Session started off. So my name is Adam Wood. I look after European software research in Morgan Stanley. It's a great pleasure to welcome Rouven Bergmann, the CFO of Dassault Systèmes. Rouven, thank you very much for joining us here in San Francisco. Appreciate you taking the time.
Thank you, Adam. It's always great to be with you.
So let me get the disclosures out of the way to start off with. So for important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. So I can slow down with the regulatory stuff out of the way. Maybe let me start off, Rouven. I mean, I think the exciting thing about Dassault at the moment, one of the exciting things about Dassault, is this movement towards 3DEXPERIENCE and this upgrade cycle that the company seems to be in. It felt that you kind of had an early wave of adoption with the deals of Boeing and several other companies kind of four or five years ago, maybe a little bit longer. It feels as if in 2023, we saw kind of another acceleration of that upgrade trend.
I mean, first of all, is that kind of how you see things? And then secondly, can we dig into a little bit around what companies are getting when they go to 3DEXPERIENCE? You're very strong in CAD. What do they get when they take that broader platform from you, the benefits?
That's a great way to start off. You know, 2023, I think, was an inflection point for us where we saw strong demand of our platform. In fact, 2/3 of the growth in 3DEXPERIENCE revenue was driven by deals larger than EUR 3 million. I think this is a very strong indicator that more and more of our installed base and existing customers, they see the value of moving to the platform. And you're right. There have been cycles, or let's say perceived cycles of adoption. But we are in an enterprise business. And our customers, they run very complex operations. They go themselves through technological cycles. And they are going through their steps of evaluation, of validating what we do, of seeing the references. And it takes some time to make decisions. But in 2023, we saw a number of those decisions that came through.
I want to be clear. Why this is happening is not because we are offering incentives or discounts on bundling products for customers to move to the platform. That's not what happens. What really happens is there's a strong value proposition behind. There's different stories for different customers because they have different priorities. Ones who want to consolidate their value network and supply chain onto one system because they need to match supply and demand in order to drive more agility into their supply chain and their ability to deliver end products to customers. Others have the need to simplify their engineering practices because it has been tailored along various programs or various brands of large OEMs. Now they see, even though they've been talking about platform strategies for a long time, but the engineering practice has been operating in silos for a long time.
So there are different stories behind each of the adoption journeys. And the good news is for us that we have the flexibility and the platform to serve all those different demands. We don't need one scenario to follow. We have multiple scenarios to come to the endpoint. And so we saw a strong trend in the auto sector. We continue to see a strong trend in the supply chain. We see a good trend in aero and space, even though we didn't have too many large deals in 2023. But we had a number of smaller midsize transactions, which they added up to double-digit growth at the end. We see an emerging trend of introducing Virtual Twin technologies into the life sciences space, which is a breakthrough technology when it's about modeling and simulation of how drugs are delivered to patients.
So there are many different use cases where the platform becomes so relevant for our customers that gives us very poor applicability. That's always how we have thought about the opportunity. I think this answers the first question, but also what do customers get? Customers get a lot of value. The journey can be very different. It can also be hybrid, on-prem, and cloud. It doesn't have to be all on cloud or on-prem. We have the parity of our platform to be deployed on-premise and in the cloud. I think that's a very strong competitive advantage. We are cross-industry. We are disciplined as it relates to pricing. I can say that because some of the OEMs who negotiated very hard with us, of course, they always want better unit prices. But you can gain better unit prices if you are also committing to larger deployments.
If your visibility to the deployment is more limited, then we will have a more iterative approach. But at the end of the day, we're not giving unit economics because of that. We retain the value of our platform. And we'll deploy it over time. So the upside is still with us.
You've talked a little bit around the industries and that shift to platform. Could you talk a little bit about it financially from your point of view? If we take a customer, you're obviously dominant in the CAD space. As customers go from CAD to take the software engineering side, manufacturing simulation, could you give us a good idea of which ones are the big drivers? Then financially, if someone's paying $100 for CAD, what would it mean going to 3DEXPERIENCE for them over a few years?
