Hey, good morning, everyone. Tyler Radke, Citi’s co-head of U.S. Software. Welcome to day three of the Tech Conference. We are happy to have Veeva here. We have Brian Van Wagener, the CFO of Veeva. Brian, I think you have rejoined the company. It has been what? A little over two years now. Just give us a quick background on your time at Veeva. What have been kind of the key accomplishments?
Well, thanks for having us, Tyler, and thanks everybody for joining today. Yes. So, I have been back at Veeva for about two years now. I was here for almost six before that. So, eight years in total at the company and have seen it really grow up. Our vision, for those of you that may not know Veeva as well, is building the Industry Cloud for life sciences. We think of that as applications, agents, data, and consulting working together. We are a vertical software company serving the life sciences industry predominantly.
That is the vast majority of our revenue, roughly 98% of revenue across two, I guess three major application areas. One is commercial, the products that allow our customers to go to market with their generally biopharmaceutical products. Second is R&D, the development side, so clinical trials and the processes that bring products to market, a nd then quality, which is related to their manufacturing organization.
We are split across those three areas into 10 different suites, well more than 50 products. We have software products and AI products, so it is a pretty diverse portfolio in the business. I think stepping all the way back to when I started in 2017, it was only a 15-product company across five different suites, and so it is the continued ability to innovate and to run a business that is increasingly both broad and deep, and I am pleased with the progress that we are making. We have really accelerated our AI roadmap in the last 18 months as the underlying model technology has matured. We are making really good progress against that.
The core business is continuing to execute well, so we are continuing to see really strong growth across our software suite as well as our services, which had its highest growth in two years last quarter. So, really good execution across an increasingly broad range of opportunity in life sciences.
Yeah, certainly. I think one of the things that stood out to me just when AI first burst on the scene, I guess it was GenAI at the time. I think Veeva took a little bit more of a wait and see approach versus the hype from vendors, and I am sure we can all guess who hypes AI out there. But now it does seem like you guys are really talking about it more. We are starting to really hear about use cases, new monetization vectors. Can you just talk to us at a high level, what is Veeva's AI strategy?
Yeah, absolutely. I think that view of maybe Veeva being wait and see was probably a common opinion. I would say we were more patient with it. One of the things that is a little bit different about vertical software versus horizontal is we make our money selling many applications, which means if you sell one, it has to work because you need to sell the next 49. For us, that reputation of having things that work as advertised is essential. They have got to deliver value, and they have got to work the way we promise. It was pretty clear early in the days of AI that the models were not mature enough for a highly regulated, large enterprise setting like life sciences.
Yeah.
I think what you saw from us was not falling victim to the distraction of some of that early flailing around that happened in AI, where we are all worried about hallucinations, and you spend a lot of money on these product development cycles, and you generally do them at the expense of something else. Our view was continue to invest in the core applications, continue to invest in maturity there, in scaling those, and then really build AI when the underlying models are ready for the environment we work in. That has really changed in the last 12 months.
Yeah.
I think you saw last year us notably accelerate our Vault AI roadmap at the time, which is agents that sit inside of our applications, generally helper agents that help you do work faster and to be more productive. About three months ago, we announced Falcon, which is a little bit broader vision of agents that do most of the work.
Yeah.
End to end, you can think of it as agentic or digital labor rather than a helper agent. That I think is pretty expansive and will ultimately cover every product area that we sell into and can be a pretty significant financial opportunity over time.
Yeah, I would love to stick on the Falcon topic because I agree, I think that was one of the more marquee announcements we have seen out of the company in quite some time. It is early, right? I think you have some early customers, but what makes Falcon different from a productivity tool? Agentic labor implies that this is taking over labor that humans used to do. Talk to us a little bit about the vision. Where do you see kind of the most addressable use cases?
Yeah. Ultimately, I think we see use cases across the entirety of the business, and so maybe I will give a couple examples of what Vault AI can look like and compare that to Falcon. A great example of Vault AI in practice is our agentic call report that sits inside of our CRM product. We have got top 20 customers using it today, one using it across the entirety of their sales force in the U.S. What that helps their sales team to do is to not have to click around into 50 different records, synthesize all that, and remember what to do with the doctor they are going to see. They just type it in, the agentic call report does all the preparation, delivers them a clear synthesis of the history of that doctor, their prescribing patterns, and what it suggests you do in that call.
