growth happening everywhere. ServiceNow and AWS expanded our collaboration with new capabilities to accelerate AI transformation. This enables use of multimodal models developed on Amazon Bedrock and additional solutions are available on the AWS marketplace. ServiceNow and Google Cloud are expanding our partnership to deliver AI-powered tools to millions of users. ServiceNow will launch on Google Cloud Marketplace, and Google distributed cloud to address demand from global enterprises.
ServiceNow and Visa are expanding our strategic alliance yet again. Visa will use ServiceNow disputes management, built with Visa's solution and together with ServiceNow, we combine AI capabilities. ServiceNow and Five9 expanded our partnership to deliver a turnkey AI-powered solution for unified experiences. The streamlined solution reduces operational costs, increases agent efficiency and simplifies contact center operations, ServiceNow and Microsoft are expanding our alliance to accelerate disruption in the CRM category. This is built on Microsoft CoPilot and ServiceNow AI agent collaboration, leveraging the unique strengths of both platforms.
It's a good time to remind you at CRM and industry workflows is ServiceNow's fastest growing business. There's a reason for that. ServiceNow is uniquely positioned to take action and fully automate workflows across the enterprise. We have a key differentiator at the architectural level. We don't have to try to translate between our AI models and some third-party system.
We sell, fulfill and service on one platform, one architecture and one data model, absolute beauty. Other software vendors need to communicate across multiple systems and stitch them together to make their software work, resulting in poor customer experiences saddled with technical debt. We're just a front-end AI agent. A customer is limited in their ability to take action across the enterprise to drive the ticket to resolution. ServiceNow enables AI agents to seamlessly collaborate across the entire enterprise with our single database and single architecture.
Our peers, agents perform simple tasks while our AI agents work together and leverage the entire system to solve complex tasks and enable true business transformation. We know that agentic AI promises exponential increase on the ServiceNow platform. This is an exponential increase in seamless automation. On a fragmented platform like many others, there will be an exponential increase in complexity and failure rates. Autonomous AI running on a fragile foundation bespoke integrations between disparate systems is a recipe for exponential disaster.
Industry analysts have acknowledged ServiceNow's completeness of vision. ServiceNow is named a leader in the 2024 Gartner Magic Quadrant for the CRM Customer Engagement Center.
ServiceNow is also recognized as a leader in the Forrester Wave task-centric automation software. The global marketplace is noticing too, just today's ServiceNow was recognized as number two on Fortune's world's most admired software companies and even better. We were number one in innovation and number one for building a people-first culture. Forbes named ServiceNow is the number two on the first ever most trusted companies list right behind NVIDIA, pretty good company, if you ask me. The American Opportunity Index, which ranks companies based on creating opportunity for their employees, ranked ServiceNow in the top five in the world, number one in software.
These are just some of the many awards, all of which are rooted in a world-class team that is completely obsessed with serving our customers.
In summary, ServiceNow is full speed ahead. We are putting AI to work for people, our innovation, growth, profitability and brand leadership position us as a class of one in enterprise software. The market is moving with ServiceNow. With this combination of start-up passion and scale maturity, we have the right strategy to deliver immense shareholder value in the years to come. That's why I keep reminding our team, our $15 billion plus by 2026 midterm guidance is only one chapter on our path to $30 billion plus in revenue and beyond.
When the world works with ServiceNow imagination is the limit. Thank you all for your time and confidence.
Before we take questions, I'll hand things over to our President and Chief Financial Officer, Gina Mastantuono. She will illuminate our unique ability to sustain profitable revenue growth, even as we add the revenue equivalent of like a number-six-scaled market-leading software company on top of ServiceNow every year. Pretty exciting stuff. Gina, over to you.