Yeah. Yeah. Yeah. I mean, for the large transactions, this obviously varies a lot. It varies a lot. For us, the way we measure the value is if we really think about the universe of potential users at our customers, how far can we expand? How many people can we touch? And that, of course, requires to perfection multiple domains. With one domain, you don't touch many users. They are very limited to those who are experts in the domain. But also, domains are converging. You talked about design and simulation. We talk about MODSIM. It becomes an iterative way of combining design and simulation. We introduce virtual twins to really create the data representation of objects.
Those data representations, we connect to all different functions across the company, so for example, to cost controlling and purchasing, to optimize the supply for certain parts or materials based on where parts are used in the car, for example, as a Virtual Twin of a vehicle or twin of components or parts, how it can be optimized. So at the end, for us, the question is, how can we touch more people with our customers? What are the domains that we need to master of MODSIM, as an example? MODSIM is the foundation to build Virtual Twins. PLM is the system where all the information is captured. And then you create the engineering bill of material, the manufacturing bill of material. You simulate manufacturing. And you are accelerating the time from idea to the first product delivered to the customer.
So these are the starting points and end points or the steps we can take. We have the supply chain that we want to connect as well. So there's a lot of space for us to cover. Now, when you think about going from legacy to 3DEXPERIENCE, the uplift can be multiple. On average, we see an uplift of 1.7 x. In very rare cases, it's actually like for like because there is a customer who goes to the 3DEXPERIENCE platform, expands their mechanical design, which was cut in CATIA, to combine mechanical design with electrical design, material science, and simulation, and systems engineering of optimizing systems of systems. Whenever there is a change in a part, you want to replicate this change in the entire system in terms of understanding the implications.
Going through that cycle that, in many cases, is required for certification purposes, just makes it much, much faster if you operate in one environment than you have to go through multiple different environments. This starts to become very clear to our clients. We had multiple RFPs in 2023. There are a number of RFPs going on in 2024 that look very promising to us where this level of integration is required to transform the companies and make them more agile. So I think 3DEXPERIENCE is now in the market for more than 10 years. I know you've been asking me always the question, why is it not going faster?
That was my next question. You've preempted it. Go on.
But today, 20% of our large clients that we either serve directly or we are supporting through a high-end partner network, only 20% of them today are using this 3DEXPERIENCE platform. Now, is this good or bad? Of course, you see, we are taking a very long runway to get to where we are. But the opportunity is not lost. We are building more and more opportunity as our platform becomes more comprehensive. And we can extract more value from our customers through that. And so we have a 5x adoption multiplier in our installed base. And of those customers who are using the 3DEXPERIENCE platform, they are capturing probably 30%-40% of the potential value of what they could do on our platform. So just looking at our top customers, we have 330,000 clients. And I'm just looking at the top end of the customers.
We have the ability to almost 10x the revenue in that installed base. And not through product bundling, it's through real value creation and understanding use case by use case on how can we create value for clients. That's very important. That's also why sometimes sales cycles can take time. They are very competitive because when they happen and when they're concluded, they are decisions that will last for a long time. And I mean, we all know, those who are following enterprise software, once you adopt a certain standard, you are very much in this standard. The cost of switching is very high. That's why the evaluation can take a long time.
Some of the stickiest software on the planet, right?
Exactly.
You alluded to RFPs being strong in 2023, continuing in 2024. Any comments you can give us around pipeline as you go into 2024 versus 2023, how confident you are these standardization deals, platform deals continue?