Super valuable, saves a ton of time for the rep, right? Might save them 30-45 minutes getting ready for that call, which adds up across a 1,000-person field force. But it is ultimately a productivity tool inside the application. What Falcon does in comparison is it does the work largely end to end. In general, these are critical enough workflows that they are still going to be reviewed by a human at the end. But this would be taking something like an inquiry from a healthcare authority that says, "Hey, we have this question about the clinical trial you are running." It is a serious question, so you need to pause your clinical trial while you are responding to this.
Instead of a person having to go, or a team of people go and receive it, figure out what the response strategy is, research it, develop the authoring, get it all approved, and submit it, we do that with an agent end to end. We think we can take the time for some of those workflows from, in that case, a month down to under a week.
You have both a labor savings from having the agent do this versus highly paid people, as well as a speed and productivity benefit to the company.
Yeah.
These are very high-value use cases, but they are different because these are more end-to-end workflows
run over potentially long horizons, versus something that is a tool helping an individual inside the application.
Yeah. Veeva's product portfolio, very extensive across commercial R&D. You talked about a regulatory use case. You also are in the clinical data management, quality safety, et cetera. Do you see a vision for Falcon to kind of touch all of those dimensions across the business?
That's our expectation right now, is that Falcon will ultimately be embedded in each of our areas.
Yeah.
It's a standalone platform that we've built outside of our main platform we call Vault. So Falcon sits outside of Vault and is a power user of our applications.
Yeah.
The early adopters that we have right now span clinical, quality, regulatory safety, and we also have Falcon agents inside our content area of commercial. We call that our MLR agent.
Yeah.
Ultimately we'll see that expand, and we're likely to have coverage across every product area, but also multiple agents within each product area.
Yeah. As we think about the technology that enables this agentic labor technology through Falcon, I know you've made a couple of acquisitions, Copli and Ostro. Are those kind of the underlying pieces of this Falcon vision, or does each area have its own approach from a technology perspective?
I think we're going to be opportunistic with M&A, but we don't need M&A to achieve our vision in Falcon.
Yeah.
I think we are planning to go out and build organically, but we found opportunities in Ostro as well as in Copli to accelerate in specific areas. Ostro is not part of Falcon, it's part of our commercial suite. It's precision AI that's compliant from a content perspective. If you are a brand and you want to have an AI chatbot on your website, you cannot have that hallucinate.
It has to work within specified content parameters. Ostro has really mastered the ability to do that at scale for these very sophisticated companies. Ostro is really leaning in early to a very new way that patients and doctors are gathering information, and we want to be at the starting point there of that journey. Copli was a really interesting company out of Europe that we felt had developed a really great product in MLR that was going to be similar to the kind of thing we were going to build in Falcon, and it made sense to accelerate that roadmap.
We can help with scaling that more quickly. We can help with embedding it more deeply with the Vault applications. There may be more of those.
Yeah
We're going to do that opportunistically rather than as the main pillar of the strategy.
Right. I guess that the MLR agent, is it kind of a common foundation across the other Falcon agents, or should we think about this as kind of specific to MLR?
Fairly specific to MLR. MLR is medical, legal, and regulatory review. It's a specific process that our customers have to do inside of the content area. All of the content that their sales reps share, that they put in an advertisement, has to be reviewed by medical, legal, and regulatory groups within the organization. That process is time-consuming and cumbersome. That's three acronyms means three different groups that have to review it. We think we can do all that in an automated way, and we're doing that with Copli right now.
Yeah.
Our vision over time is to eliminate about 70% of the manual labor that's being done on MLR across the industry.
Yeah. Thinking about that, I think the labor TAMs certainly are way larger than software TAMs, right? How do you think about pricing agentic labor in that context?
I think we've got a couple different pricing strategies. We talked about the Vault AI, which is agents inside the applications, and then Falcon which is the agentic labor. On the Vault AI side, we are thinking about that right now as consumption-based. It is based on tokens.
Okay.
That is going very well. That may evolve over time, but for now, that is our strategy. In Falcon, I think that is more likely to look like a business transaction. So in MLR, that might be a document reviewed, or in the health inquiry, it would be an inquiry that was resolved and filed. It will be some business transaction that is relevant to that agent. A lot of that we are still working through.