Gina Mastantuono

Thank you, Bill. Q4 was a great quarter, capping a year of incredible innovation and execution. Once again, we exceeded all of our constant currency growth and profitability guidance metrics, demonstrating the strength of our platform and the extraordinary efforts of our team. . 2024 was a pivotal year in driving customer adoption of our generative AI solutions.
By embedding powerful AI capabilities into our offerings, we've enabled organizations to reimagine how work is done, unlocking new levels of efficiency, creativity and value. Customers across industries are embracing these solutions at a rapid pace proving the immense opportunity, AI and data bring to our business.
Turning to our results. Q4 subscription revenues were $2.866 billion, growing 21% year-over-year in constant currency, exceeding the high end of our guidance range by 50 basis points. RPO ended the quarter at approximately $22.3 billion, representing 26% year-over-year constant currency growth. Current RPO was $10.27 billion, representing 22% year-over-year constant currency growth, a 50 basis point beat versus our guidance. From an industry perspective, manufacturing posted impressive growth, surpassing 50% year-over-year.
Public sector grew nearly 40% year-over-year, driven by the strength in U.S. federal and state and local government.
Transportation and Logistics also reported significant growth during the quarter. We delivered a very strong 98% renewal rate in Q4, highlighting the trust and value customers consistently place in the Now platform. The number of customers generating over $1 million in ACV rose to 2,109, even more remarkable, the number of customers contributing $20 million or more in ACV was 35% year-over-year, fueled by continued success in securing large deals.
We ended the year with approximately 8,400 customers as our strategic focus on landing the right new customers continues to deliver results. The average deal size of our new logos has consistently grown, marking the sixth consecutive year of expansion. And in Q4, we landed our largest new logo deal in company history. We closed 170 deals greater than $1 million in net new ACV in the quarter, including 15 with new logos. What's more impressive is that 19 of these deals exceeded $5 million and 3 deals were over $20 million.
Our Gen AI net new ACV to date continues to trend ahead of any new product family launch for the comparable period in our history. In fact, AI net new ACV stepped up meaningfully in Q4 from Q3 as we saw momentum accelerate for our plus SKUs. As Bill noted, the number of now assist service desk deals, which include ITSM, CSM and HRSD grew over 150% quarter-over-quarter.
Furthermore, the number of customers who bought 2 or more of our Gen AI capabilities doubled quarter-over-quarter. Our plus SKUs also maintained a greater than 30% price uplift over Pro in Q4, a strong indication of the tremendous value they deliver. For example, a multinational conglomerate customers are a 45% reduction in live chat from July to December would now assist and the company is now targeting to save millions by the end of 2025. For the addition of our AI agents to many of these SKUs, we think the value proposition becomes even more compelling in 2025.
We're also seeing great early traction with our RaptorDB Pro offering. Since the announcement in just September, we already have 5 customers over $1 million in ACV, a leading multinational technology company that adopted RaptorDB Pro in Q4 is already achieving performance improvements that exceed expectations, with database response times up to 3.5x faster than previous benchmarks and a database size reduction of over 30%, driving significant cost savings.
Turning to profitability. Non-GAAP operating margin was 29.5%, driven by OpEx efficiencies and our top line outperformance. Our free cash flow margin was 47.5%. For full year 2024, operating margin was 29.5%, up 200 basis points year-over-year, and free cash flow margin was 31.5%, 50 basis points above our guidance and up 100 basis points year-over-year. Total free cash flow for 2024 was a robust $3.5 billion.
We ended the year with a healthy balance sheet of $10 billion in cash and investments. In Q4, we bought back approximately 293,000 shares as part of our share repurchase program. As of the end of the quarter, we had approximately $266 million remaining of the original $1.5 billion authorization. Given our strong cash position and our strategy of managing the impact of dilution, we announced today that the Board of Directors authorized the purchase of up to an additional $3 billion of common stock under this program. Together, these results continue to demonstrate our ability to drive a strong balance of world-class growth, profitability and shareholder value.
Moving to our outlook. As we enter 2025, we're not just planning to win today, we're making bold strategic decisions to position ourselves for sustainable strong growth in the years ahead. This year, we will incorporate more elements of consumption-based monetization across our AI and data solutions. For instance, we will include AI agents and our plus SKUs for going upfront incremental new subscriptions to instead drive even faster adoption and monetize the hockey stick of usage over time. We're optimizing our go-to-market approach and evolving our business model to create even more integrated solutions and sees the massive agentic AI and data opportunity in front of us.
Our guidance prudently reflects the flexibility to make these moves while delivering higher incremental free cash flow building on the accelerated margin trajectory created by our outstanding 2024 outperformance. With that in mind, for 2025, we expect subscription revenues between $12.635 billion and $12.675 billion, representing 20% year-over-year growth at the midpoint in a constant currency basis. This reflects an incremental $175 million FX headwind since the end of September and assumes a more pronounced back-half weighted seasonal linearity in our U.S. federal business due to the change in the presidential administration. We expect subscription gross margin of 83.5%, reflecting the diminishing impact of the change in useful life of our data center equipment as well as incremental data center investments related to public cloud partners, to expansion, NAI.
We expect an operating margin of 30.5%, up 100 basis points year-over-year driven by OpEx leverage across the board, including sales and marketing. We expect free cash flow margin of 32%. This reflects an incremental 50 basis points of margin expansion on top of our 2024 outperformance and in spite of an incremental cash tax headwind of 100 basis points year-over-year. And we expect GAAP diluted weighted average outstanding shares of $210 million.
For Q1, we expect subscription revenues between $2.95 billion and $3 billion, representing 20% year-over-year growth at the midpoint on a constant currency basis. This reflects an incremental $40 million FX headwind since the end of September and the previously noted Fed seasonality assumptions. We expect CRPO growth of 20.5% on a constant currency basis. I would note that CRPO saw an incremental $205 million FX headwind since the end of September. We expect an operating margin of 30%, and we expect $210 million GAAP diluted weighted average outstanding shares for the quarter.
In summary, 2024 was an amazing year, but we're just scratching the surface of what's possible. Our ability to seamlessly orchestrate and govern AI agents, coupled with our unified data architecture positions us uniquely to bring all your data, people and processes onto a single platform so you can take action on it. The moves we're making in 2025 aren't just about maintaining our lead, they're about expanding it. We're setting ourselves up to define the future of agent powered automation, solidify ServiceNow as the AI platform for business transformation and deliver strong growth year after year. Finally, Bill and I want to express our deepest gratitude to our employees around the world for their unwavering hard work and dedication.
You are the heart of our success.
With that, I'll open it up for Q&A.