As I said, I think the key data point for me was to see that 2/3 of the growth comes from deals larger than EUR 3 million from our installed base in 2023. This we have not seen before. This gives me confidence that we are at the next level of the adoption curve. The pipeline looks certainly in the installed base and the demand looks very good for 2024. In a way, the similar structure, it's more geared towards the second half than the first half. I think the demand for 3DEXPERIENCE is strong. Customers are taking note of what we're doing with BMW. They're taking note of what we do with Renault. Many experts in the auto industry, they have written off Renault for quite some time. But now they are starting to become a poster child of what transformation is about.
People take note of what have they been doing that others still need to do. I think the 3DEXPERIENCE platform and their bet on the 3DEXPERIENCE platform plays into that. The feedback we have from them is fantastic. Again, these are the references that our customers take note of. Those who are still yet to make the next step to the 3DEXPERIENCE adoption, they will look at all of those references and understand what are the motivations and what are the strategies of the different players. But of course, the next step for us as well is to go much more strategic and focused to the suppliers, the large-scale suppliers, which we are working already a lot with. But there's much, much more we can do.
Perfect. That's really helpful. You alluded before to the variety of implementations that you can run. So it can be on-premise. It can be hybrid. It can be full cloud. I think there's sometimes a little bit of confusion around payment models and delivery models with the market. And sometimes, because a competitor has forced a migration to subscription, there's a kind of feeling that, oh, there must be ahead in cloud, which is not the case. They've just moved a payment model. Could you maybe just talk a little bit around where your products are in terms of cloud readiness and the willingness of customers to go there and your decision not to force a subscription transition and to allow customers still to have the choice why you feel that's a better decision when almost everyone else in the industry has forced the change?
Yeah. It's interesting because many of our customers approach us with the condition that they do not want to be forced to go to the cloud. For them, it's important that they keep control. I think data sovereignty is probably, you might have heard that Microsoft announced massive investments in Europe in local data center structures. Why do they do that? Because of the requirements of European customers for data sovereignty and also regulation. So don't underestimate the importance of that. So there will be local clouds, essentially, whether they are on-premise or localized in countries. That's the model that we are going towards. One of the advantages that we have is we operate also our own cloud infrastructure.
We've been an early investor into a company to build and optimize cloud infrastructure for Dassault Systèmes deployments that gives customers the flexibility to operate this end-to-end, but also we operate it for them. So we don't want to be distinctive to customers whether this is cloud versus on-premise. They have the choice to go either direction. In the data, there's a lot of value. So our preference is to go towards cloud, to use 3DS OUTSCALE as the infrastructure on which we run the 3DEXPERIENCE platform. But we also offer the choice for the customers to operate these clouds on their premises. And we price for value. Whenever we do that, it has a price point. So it's not margin dilutive if we take that approach. I think that's also important for investors to understand that the cloud strategy for us is not changing our margin profile.
Our platform today, as I said, you can run it on-premise and cloud. There's a parity in terms of functionality and value you have. So the roadmap over time of which the platform is adopted is really in the hands of our customers. This has been a key element for JLR in that negotiation. We wanted them to go to the cloud as much as possible, as fast as possible. Others will take more time to go there. I don't want to mention other names. But Renault, for example, is fully on our 3DS OUTSCALE cloud. So we have the opportunity to do both.
But for you, customer choice, full parity between the solutions, cloud on-premise. Then you let the customers decide how quickly they want to make that move. It doesn't affect your business model from a margin perspective.
Yep. And the predominant business model is, of course, the subscription model. And the subscription model gives us the opportunity to do much more value-up strategy. Whenever we see an opportunity to expand, we give the incentive for the customer to use as much as possible on the platform in the subscription fee that we pay, that they pay us. And after the subscription term reaches its end and we renew the subscription, we will measure the usage. And based on that, we will adjust the baseline of the prices on the usage. And as you give freedom of usage, you give incentive to use more. You automatically equip also new programs. And innovation has always fueled our model. And that's what the 3DEXPERIENCE platform enables us to do. When we lead with the 3DEXPERIENCE platform, it's not that we replace what exists. We are significantly expanding.