It is really important when you are doing something new like this to evolve a little bit and be agile in your pricing strategy. So our current state is we are very focused on product excellence, getting the products to the point that they are really mature with a set of reference accounts, and then building the commercial model on top of it.
Yeah.
But there are still a couple more steps in learning and likely some evolution there. What I do think that that model of pricing allows us to do is to get away from some of the really low or even negative margin that you have seen emerge in AI and other spaces, where you end up just getting marked down to what the underlying model cost is from Anthropic or OpenAI or whomever. I think by creating value through the Falcon agents and by talking about the business transaction, you can create a product that still delivers high margins.
Yeah. It kind of ties into that business outcome-
That's right.
... type pricing strategy.
Yeah. I think the thesis there is if you're creating a lot of value for customers, we should be able to capture a fair share of that.
Yeah. No, it's super interesting. Shifting gears a little bit to the core businesses, commercial and R&D. I think certainly the commercial growth has surprised a lot of folks. I mean, consistent double-digit growth. I think a couple of years ago we were concerned this would be slowing into the single digits just given the maturity of the CRM franchise. Maybe help us unpack, what is driving that sustained double-digit growth? Would you say it's more of your own execution? Obviously, the end market currently is quite strong in terms of trial volumes and everything. So, how would you just sort of frame that growth profile?
Yeah. We had commercial subscriptions growing about 13% year-over-year in Q2. That's a little bit higher than our long-range expectations, which were more in the high single digits for commercial. I would say it was a pretty broad-based strength that we saw there, Tyler. Some of that is the continued strong performance of our Crossix business, which is on the marketing and media optimization side. It helps customers to measure and optimize their media spend, as well as to place their digital media spend more effectively. That business is a growing end market and a place where we can continue to capture share. So we see that as a durable grower for many years. It's accretive to overall subscription growth and certainly to commercial subscription growth. So it's a key driver of that growth rate for the long term.
We saw, even outside of Crossix, strong growth across the commercial business. Still growing in CRM, the rumors of its demise notwithstanding, we're continuing to grow in CRM. We're seeing large accounts, in some cases, growing their field forces. We're continuing to win with new accounts in the SMB space as new companies form and commercialize. We're continuing to grow in our data business, where we have a number of products that can grow for the long term. We still see add-ons in places like content, which are very mature at this point. So it's no one single thing. It was really broad-based strength across it, and I would say that that's the Industry Cloud story coming together. It's that these products work so well together that customers are continuing to invest in their commercial capability broadly.
Yeah. I would love to spend a little bit of time on Crossix, because I think it is an area that when you acquired it several years ago, many years ago, it sort of was not a small deal by any means, but sort of flew under the radar a bit, and now it is kind of burst onto the scenes, just given the success of that. Where are those dollars coming from? Is this share gains, like consolidation from other platforms? Again, is there some element of cyclicality that you would be caution of us over-extrapolating the recent success of it?
Yeah. Crossix is, it is a great acquisition story, but it is a great business, too. The original deal thesis from when we brought the Crossix team in, I guess about six years ago now, was that the IP they developed would be the foundation for what is now our Compass data business. We felt the Crossix business was attractive in its own right, but I think it has surprised us, with the investments that we have made and the good execution from that team, just how significant that business can be. When you think about marketers and advertisers, of course, they are going to measure their spend and try to make sure they are spending more effectively. That service can be done by third-party providers. Sometimes it is done by the agencies they work with.
But there was really a need, an unmet need in that market, for a strong third party that could do that very effectively. To do it in a way that is now integrated with CRM, right? To build some of those linkages to the CRM and the marketing platforms over time. So that media optimization business is the larger area of Crossix and has been a big driver of growth for the past several years. Last year, Crossix was a key part of the outperformance of the full fiscal year, and that was really driven by the other side of Crossix, the Audiences business.
That had been sort of a nascent business when we took over the Crossix team. We made a number of investments in the product over the last two or three years, and we saw that really explode. We saw both a healthy end market again, but also a lot of share gain in Audiences last year, which is the other side of digital marketing. If optimization is about measuring and then optimizing the next year's spend, the Audiences business is around placing your digital spend up front with the right segments. Our product is pretty unique in that it allows them to develop really precision segments for the specific therapy they're selling. You want to find cardiovascular patients in the western half of the U.S.? We help you build that segment.