Question and Answer Session

Operator

(Operator Instructions) Kash Rangan, Goldman Sachs.

Kash Rangan

What a superb quarter fiscal year end as well. Bill, a question for you. Can you talk a little bit more about your decision to do a hybrid consumption and subscription? And the reason behind your confidence that you can walk both those tracks very well. And Gina, if you could do a pro forma, had you not done the consumption? What would the subscription revenue growth outlook would have been hypothetically?

William Mcdermott

Thank you so much, Kash. Our goal is to combine both subscription and consumption pricing. So customers can start with a base subscription, which they like. They want that flag in the ground so they can predict their spend and their current ROI schemes. But then they obviously want to take advantage of agentic AI and yet at the same time, the industry is early in its formation.
So we're actually innovating faster than they have deployed it. So they want to scale with us in harmony and in partnership. So with our Pro Plus version, they'll get access to our agentic AI agents and will give them a meter-based pricing methodology where they will take out the sole crushing business process work that is tedious and complex that people actually don't even want to do and agentic AI agents will do that for them. They will see a very nice ROI on that. And by definition, if the meter is running up, that means they're using it and deriving financial gain from it, and they're happy to pay and share with us the profits.
So in a certain sense, cash, it's the Goldilocks model where you get it both ways. You get the same revenue and predictability you've been used to. But now we have a hockey stick formation that can turn into a sensational growth shoot for the company. So I think we really got this right and with Amit's leadership and our great engineering team, the way we're pressing innovation into the market is truly, truly stunning, and I'm so, so excited. I just want to give you one thought that nobody ever talks about.
And that is, no matter where you travel in the world, and I just got done with a global road show, there are millions of technical jobs that go unfilled year after year. With a agentic AI agents, you can now fill those jobs and at the same time, improve human productivity and the productivity curve of these businesses in a dramatic way. And nobody actually talks about there not being any such thing as artificial intelligence without human intelligence. And that's why we are putting AI to work for people, and that has landed in a very pleasant way with our customer relationships. So everything we're doing is on the right side of customer success.

Amit Zavery

So Kash, thanks for the question. This is Amit here. Just to add to what Bill was saying. The way we are thinking about this, the Pro Plus we have 1,000-plus customers already who will be able to now take advantage of the innovation we're bringing out with agentic, and it is subscription. So they're still paying for that kind of capacity.
If they go beyond the capacity and the assist, which is the meter gap capacity, if they go beyond that, they're still buying assist packs. So it's not like one assist per time in terms of consumption. So it's not really subscription packs in a way, but just that they can now go and tier those things much more easily versus having to do everything upfront, right? So they get the predictability and the confidence of the usage and the value and they grow with it. and they assist with a genetic really scales very, very fast because each assist each agentic call requires many assets.
So we just started getting them seeded and get going and the subscriptions really keep on growing as they start using more and more. So it's not completely like pay as you go or individual assist. It's really packs of assist in a way. So subscription pricing and we are giving them some flexibility as well and then giving the ability for customers to see value instantly.

Gina Mastantuono

And then on the second part of your question, Kash, as always, we're beginning the year with a thoughtful guide, right, that sets a strong foundation for success. And I talked about factors that we prudently factored into our guide. We just talked about the monetization of the agentic AI. I won't go into that further because I think Bill and Amit explained it quite clearly. The other piece is that we anticipate a more back-end weighted deal linearity in 2025 for federal, reflecting the outcome of the US election. And so what I'd say is that we're giving you a 20% constant currency guide while de-risking the guidance. And to your question on what would it be if we didn't do that, I would say higher.

