Hello, and welcome to the Dassault Systèmes Third Quarter 2025 Presentation. For the duration of the call, your lines will be on listen-only mode. However, you will have the opportunity to ask questions at the end of the presentation. This can be done by pressing the pound key, five on your telephone keypad. I will now hand over to your host, Stacy Pollard, to begin today's conference. Please madam, go ahead.
Hi, this is Stacy Pollard with Dassault Systèmes. Thank you for joining our Third Quarter 2025 Earnings Conference Call with Pascal Daloz, our Chief Executive Officer, and Rouven Bergmann, our Chief Financial Officer. Before we begin, some quick notes. Dassault Systèmes' results are prepared in accordance with IFRS. The financial figures discussed on this conference call are in a non-IFRS basis, with revenue growth rates on a constant currency basis unless otherwise noted. Some of the comments on this call contain forward-looking statements that could differ materially from actual results.
Please refer to today's press release and the risk factors section of our 2024 universal registration document. All earnings materials are available on our websites, and these prepared remarks will also be available shortly after this call. Now, d ue to technical delay, we've lost some time. D ue to the technical issues, Rouven is going to summarize more quickly than we'd originally planned, in order to save time to address your questions. Both Rouven and Pascal are here, and Rouven will take over.
Thank you, Stacy. We apologize again also for the technical delay, but we promise you we will make up for it, to make sure we have enough time for your questions. As Stacy was just saying, I'm going to give you a brief overview on the quarter, but also, I want to spend some time on the guidance, before we then open up for Q&A. Let me start first of all, and thank you again for joining us on today's webcasted call here. G ood morning to you in the United States. I want to start with three key messages. The first one is top-line growth and margin expansion are our top priority. The second message is the 3DEXPERIENCE platform is driving our business model shift to subscription and recurring revenue growth. This engine is working with 16% growth in subscription this quarter.
The third message is, we are mission-critical to our client. In fact, in 2025, we are winning significant contracts with many of the top industrial companies in the world. This is laying the foundation for long-term value creation with cloud and 3D Universe. It is these powerful long-term partnerships that give us confidence in our long-term targets. Now, a few more comments. Our financial results for the quarter were solid, with 5% revenue growth and an expanding operating margin up 100 basis points and 10% EPS year-over-year growth. Industrial innovation is driving the growth, which was up 9% in the quarter and 8% year-to-date, while Medidata and Centric were softer than expected. As discussed previously, the repositioning of Medidata is ongoing.
The change of the model to reduce the dependency on clinical trial activity will take time, as we are doubling down on the enterprise and PLM opportunity in life sciences. For Centric, we are accelerating the SaaS transition. To this effect, we also have promoted a new leadership team. Now, looking at the full year, we adjust our revenue outlook to 4%- 6% growth XFX, in line with our current trajectory of 5% top-line growth year- to- date. At the same time, we maintain our EPS growth target of 7%- 10% XFX growth, thanks to the strengthening of the operating margin, driven by efficiencies we are generating in the business. With this in mind, a few more details on the quarterly results. Q3 and year-to-date total revenue and software revenue are both up 5%.
Recurring revenue was strong, up 9% in the quarter, and it highlights a very solid acceleration when compared to 7% year- to- date for recurring. Subscription revenue growth was 16%, and it was driven by new deals signed in the quarter and the increasing visibility from large contracts that are ramping. As a result, subscription revenue now represents almost half of the recurring revenue base, which is up 3 points from last year. Starting in 2026, subscription will surpass maintenance revenue in absolute terms. 3DEXPERIENCE revenue was the growth engine behind that, which was up 16% in the third quarter. The signings at Ford and Apple contributed to the strength in subscription growth. Upfront license revenue declined 13%, as our clients continue to adopt the subscription model at an increasing rate.
The best proof of this is that recurring revenue now accounts for 84% of our software revenue year- to- date. The operating margin improved 100 basis points year- over- year for the quarter, driving strong EPS growth of 10%, thanks to productivity gains and cost discipline. In fact, the expense growth was up 3.1% in the quarter, and we continue to rebalance resources to support our growth strategy. I stop here on the overview, and I will now turn to the financial objectives. The main net is that our year-to-date revenue is up 5%, and for the full year, we now adjust our outlook to reflect this trajectory. We expect a growth of 4%- 6% XFX for both the total revenue and software revenue, with a 6%- 8% growth previously.