Wow.
It's precision targeting in healthcare that's still HIPAA compliant and anonymized.
It meets all those regulatory requirements. It's super unique, and as we have success with that in helping customers get higher ROI on digital spend, they of course want to invest more behind it.
Yeah.
It's a market where success can feed growth of the market.
It sounds like there's some real kind of secular drivers on that, more than cyclicality.
Yeah. That's right. That's why I think you hear us consistently say we think that can be a durable grower.
Yeah.
We think there's headroom for growth from a share perspective. We also think the market itself will continue to grow.
Okay. Shifting to CRM, obviously you get a lot of questions around top 20s, and I think at least we won't be having to track those top 20 wins for much longer, given we're almost through those decisions. How have the go lives on Vault CRM been going versus your expectations? I guess for the folks that have gone live, what are the benefits that they've seen versus the legacy Veeva CRM?
Yeah. For those that may not be as familiar with the Veeva story, our very first product was CRM built on the Salesforce platform. We are in the process of migrating customers now from that Salesforce product to a Vault CRM product on our own Vault platform. That means we are now competing with Salesforce, and there has been a fair amount of noise, I would say, around that for the last couple of years. We are making really great progress against that. Sometimes the top 20 is used as a yardstick for the overall migrations. Right now of the top 20, 12 have selected Veeva, six have selected Salesforce, so there are two to go. We like our chances in both. We expect that those will largely be decided and announced by the end of the year. That is going really well.
More importantly is not the decisions, it is the execution against the decisions that have been made. We talked very early in this about a decision for Veeva is a little different than a decision for Salesforce. We are very confident that we are going to get customers live and happy and having success, and we are seeing that play out. We have got more than 180 customers live on Vault CRM. We are not aware of any customers that are live on Salesforce CRM products. We have got numerous top 20s live, five that are live in different regions, two that are live globally across all regions. So it is going really well.
The migrations are going well, and I think the feedback we have heard from customers is that the migrations have been very smooth, they have been on schedule, they have been on or under budget, and that the product is delivering on its promise of innovation in AI, but also in the core areas of the application. I think increasingly as we get further down the road of that migration, we can shift even more of the resources away from the support of the legacy product into the new product. That acceleration in innovation is the thing that I think we get most excited about internally over the next few years.
Okay. Not to introduce a third CRM name, but you did recently introduce Aspen CRM. Talk to us a little bit about the strategy there. You talked about this earlier, lots of products, lots of things going on. How much of a priority, how much investment dollars are you putting behind Aspen?
Yeah, just to keep you on your toes, we brought another name in. Aspen is something different for Veeva. We announced it about a year ago, I think, at our investor day last year, and that is a move outside of life sciences. Everything we have been talking about so far is the life sciences space, which is 98% of our revenue today. Aspen is a move into horizontal software. Our first product in Aspen is going to be in CRM. We have got early adopters that we are working with right now. The team is making great progress. We run that as a startup inside of Veeva. It has got a lot of autonomy, it has got a lot of focus. It reports directly to Peter, and we do that pretty deliberately.
It allows them to be able to work very differently from big Veeva and function more like a startup with short-cycle roadmaps. It also means that Peter is working directly with them versus filtered through other things. That team is doing some really exciting work, and we think that the CRM market is ripe for a new entrant, and not everybody is going to agree with that. I think we are comfortable taking an unpopular opinion on that. I think as we look across CRM, I do not talk to many people that say, "I love working with my CRM. It really helps me.
It is well-designed, it is made for me, it is working effectively." We think there is an opportunity to do that much better, to make a product that works better like Zoom did with web conferencing several years ago, and to sell that in a way that is much cheaper than the products on the market are today, and that is much more nimble and easier to evolve and sustain over time. It is early days there.
I do not view that as a near or even midterm, meaningful financial opportunity. That is more of something that plays out in a post-2030 timeline that we are planting the seeds for now. We are really pleased with the progress that team is making.
Okay. Brian, I would be curious your take just on some of the big announcements from your peers in the CRM space around headless and partnerships with Anthropic and seemingly giving up the UI layer, may be one interpretation of it. How does Veeva view that strategy? How important is it for you to own the UI layer for your customers?