Operator

Mark Murphy, JPMorgan.

Mark Murphy

So Bill, regarding the DeepSeek models, I'm very interested how you're assessing this moment where we have a Chinese AI model. It seems to perform well at a fraction of the cost, and I'm curious if it's something like DeepSeek, is that an option for ServiceNow? Or is it too risky with the data protections. And just more broadly, what do you think ServiceNow can do to exert its leadership as this cost of one of your input costs, which is the underlying models, it seems like it's starting to drop at a faster pace.

William Mcdermott

Yes. Thank you very much for the question, Mark. It's actually a fantastic news for ServiceNow because as these models are being commoditized at a rapid rate and probably more rapid than anyone could have dreamed of. It's super, super exciting platform. and app vendors like us since the competitive differentiation will happen at our level in terms of the applications, the business processes how you can help these companies run better and then ultimately get a business outcome.
And again, what I think helps us a lot financially, you can see the gross margin benefit of that, but you can also see our customers that have been investing in these LMs, getting a lot better deal now, so they kind of past the hardware, the infrastructure and the LLM and that price is dropping on everything, they'll have more left over to invest in ServiceNow. But we're truly set ServiceNow apart is our platform's ability to orchestrate and operationalize AI to create meaningful outcomes. Our customers are using ServiceNow to turn these AI insights into real tangible actions whether it's automating workflows, resolving issues proactively or enhancing productivity across the enterprise. And as it relates to Deep Seek, I would just simply say, we have a deep belief here in basically measuring twice, if not 3 times in counting once. So we believe very strongly in understanding what's really going on, and we're highly committed to responsible AI, and I know and his team are taking a look at that right now.

Amit Zavery

Yes, to add to what Bill said. We definitely are going to -- we are supporting third-party large language models. And we continue to work with many of those providers and doing a lot of integration with those technologies as well. The way we've architected our platform today, it's very easy for us to keep on adding any innovation which happens in that place and give that outcome from our applications. So it's really the combination of workflows with AI and the data is what our power is.
And we would take any technologies out there, which helps our customer and get them the best outcome. So that's really go, and we continue to work with all the large language providers, and we will look at Deep Seek. We have teams kind of investigating that. But the goal is that architecturally, we can support everything out there as needed.

Gina Mastantuono

And then Mark, on the cost side -- on the cost side, I would just say this is not unexpected. We've been talking about innovation and investing in innovation in AI for quite a while, and yet we continue to accrete margins. Our guide next year is for 100 basis points on operating and 50 basis points on free cash flow on top of overachieving in '24. And so really great cost discipline at ServiceNow while we continue to out innovate.

Mark Murphy

Well, congrats on having that foresight and delivering that.

Operator

Keith Weiss, Morgan Stanley.

Sanjit Singh

This is Sanjit Singh for Keith. Bill, when you think about this year, I kind of have sympathy for a lot of enterprise customers, every vendor in software is going to be pushing their agentic capabilities. So Bill, could you just detail a little bit about your agent strategy in that more fragmented landscape, what's going to give ServiceNow, the license to win? And also importantly, what are you contemplating from a go-to-market change perspective that's associated with this move to agents and the move to a hybrid subscription consumption model?