When you look at this adjustment in absolute terms for the full year, this reflects an adjustment of about $140 million to the midpoint. Of that, there's an impact from Q3 of $30 million and an FX impact of about $20 million. The remaining data can be explained by three factors. First, the lower growth for Medidata in line with Q3. I guess we'll discuss that. Medidata growth was -3%, and life sciences growth was -3% in the quarter. I just add that because we were not having the time to go through all the details. The second part is the impact of the SaaS acceleration of Centric. The third element is an increasing macro volatility, with the potential to impact the timing to close large transactions.
Here we are reflecting the potential of delay in decision-making on some of our large contracts, with the potential to impact our Q4 closing. Please also remember, we had a high comparison base in Q4 last year. Now, looking forward, the change of model for Medidata is ongoing and we are confident in the accelerated SaaS transition for Centric, given its strong positioning in a very large market and clients are endorsing it. For industrial innovation, we have built a strong foundation in 2025 with signing significant contracts. Also, we expect in 2026 to expand on these partnerships, to continue to transform our customers with virtual companions and generative experiences. To add to this, SolidWorks momentum was strong in Q3 and also year- to- date. Now, looking at the recurring revenue outlook, it remains stable at 7%- 8% and it's underscoring what I said at the beginning.
We're implementing a sustainable recurring growth model with increasing visibility. Above all, we have the strengthening of our operating model highlighted by margin improvement. As such, we are maintaining our EPS growth expectations for 7%- 10% XFX of €1.31- €1.35. To achieve this, we expect Q4 OpEx to continue the trend of Q3, delivering margin expansion of 100 basis points, driven by ongoing productivity initiatives. Now, briefly to Q4. As you can see, the revenue range of 1%- 8% is fairly large. This predicates the potential uncertainty in timing of large deal closing, mainly for the upfront license business where subscription growth of 8%- 12% is solid, also comparing to a high base last year.
Operating margin is expected in the range of 37.2%- 38%, and EPS growth of 17% XFX, to reach €0.41- €0.45 EPS for the quarter, reflecting the ongoing operating leverage. Now, as I reflect on our performance so far this year, I want to highlight that our operating model is resilient and we apply strict financial discipline to support our long-term growth. We are occupying a leading market position, which makes us mission-critical today and tomorrow for our clients. Profitable growth and improving cash conversion is a top priority, with a clear objective to show results, improving results starting in 2026. AI and cloud are our two main growth drivers, and we are confident that we will deliver on their ambitious growth targets. We are committed to invest right for innovation, customers, and shareholder value. With this, we are turning to the Q&A.
Thank you. As a reminder, if you wish to ask a question, please press pound key, five on your telephone keypad. Our first question comes from the line of Jay Vleeschhouwer at Griffin. Your line is open. Please sir, go ahead.
Thank you. Good afternoon, Pascal and Rouven. A few things, of course, as always. Rouven, you mentioned in passing some productivity initiatives. Could you elaborate on that? Are there any operational adjustments that you are making with respect to any or all of go-to-market, perhaps affecting CSE or CPE product packaging? In other words, anything having to do with the portfolio, perhaps even any levels of management. Anything in detail of that kind that you can talk about in response to some of the growth inhibitions you spoke about would be very helpful.
Okay. Thank you, Jay. I can start. Thank you for the question. The productivity initiatives are ongoing for quite some time. We are focused on putting the right people at the right place to support our growth initiatives. Now, with AI and 3D Universes, it's huge. We are creating new categories of applications for our clients. This requires making sure we have the right skills, as well as the right people to support these ambitious roadmaps in R&D, for example. This is very well managed by our R&D organization. As it relates to the go-to-market, there are several points to take. I think you heard us talk about the indirect business, which is a key focus on rejuvenating and recalibrating our partners. I think you see already good evidence of this in our SolidWorks numbers for the role partners. There's good momentum that we are creating.
We are working actively on the process partners, to have the changes that we are planning to also drive impact, which is not yet so visible, but we are working on that, on the process partners. As it relates, for example, to our life sciences business, here the objective is to evolve our go-to-market and offering to be enterprise, to embrace the enterprise, to be enterprise-wide, to expand from the focus of clinical development across the whole value chain of life sciences and the entire patient journey. Therefore, we are recalibrating our go-to-market and our offering. This is in process, and this is what we are expecting. We are seeing results right now already, but of course, this is to be expected to turn in 2026 also into growth for life sciences overall, which we don't see yet. We are also looking at our service business.
S ervice used to be a time and material business. Today, it becomes an outcome-based business. W e talk about Virtual Twin as a Service, as an offering which is really delivering outcomes to our clients. It requires a repackaging of our software and services to be combined to deliver real outcomes for clients. This is in full execution, this transformation. This is to mention a few on the business sides. You heard us talk about Centric as well.