That's the key message, significantly expanding, whether it is on-premise or on the cloud.
Perfect. That's very helpful. And I've been trying to do this work myself. When we look at competition, there's a lot of the companies disclose different elements of the business. There's different mixes to you. Companies are going through cloud transitions. So it's actually very hard to get a like-for-like comparison of you versus peers. In that PLM space, could you maybe help us from your point of view? What are you seeing on win rate trends on these platform deals? Are you seeing more competition? Do you feel your platform has improved? And now that your win rates are improving, any help you could give us on that side?
I mean, all of those deals are competitive. In fact, it goes through multiple iterations. But at the end, we feel that how decisions are made is based on what engineers like to work with most. And this is where we are very, very strong. Engineers, they love our products. And they are advocates for the next generation of products to be adopted. Sometimes organizations and processes can be in the way. And so when there is a platform adoption, it also comes along with reorganizing and changing the way we work. Of course, there is competition. But our win rates are fairly stable. I think the recent wins in the OEM space, they document this. And they are not only in Europe. We've been successful at GM. We've been expanding our footprint at GM. We continue to gain traction at Honda, Toyota. These are all our customers already.
But they are expanding the conversations with us. Not that this is already everything reflected in our P&L. No, they are expanding conversations. They see the need to drive transformation, simplification, speed of execution, better connectivity to their value networks, new value networks, material science. All of those aspects are drivers of the adoption of our platform. And well, this is also where we differentiate with the competition. The competition offers parts of what we do. And when we used to go to market on a brand-by-brand basis, we were going competition by competition. But now it's a platform play. And there are different dynamics. There are end-to-end processes that we have to demonstrate in these assessments where we have a strong value proposition. And we've been competing very well.
So there's another part of our business, which is the consumer-centric industries where we say it's a new generation of PLM system for the fashion and consumer industries where you have fast-turning collections based on customer and consumer sentiments that can vary based on programs, on seasons. And today, we have the major and leading brands in the world on the Centric PLM platform and really expanding not only through industry diversification from apparels to fashion, consumer electronics, food, and retail, but also expanding our platform from collection management to planning and dynamic pricing. We, in fact, signed in 2023, starting to sign large enterprise deals in double-digit million EUR in this category. And also here, there's a lot of homegrown systems. But there's SAP. There's Oracle. And there's PTC in the market that we are displacing. And this market is large. It's a very large TAM.
It's a volume business. Precision and digits behind the comma matter. When you have an integrated system there, there's a lot of value to be created through that Centric PLM platform. I think you will hear more from that in the next year.
Well, this sounds, I mean, this has been more than a nice, a very good growth driver for you in 2023. I guess you expect that to continue. Is there any numbers you could put around that in terms of how big Centric is, how quickly it's growing, any kind of flavor or?
Centric is around EUR 200 million today. It's growing 30%-40% year-over-year. It's on a trajectory to further accelerate growth and to be at EUR 1 billion in the next years. We have hunted very successfully the market, winning a lot of brands. Now, we are expanding to large retailers who also have their own brands, like Decathlon. It's a very famous example. 80% of their revenue comes from sports brands that they have their own brands. Last quarter, we won Wilson, Fila. All iconic sports brands are now starting to be on our platform. We are not only going the industry diversification, but we also are expanding towards domains and really capture more share of wallet now within all of those brands that we already have as customers. We've been hunting.
Now, we are farming while we continue to hunt.
Perfect. That's a very nice story on Centric. You've talked a lot about platforms around PLM. Maybe let's shift that discussion to the Medidata business. And I think there, you've started with electronic data capture for clinical trials. But I guess the vision is similar in PLM where you start with CAD. But the vision is to have a much broader platform of what you can deliver to pharma, biotech, CROs. Could you just help us a little bit with what that vision is, where we are on the delivery?