William Mcdermott

Yes. I think the number one most important thing is our AI orchestration serves as the control tower for business transformation. And if you think about our core strength in IT, and how we handle all things IT from servicing the business, running the operations, the assets, the security of it all, and then moving into the employee and the experience and giving them all the things they need and think about the customer, it's literally sell, fulfill and service all on one platform. And then from a creator perspective, with natural language wiping out 60% of the productivity drag to actually get them to doing the code and inventing something. We're handling all of that today on one platform with one architecture and one data model.
So we'll give you our agentic AI solution in all of those domains. But best of all, because everyone is pounding them with their siloed solutions, it's become highly credible to customers that somebody has to be the controller or the orchestrator for all of these other agents as well if they decide to implement them. And as soon as they see the architectural advantage of what ServiceNow has. They say, I go for this one. And there's another thing that's happening in combination, thanks to Amit and the great engineering here.
We now have RaptorDB, and I do remind you that Amit had an amazing career at Oracle before GCP. So he understands all of this at a very deep technical level. And frankly, we have some amazing talent in our company on database. And RaptorDB is a sensation not only in its speed and its ability to be fully automated on the ServiceNow instance and make big companies go three or five times faster but we also now have connectors into Snowflake into data bricks into oral into all of the hyperscalers, including the announcement today with GCP, we had AWS and Microsoft. Interestingly, AWS is number one and number four biggest deal ever in the enterprise, with ServiceNow customers running in AWS.
So they know us well. Everybody knows us really well. So now you're controlling east to west and you're going north to south up and down the stack, and then you ultimately integrate in all these systems of record where we have been moving that data into ServiceNow for years. But with RaptorDB, some customers are saying, well, why don't I just move it all into ServiceNow and skip the fragmented step of hundreds of broken fragmented installs that have been put in there departmentally over the last 55 years. And we tell them, have a way you want to leave the data where it is, we'll make a 0 copy.
You want to move it into ServiceNow, that's fine, too. You want to run it in the hyperscalers, including running our ServiceNow instance with some customers want to do, we let you do that, too. And as Amit said, we already integrate with all the LLMs. So I mean, seriously, if I'm sitting back as the CEO, I spoke with one that runs the largest insurance company in Asia, and I think that I'm going into an insurance call telling them how we can improve insurance claims and all this sort of stuff, and he said, Bill, I want to be a data company. Actually, I want to rely on the old business model of my agents shaking everyone's hands. I want agentic AI to do that.
I want to be a retail company. And then I explain our story and he tells me I have no idea. I said, well, we're still a young company, but now that you do, what do you want to do about it? I want to go with you. I said, great, I'll send them the team. Let's get it done. So when they hear this story properly presented, it's game set match.
So our challenge is simply to help you and the media and the customers get the memo because once you get the membo, this is the only viable option.

Operator

Matt Hedberg, RBC Capital Markets.

Matthew Hedberg

And Gina, congrats on your expanded role. Really great to see that. Bill, so obviously, a lot of -- Bill, obviously, a lot of enthusiasm on AI, which is great to hear. Just given your global perspective, you have such a good pulse on overall IT spending trends. Did you see things improve from what we last talked on September results?
And how do you feel at this point versus, say, this point last year, like are things improving out there in general?

William Mcdermott

I think that -- Matt, thank you for the question, by the way. And I couldn't agree with you more on Gina's promotion. She is stunningly fantastic and it's an honor to work with her and has been for the 5 years we've been together, and I'm just so proud to see her richly rewarded for what she deserves as a leader. So I think it's great. And congrats, Gina.
On the IT spending environment, I think it's similar in the IT spending environment to what it's been. But I do think the big change is the move from AI fascination to AI business model innovation. And it's clear to me that the best CEOs in the world have AI at the forefront of their mind. In fact, I actually see them even extracting AI from the CIO and the CTO remit, and it's not because they don't respect or really take pride in their CIOs and CTOs, but they believe they have to be AI-led companies. And so they're actually creating AI leadership and data leadership within the construct model of their org charts and those individuals are showing up in their boardrooms because these CEOs really understand that if they don't capitalize on this moment and they end up with second mover advantage.
It might be a liability. So they're really on their tippy toes. And therefore, if you have a great agentic AI story, you plant that flag in the essence of a great business model like we have, and you do the end model on what it very appropriately said on our agentic AI portfolio would now assist and you have a win-win relationship on consumption, value, and we both benefit from that. It is an extremely pleasing story. And on that dimension, they will open up the pocketbook and that is the key because they're getting value from it.
And the other piece is, in the public sector, I see a complete renaissance happening which I think is truly outstanding.
Nowhere in the world do I go where they don't want to talk about those and what's happening in America. And so the administration has done a very good job of getting everybody aware that something big is happening here. And because we have such a strong foothold in the agencies, in the military establishments and the defense establishments from all of them that you can possibly think of that really matter, including veterans, affairs, we have a great reputation. And so they know when we come in, we make work simple for people. And so if you think about sending everybody back to the office and telling them good luck, you're going to announce swivel chair between 17 different application experiences, which will eat up a 30-year productivity a day, they'll probably take the package and leave.
So the ServiceNow solution becomes even more relevant. And the fact that we could take cost out, bring productivity in and inspire the best in people is extremely inspiring.
The other thing that's really inspiring, and this is true in the private sector too, but in the public sector, they know they're inheriting more than half a century of disasters. So they're not interested in the rip and replace. They're also not interested in layering silos. They really want to rethink how things are running. So because we can extract the data from the systems they have.
We have RaptorDB and the connections to all the best companies in the world, and we could orchestrate the AI transformation. They're like, how do we get started? Because only you guys came in here with a price that I can afford, and you're going to have me up and running in weeks instead of years and congressional trips. So we're a whole different allocator when it comes to serving the government than anyone else.

Operator

Brad Sills, Bank of America.