I think of it less about do you have to own the UI layer than are you meeting the customer and the user where they are for the business process that has to be done? So, our strategy is maybe a little bit broader than headless, which we have called dual mode, which is just to say, yes, you can access our applications through Claude or ChatGPT or via an MCP server and navigate them that way. F or some use cases, for some queries, that is going to be the right thing to do. But most of my history at Veeva was leading our sales operations team, so I could say personally, I would never want to give up the entirety of the UI. The UI is how you focus your team on the job to be done. These are the steps that you need to follow.
These are the priorities for right now. UI adds a lot of value to many workflows and many processes. I do not think it is either/or. I think it is the ability to support both, and that is how we are thinking about the development of our products, that our customers would be able to use those MCP servers to navigate through their preferred AI pane if they want to do that, and also have a UI in the application that it is using. T hen the third is Veeva Falcon, which becomes a user outside of the individual human that is using the application. There is a lot of ways that the applications will get used, we think.
Right. Maybe we could shift a little bit to the R&D side of the business. I think for basically since the IPO many years ago, that had been a 20%, in some years much higher than that, type of growth. This year, the guidance assumes that that growth rate does take a decent step down, I think, into the mid-teens or so. Help us understand what is driving that, because on one hand, we look at the TAM, right? These are greater than $10 billion revenue markets. You are not highly penetrated there as you are in core CRM. So why is growth slowing if there is still so much more room to go?
Yeah, I think overall it is playing out largely as we would expect it would be the headline there. When you think about R&D, we have a lot of products, but there have been sort of five or so that have driven the majority of growth to date, and then some add-ons that sit around them. So eTMF, CTMS, Veeva QualityDocs, QMS, and our regulatory suite have been the five that really fueled growth to date in R&D. They are large products, but they are getting to the point where they are at maturity. eTMF has 20 of the top 20 as customers, so there are some ramps continuing to come in, but there is no more selling to do on eTMF. It is just about making sure we maintain that position of product leadership. CTMS is not that far behind it. The quality products are not far behind it.
Those products are at a pretty high level of maturity, and growth from here on out is driven by a different set of products. It is driven by products like EDC and eCOA and RTSM. I am sorry for the acronym soup.
Yeah.
But it's driven by a number of products across clinical quality and safety. I think all you're seeing is that those S- curves don't just stack up exactly right.
That it takes some time to get the new products to the scale that they're fueling the overall business, and that's largely what we expected. We alluded to that at the start of the year pretty directly when we gave our guidance.
Yeah
Q3 is where you start to see some of that step-down happen.
Yeah. How do you think about, clearly you have long-term targets out there that embed some level of S- curve, I'm sure, but how do you think about this sort of gap in those S- curves, and is there something we should track? I know you guys do a good job of usually calling out these top 20 wins. Are those the leading indicators of these newer products starting to see ramps, or how should we get comfortable around that, hey, that S- curve is coming not too far down the road?
Yeah. I think the main thing is obviously to look at the momentum of revenue in the business, and we're pretty transparent with how we call the business. We don't put a lot of hype into our numbers, so we call it like we see it, and we're very direct on that. We'll give guidance for next year as we get closer to next year. We remain confident in the 2030 targets that were our long-range targets we announced a couple of years ago, just not quite a couple of years ago. So I think we're pleased with the progress that we're making on these new products and the trajectory that they're on, and that we're still on pace.
Yep. Got it. Maybe as we think about tying this back to the broader vision, and you led off saying Veeva is a vertical software maker, right? You've got to make sure all your products work together, and I think that was a big piece of also this Veeva CRM to Vault CRM, to kind of have it all on a common code base. Talk to us about, are there customers that have gone all in on Veeva? What does that look like? Is it a percentage of their wallet share? And just how do you think about how big this company could be if that vision's achieved?
Yeah. I tend to think about it more on an area-by-area execution basis. We have a lot of different types of customers that we work with. So when you think about any large company, they have different ways of working. Some are more top-down, and they drive things programmatically. Some are more decentralized, and they would operate application by application for us. I think part of what's important for us to do is to be able to work in either mode, to be able to work with the CEO and the board when that's a conversation that they want to have around transforming their company, and to be able to work with the director of an area of clinical operations when they want to modernize one clinical application. There's no one right or wrong way to do it, and we work with our customers in both ways.