Yeah, absolutely. That's exactly the right way to frame it, Adam, because we've been inventing the category of EDC. Now, others are catching up while we are inventing what's next. Today, we are by far the market leader. We have 8,500 concurrent clinical trials on our platform, over 30,000 clinical trials that we have done to date. But today, when you think about the challenge of the industry, the drugs are still too expensive to develop. It's very hard to enroll patients. Data quality is very hard to maintain to prove the evidence. Hospitals are understaffed with their ability to provide care to patients. So how do you really rethink the system right now? So the decentralization of clinical trials is a key element of that. And that's what we address through our Patient Cloud application.
So when we talk about our value proposition, where we see this market is going, the majority of the data in the future will be not captured at the site. It will be captured with patients directly. And that's where our Patient Cloud portfolio is market leading. And we have proven that to be the market leader during the COVID period. Today, the most complex clinical trials are run on our Patient Cloud platform. And the sponsors who have the widest adoption of Patient Cloud and patient direct data capture technologies are using us. So we're going from EDC, site-centric solutions to patient-centric solutions. And now, we need to augment the evidence generation with relevant data sets.
We have now completed more than 65 Synthetic Control Arm projects where already in 15 cases, these control arms have been used for regulatory purposes, be it for the submission or be it for the design of a clinical trial to be approved by the regulator. So we are using our data set to advance evidence generation. And the vision here really behind this is that we are going to package clinical trials with preconfigured data to be able to measure the standard of care as the evidence is generated so that, yes, you will still probably have a certain distribution between treatment and control arm. But you have supplementary data sets to help you to design the trial in a way that it is better the patient burden is less, that you reduce patient dropout rates.
The worst thing that can happen to a clinical trial is that patients are dropping out because they realize that they are in a control arm. They don't receive the actual treatment. They realize, "I'm wasting my time. I need to do something else. I'm not going to stay." Or it is so painful that they do not want to continue with that. They stop participating in this. That puts the clinical trial results at risk. In order to find better designs, you need to learn from the data. You need to leverage data that already exists to show the efficacy and prove the evidence. That's the direction that we are going. With this, I know we are, in a way, ahead of where our customers are.
Some biotechs are leveraging this technology today because they know for them, it's a way to get to the endpoint faster. Large pharma companies are still hesitating as they have been with many things in the past. We know that from our own experience from years ago. It's a very conservative and risk-averse industry. So I mean, I think that describes very well where we are. I mean, we are inventing what's next at this point in time while we are defending our space. And of course, we continue to invest into our EDC system and where we are market leader today.
I don't think any presentations happen without some discussion about Gen AI. Is that the area of Dassault's business today where you see the biggest opportunities around the data set that you have in Medidata? Maybe you could just help a little bit on that. And then are there areas else?
I wouldn't say that's the biggest opportunity. It's one of our opportunities. What we do with the large OEMs around Virtual Twins to connect the Virtual Twins to optimize the design of cars, to optimize cost, to optimize sustainability, there's massive value to be created to create virtual representations of the real world or whatever objects you want to imagine. Think about predictive maintenance on how you improve maintenance schedules, improve the designs to reduce cost. There's a lot of value in our core industries. Of course, we augment designers with Generative Design to improve designs. I think this is clear. The value capture is yet ahead of us on this part. For Medidata, we have been working on our data set for quite some time.
At the end, to get a drug approved by the regulators, it's based on the quality of the data and the efficacy that you can prove with your data set. So we have been. That's in the name of Medidata. It's about data and finding the algorithms to identify patterns and to identify the biomarkers to differentiate certain patient groups that have a positive response to a treatment versus those who do not. That's ultimately how you, what we've always been aspiring and going after together with our customers. It's not something that we just do on ourselves. We do this together with our sponsors. But you're right. I mean, that's a massive opportunity. But it is an opportunity for this industry to transform.