Brad Sills

I wanted to ask a question around customer employee workflow. It continues to represent real key specific incremental growth. We certainly hear about real strong traction in the channel anything to call out there? There's a lot in those two offerings. So would love to get any thoughts from you on what's driving the strength there? And my notion is that perhaps now assisted starting to kind of pull customers into those applications? Are you seeing any kind of halo effect or pull-through, if you will, as customers are getting ready for now assist in those categories?

William Mcdermott

Yes. It's a great question. Brad, the whole agentic AI revolution is a lot like the chicken and the egg, right? Is the agentic AI pulling the core or is the core pulling agentic AI? In our case, both are working for each other, which is truly a beautiful thing.
If you look at customer workflows, they're performing extremely well, as I said that, and the industry workflows are the fastest growing in the company. And in most cases, when you talk about CRM, you're personalizing that to an industry and a specific use case, which is perfect for us. And if you think about it, field service management, a really incredible quarter, large DoD agency, multinational imaging and electronics company, a European health care company, a semiconductor manufacturer air transportation, communications and IT top energy. I mean I can go on with the list. Suffice it to say this and I want you to register this carefully.
We have the best CRM solution in the market. CRM is not SFA. CRM is how do I sell to you in every channel, how do I fulfill what I sold to you and how ultimately do I service you and we can do it on one platform. And our unfair advantage in IT is we know all the assets, we know all the operations, we know all the customer stuff in the ServiceNow platform today. And now that we can sell fulfill and service on one platform, the world is standing up and taking notice.
They have had it with literally, in some cases, big companies. They have hundreds of disparate implementations from different suppliers out there that are not integrated to begin with, but when they're not even on the same instance, can you imagine the chaos. So this is extremely pleasing. We're going to run the table and it's going to be a big, big thing. And that's where we're seeing tremendous uptake in agentic AI.
I want to tell you a story about ServiceNow because we drink our own champagne. Hear me out on this. We have AI agents servicing 85% of customer self-service in the company. We're automating 37% of our customer support case workflow through routing categorization and similarity detection. And agentic AI is transforming complex incident data into automated requests.
It saved us 400,000 labor hours already. And we have 80% case deflection in the last 6 months. I'm not talking chatbots here. We have them 6 years ago. I'm talking agentic AI agents, and they're driving developer productivity by more than 20%.
And here's the big thing on go-to-market, which I think a lot of people find interesting. The AI agents are helping to increase our prospect conversion rate by up to 16x and that's because they can curate the best leads based on the heuristics of all the data sources and put the top ones in front of the rep so they could sell more. And I could just tell you like seeing things going 99% fast to the customers answer and going from four-day response to eight seconds, is something that a lot of customers are getting excited about. So we have now on now here, a great team that does it. We did 8,500 now on now engagement with customers last year. It's estimated this year, we'll do 20,000.
And with regard to the employee, from recruiting, hiring and onboarding to providing all the services, just think about combining all the functions of a company to serve the employee and everything on the mobile but you get to know them so well, you're actually pushing solutions to them. They don't have to ask for it. But when they do ask for it, it's instantaneous. And this is freeing up 3 million hours of capability within ServiceNow. So literally, when Gina talks about -- we just did 200 basis points of op margin improvement we're thinking about more of the same because we run the whole company on ServiceNow.
And where we do have a system of record somewhere, it's literally, you have to go into the basement and take the elevator down like several floors to find it. Nobody even knows. So this is a whole different way of approaching enterprise software.

Operator

Samad Samana, Jefferies.

Samad Samana

Gina, I'll echo the congrats very well hurt. Bill, maybe first a question for you. Just you mentioned now a couple of times how the deflation in costs at the computing level will unlock budgets in the application layer and the platform where lease for ServiceNow. I'm just curious if we've been getting feedback from either early customers or those that were in the pipeline that there is a budget constraint or a tough decision as being having to be made, and that's what's leading to that confidence in the unlock? Or is it more of just a high-level observation.
Just maybe help us understand what gives you that confidence that those dollars will shift over? And have you already seen a change in the conversation?