We do have some customers that have said, "We are going to standardize on Veeva. If there's a Veeva application that's fit for purpose, we're buying it." We have memos of understanding and contracts of different structures to do that. We have customers that have said, "I'm going to standardize in one area." We have customers that continue to go application by application. I think ultimately, our goal is to be the standard for the industry in every application we build. So that yardstick of eTMF supporting 20 of the top 20, that should be our yardstick.
It should be 50 of the top 50. We've got to earn that. It's going to take time to do that, but our goal is to be that kind of industry partner to all of our customers across every area.
Yeah. I did want to touch a bit on profitability. Brian, I think there's some debate in the software industry. Is AI a profitability booster or detractor for software companies? On one hand, you all are taking advantage of the efficiency that AI gives you in terms of operations, coding, et cetera. On the other hand, AI revenue has more COGS, and certainly AI talent is expensive. How do you think about as Veeva becomes more of an AI company, what does that mean for profitability, knowing that your long-term targets are, I'd say a little flexible, with the plus above, but they're a bit below where you are today.
How should we think about Veeva's profitability, moving forward?
Yeah, I think on the long range, we don't give long range margin targets. We have a floor that we talk about to give people confidence we're not going to have an excessive investment cycle. But I think we've always thought about balancing growth and profitability. If you look at the early days of Veeva, I mean, Veeva was profitable very early on.
A little bit unusual at the time for a tech company. That's pretty deeply embedded in the DNA of what we do. When I think about AI, I think what I think about is how do you build products that deliver clear and differentiated value to customers? Because if you do that, you can charge for it. If you're delivering commodity AI, then you're going to get marked down to the underlying commodity of the token input. But our Vault AI agents in the applications are definitely not commodity AI.
Right.
We have full product teams building these things to work in pretty sophisticated ways. The Falcon products absolutely are not commodity. They're going to be taking significant amount of manual, repetitive labor cost out of our customers' operations that they can use to invest back in growth or pull through to the bottom line. They'll make decisions on how they want to redeploy those resources.
But I don't worry about the ability to drive a profitable business when we deliver a product like that that works. Where exactly that lands, I think is a little bit too early to say. Peter, our CEO on the last call was pretty direct in saying there is not a reason to think we cannot get software-like margins out of the Falcon business in particular over time. That does not mean we will. We have got to go and create enough value that we can do that, but I believe we can.
Yeah.
It starts with picking the right use cases and building the agents to deliver enough value that we can capture a fair share of that, for Veeva.
Yeah. No, that makes sense. There are a lot of events coming up for you in the next few months. I think you have both your R&D summit and commercial summit, if I am not mistaken, as well as Investor Day, probably in early November, if typical timing holds true this year. What would you encourage investors to focus on, both from product announcements as well as what to look forward to at Investor Day?
Yeah. We're excited about all of those. We've got our Europe Commercial Summit coming up a little bit later this year, North America R&D coming up, and our Investor Day on November 5th, Gunnar? November 5th. We'll hope to see some of you there too. We I think have covered so many of the topics that are of focus for us. It is a broad-based business at this point.
Yeah.
Ten application areas, 50 applications, an entire new area of AI that we're opening up in Falcon. I think really just leaning into the progress that we're making in those areas, the roadmap that we're talking about for Aspen. We're looking forward to giving more detail on those things obviously as we get a little bit closer. It is an exciting time for Veeva. You go through these waves in companies like ours where you are innovating and expanding the aperture of the business and then where you're focused on execution of the opportunity in front of you. It feels a bit like we're doing both right now, which is exciting, but we're certainly expanding the aperture. There's a lot of new things happening and I think also a lot of really strong execution of the things that we've been known for the last couple decades.
We'll look forward to sharing more about that with you at Investor Day and some of the announcements coming up there.
Yeah, look forward to it. A lot of exciting things ahead, Brian. I think it's a great place to wrap up as we're out of time. Thank you for a great discussion. Thanks everyone for joining, and I think we have the lunch keynote kicking off here in a few minutes. Thank you.
Thank you, everybody.