But there's no other place to go than to us because no one has that data set that we've aggregated over 30,000 clinical trials where we represent 75% of the drugs approved by the FDA in 2023, so very, very relevant data. And that cannot be replicated, for which we have secondary data use rights. Of course, all the data is anonymized and aggregated. But that practice has been built for the last 10, 15 years together with our customers. Why could we do that? Because customers trust us. It's a gift-to-get model. You share to receive the benefits. You don't receive the benefits. You're not willing to share as simple as that. And customers are willing to pay for it. So we've been thinking about this, as you can hear, very deeply for a long time.
I think with this awareness and understanding of the power of AI, I think we are getting closer, clearly closer to the point of commercialization of this type of approach.
Perfect. Looking at the clock, I just wanted to open it out to the floor to see if there are any questions from the audience. We've got any questions from the floor at all? Take one here, please.
Thank you. Just with five years on from the Medidata acquisition, what were the parts of the business that you think performed better than you expected or worse than expected, notwithstanding the fact you did give a very solid answer before around Medidata? I just thought it's a good chance to look back at what you acquired.
Sure. I was part of the Medidata acquisition. So when you say of what we acquired, I have the benefit of understanding the world before and after the acquisition. I think what I'm most excited about is that we were able to continue to keep that spirit of innovating what will be next. The transitioning from a site-centric to a patient-centric approach is what will be the future. And I think we have built the foundation to make that shift. And during COVID, that was clearly evidenced. And we will continue to invest into this direction. And it will lead us further outside the clinical trial space because Patient Cloud is applicable not only to prove the efficacy of new treatments but also the value of drugs that are already in the market through companion applications. That's the next step to take. And we are just in front of that.
So this is probably, for me, the part that gives me the most confidence about the future. The part that I would have loved to see accelerating much faster is our ability to monetize our data and to scale our Data and AI business. I think we have made good progress. We have seen the adoption of Synthetic Control Arms by the FDA and some biotechs. But we have not yet been able to get our large sponsors to adopt this technology. And that's something that we have to do.
Anything else from the floor? Could I maybe sneak one last one in? And I'm sorry I'm only giving you a minute. But we've obviously had a, it feels like we've had, a little bit of a boom and bust in the clinical trial market, COVID, the boom, and then a bust with difficult comps and less investment in the industry. Medidata slowed through 2023. And you're guiding to kind of flat-low single-digit growth of that business in 2024, a competitor's guiding to faster growth. Could you just help us? Is that market that's driving that guidance? Because investors are obviously nervous that there could be some share shifts happening.
Yeah. So our business has always been sensitive to the way we monetize our solution is on a trial basis. The number of trials our customers are running, on average, determines the price they pay. Now, of course, the number of products they use as well. So it's the intensity, which is the number of trials, and density, which is the level of which they adopt the platform and solutions that we offer and services we offer. So that part of market growth, which was fairly consistent since I've been observing the market since 2015, 3%-5% on average clinical trial growth, has resulted in strong double-digit growth for us for years. Now, we have seen a strong uptick during COVID as the number of clinical trials surged. And we captured approximately 80% of those COVID-related trials, which gave us a big boost in terms of revenue growth.
Now, we are on the other side of that swing where those trials are ending and less trials are starting. We see that in our revenue model where we simply have less trials starting than the ones that are ending are more than the ones that are starting in the future, right? In 2024, we see the normalization of this trend back to the 3%-5% growth of clinical trial starts in 2024. We have now the carryover effect from the decline in 2023 where we have less trials that we signed that are producing revenue in 2024 from 2023. As the number of trials are normalizing in 2024, we will be able to reverse that trend coming into 2025. In a way, that dependency on volume will be behind us.
The COVID bubble going up and going down will be consumed in 2025. Based on what I described before on our platform strategy and the value we capture, our plan is to be back on double-digit growth.
Perfect. That's very helpful. Again, Rouven, thank you very much for joining us.
Thank you, Adam.
Thank you.
Great to be here.
Thanks, everybody.