William Mcdermott

I think the conversations are clearly -- and they have been for some time now. there's no magic ones here. It's all -- you've got to work hard. And working hard means, you know the industry, you have the best-in-class solution. You have a well-composed design thinking, business case.
Is it desirable? Is the dream there? Is it feasible? And meaning can you do it in the right amount of time. And then ultimately, as a viable, does it have the return on investment that customers need to say, yes.
And you can't skip steps in this process. You have to do it the right way. But what agentic AI is it opens up a fundamental reinvention of the business model and how you can innovate entire corporations. Just think about entire corporations rethinking what business they're in, what industry they're in, what offerings they can give to the customer, that's on the revenue side. But on the finance side, why do I need to hire thousands people?
Why don't I just put these agents to work to complement the people I already have or perhaps as I attrit people I don't backfill them. I backfilled them with agents and my revenue per employee explodes through the roof. So I got you on the margin and I got you on the growth. And that's why I am a guy that truly believes that this revolution is even greater than the Internet, it's even greater than the cloud and the iPhone moment with mobile business because it encompasses all of them. And that's why it's a $20 trillion GDP movement between now and 2030 based on the business impact it has.
It's a 1:5 ratio. For every $1 wisely invested in AI, you'll get $5 on the other side. So it really does come down to proving that your tech is there, your understanding of the business is there. The value is there. And once that's clear, there's nothing standing in the way of a decision that says, let's go.

Samad Samana

That's great. And Gina, maybe a follow-up for you. Just on I think there is already a consent development. I think early on in the proposed where if there's a certain amount of other tokens or coins associated. What have you observed from the early cohorts of customers how many getting into a level where it required the consumptive element? And maybe just help us think about how that will impact CRP or what goes into CRPO going forward?

Gina Mastantuono

So Samad, thank you for the question. What I'd say is that you're absolutely right. So our analysis already had a consumptive piece to it, right? You have deep base. And then if you use more than what was anticipated, you would have to buy more tokens.
Up till now, that extra token monetization has not been significant. What we're talking about now is very similar in that the agenetic capabilities that we're bringing out now, we are not -- we're not making a new SKU. It's going to be in the Now Assist. So those falling customers that have already bought it, have it automatically. They get to use it.
But once they start using the agentic for the assist, that's going to run the meter a lot faster. And so the expectation is that, again, the baseline subscription is required and necessary but the monetization of the hockey stick when those ages really get deployed and are really driving value when you're going to see that hockey stick of growth. So not very different than what we've been doing before. It's just now being applied to a agentic AI as well as the cord now assist. Does that make sense?

Samad Samana

Yes.

Gina Mastantuono

And then, Amit is going to add something.

Amit Zavery

Yes. I think with agentic, as I was mentioning earlier, the amount of tokens you require for a lot of this task because you're breaking the particular intent of the user into multiple pieces of the task and each of this requires a lot of work in the back end. That's where this amount of assist you require just keeps on going up very, very fast. And this is where this idea that today, your subscription will have some amount of it and they keep on adding to it as they use more of it. So it makes it very easy for our sellers to go and talk about this thing.
Customers start using. We have 1,000 per customer already using it. So it becomes a very easy scalable and win-win for both us and the customer.

Operator

Mike Cikos, Needham & Company.

Michael Cikos

I just wanted to cycle back. I know in the press release and the prepared comments, you had cited this go-to-market optimization. Can you just provide a little bit more color on that front as far as what ServiceNow has in store. And I think, Gina, to the top of the Q&A with Kash's question on impact from consumption or federal, but do the go-to-market optimizations in any way, play a role in your guidance philosophy here as well?

William Mcdermott

Yes. I think it's very important that you understand that we are building strength on strength. We still have everything just as it was in the fundamental business model of ServiceNow. The subscription-based pricing, the focus on industry-specific solutions, packaged and target for personas across an enterprise, on the best platform in the business. That's where we are the control tower for AI transformation.
In addition, we have brought about RaptorDB, which is an incredible database. Based on postgrads QL, open-source standards, we acquired a company called Swarm64. We got the best database watchmakers in the world to put some secret sauce into making that a ServiceNow sensation. On top of that, we have the connectors into all the other major data sources in the world from the hyperscalers to Oracle, all the way through to Snowflake and Databricks where we'll be able to use that data and the automation of work. That is yet a net new business for ServiceNow.
It also happens to be a $250 billion TAM, and we're in the early innings, but we love what we're seeing. And on top of all this, we have this beautiful platform that has been enabled with what we call Now Assist, which is our genic AI portfolio. So when you as a customer, procure our Pros version of the platform that you've grown to love. You now are going to get these agentic AI agents with that platform. We will also give you some ability to look at thousands of different things you can do with it.
That's included in that subscription price. But as you do more and more with it and more and more with our agentic AI agents, a meter kicks in, and we make more money on top of the core. And the customer gets tremendous return on their investment because now instead of people doing all this stuff, they have their agents working for them, and they wouldn't do it if it wasn't a great business case anyway. And the consumption is smart because if you're using it, you're doing it and you're getting value from it. So now we're all on the same value page.
So that's that. So we still are going to market with Paul Smith, who runs our global field. He also became a President, congratulations to Paul. And we have a gentleman that is named Paul Fipps, that is running the regions to give more bandwidth to the executional excellence of running the number every quarter. even as we build out new dreams in harmony with Amit and Paul to invent the $30 billion version of ServiceNow.
And so what I'm really super happy about is we're not just focused on big enterprises. We have a great commercial business. And we also have a great direct-to-consumer business that we're building out because we're very happy with $50,000, $150,000, $200,000 deals on top of these $100 million. And we want both high value and high volume. And when we put this all together, you're dealing with a company that's performing at the Rule of 54.
You're dealing with a company that's blown through $10 billion on its way to $15 billion plus by 2026 that has a bold ambition for $30 billion plus going forward, but I would say a great management team. And that's all part of the go-to-market incidental because no grass has grown underneath our feet. We're not measured by how many e-mails we got done in the office. We could do them in the back of a call on our way to a customer meeting. So we brought in the best engineers engineer and the great leadership of Amit and who incidentally is also our Chief Operating Officer, in addition to being the President and lead of all of our engineering efforts, and he is an expert in product packaging and pricing.
So that interplay with our innovation and go-to-market has never been sharper than it is right now. So that's kind of an outline of what's going on, and we're putting agentic AI to work in all these processes. I'll give you an example. We just did a sales kickoff meeting in Vegas. We had about 8,000 of our closest friends there.
And we showed them how AI is going to do their territory management for them. So they'll take all the use cases, all the data that exists already and they'll compile their road map of first things first with their customer relationships all the way down to here's the solution. He is the best-in-class offering and he is the customer you need to take it to. That's cutting out so much of the busy work that the sales professionals are super excited. And that's true of all position types in the company.
Does that help you?

Michael Cikos

It does, Bill.

Gina Mastantuono

And Mike, just from the -- your specific question on guidance philosophy, of course, everything we're talking about is factored into our guidance. top to bottom, east to west and north to south, all of these impacts and thoughts are very thoughtfully put into our guide. So very specific answer to that question for you.

Operator

Derrick Wood, TD Cowen.

Derrick Wood

Great. I guess, keeping on the same theme, Gina, I just want to unpack your comments around foregoing new subscription revenue for basically incremental consumption of agents, but it will take time to build up on the consumption side. Is the message here that you expect some kind of near-term headwinds in seat growth as more workloads are being pushed to agents and that it just takes time to start really moving the needle on the consumption side. And if that's the case, just how long do you think this will take for this new hybrid pricing model to be net accretive versus kind of your previous Pro Plus sales motions?

Gina Mastantuono

Yes. So Derrick, thanks for the question. Just very, very clearly, not foregoing new subscription revenue. At the end of the day, Now Assist continues to grow extremely well. We talked about 150% growth quarter-on-quarter, fastest new growing product launch of our history, and we absolutely expect that to continue to grow.
The agentic AI hockey stick of value is going to take some time as customers start to use it. And so we're absolutely not foregoing. We're continuing to have strong subscription revenue growth, 20% constant currency at our scale at $12.675 billion never been seen before. And on top of that, you will get the hockey stick of a agentic AI multiple growth in as customers really start to get value in it. So it's an and not an or, and I want to make sure that, that's clear outside the gate.

William Mcdermott

And may I just say one thing, Derrick, to build on what Gina said, and I Amit covered this earlier, but I don't blame you for asking questions because one thing I want to make sure everybody understands, if we didn't have any consumption attribute to the pricing model, everything stayed the same. You still have to subscribe to the Now Assist portfolio, and that is the same subscription model we have today. The nuance here is we've actually added what I would call an accelerator to the consumption idea because my idea really is that they'll buy more subscription because they're actually getting the agents included in it. And then as they consume that subscription, and I met very nicely said, they will quickly because they're going to get so much value from it, we actually then kick in an accelerator or a turbocharger to the existing business model. We still have the existing business model easily as it is.
And someone asked a great question, it could be that knowing with the Now Assist portfolio and the speed of our innovation, they get the agents for the subscription, that might actually significantly increase the subscription. We just got to let it play out. But in any case, once they start consuming those agents and those processes are being done by the agents and those assists are happening at mass scale. From an investor standpoint, you're going to really like ServiceNow a lot more.

Derrick Wood

Okay. So yeah, it sounds like you're really building up a springboard that's net new that the hockey stick hopefully around the corner. That's clear. Thank you.

William Mcdermott

Thank you, Derrick.

Operator

And ladies and gentlemen, this concludes today's call, and we thank you for your participation. You may now disconnect.