For Centric, we are focusing on really transitioning this business to become SaaS and cloud- first. This decision has been made. Now we're executing along those lines. From a cost structural standpoint, really here, the focus has been for us to capitalize on past and existing investments, and to ensure that we can grow revenue faster than expenses, which then automatically translates into EBIT expansion. You see that in Q3, and we are committed to do that in Q4 and going forward.
Okay.
That's to be [crosstalk].
Thank you. You mentioned some large signings, which I assume you were alluding there to industrial innovation large signings. The question there is, are you seeing the customary programmatic adoption that D S has seen for decades in aero and auto behind those signings, or something perhaps more structural or transformational in terms of the adoption that those signings are based on? In other words, are you seeing a mix in terms of why customers are doing the signings? Is it just a conventional new aircraft program from Boeing or whomever, versus something broader and on a more ongoing basis that would not only affect CSE, but eventually move out into CPE tier two and tier three?
Jay, Pascal speaking, it's a very good question. In fact, you're right. For a long time, the deployment of our solution was extremely linked to the programs, whether it's an aircraft or an auto [crosstalk].
[crosstalk]
With AI, I think we are seeing a different trend, which is much more, how we could accelerate the transformation, leveraging the platform which is ready for AI across the entire chain. Just including the large OEM, you are right, but also the suppliers. This morning, I was mentioning the Ford case. As you may know, we signed a five-year contract with Ford, to deploy the platform across all the programs. It's not only the EV, but also the traditional one. This is also including the supply chain.
It's exactly your point, which is, this has a drag effect automatically on the supply because at the end, this AI story is something you cannot segregate, if you want, between different companies. You have to look at the extended enterprise if you want to unlock the value. This is opening this. This is definitively accelerating the trend for the adoptions.
Okay. A couple last things, if I may. Could you comment, Pascal then on the rationale behind the new renaming within the SolidWorks business? In other words, you're returning to your roots of SolidWorks branding. Maybe talk about why you did that, what go-to-market effects that might entail. More broadly, the company has been adhering to a solution- selling concept for years, but perhaps, do you reconsider that or evolve that in some way as you've done just now with SolidWorks?
I'll take it with what is the most proven, but feel free to add whatever you want. SolidWorks is an umbrella for the mainstream market, right? That's the reason why we are creating the Works family, whether it's DELIA Works or [Solid] Works. At the end, SolidWorks is really the umbrella because the large community we are serving are the SolidWorks users. Now, we do not want to be a solution or industry-oriented with SolidWorks. We have to be industry-aware, which is very different. When you are industry-oriented, basically you package your solutions to fit precisely for market segments. This is what we do for most of our product line. We do not intend to do this for SolidWorks. SolidWorks, nevertheless, has to be industry-aware, which means there are certain features which are extremely relevant for one industry.
There are certain model foundations with artificial intelligence, which will be extremely relevant for certain industries. We want SolidWorks to take the benefit of this. This is the way to approach it. This is a way to package it. Last but not least, and I know, Jay, you are a SolidWorks lover for a long time, and m y motto is, "Red is red," meaning we need to keep the ease of use of SolidWorks across the board, because this has been for a long time, a key differentiation for SolidWorks.
AI in a way is giving another way to ease even further, the usage of the solutions because this is basically managing the complexity of sometimes very advanced workflow, like what you can do in simulations. Those are what we are currently doing with SolidWorks. By the way, you should celebrate the anniversary of SolidWorks because this year is the 30-year anniversary of SolidWorks. I think over the last 30 years, we have built an amazing community.
Okay. 3D U niverses are still very early obviously, but you did give a very specific outlook for it at the meeting in Paris in June. Between now and then, that is to say 2029, what kind of additional or new metrics do you think you will be providing us, to help us understand the progression towards that very large goal that you set for several years from now? Are you going to give specific revenue milestones, or any other kinds of metrics like that to substantiate your ambition for 3D U niverses?
There are two ways to answer to your questions, Jay. One, Rouven, during the Capital Market Day, made basically an explicit commitment and said over the time, over the period of the plan, half a billion will come from there, which is the combination of the AI and also the acceleration of the cloud, right? At the end, in a way, we will give you, maybe not on a quarterly basis, but on a yearly basis, where we are. The second thing is probably more important, which is the attach rate. Remember, the way we have structured the portfolio is the following, w e have roles, processes, and solutions, and we are with AI, creating a new category of solutions. We do not want to replace what we do. We want to complement and extend. In front of the role, you have the virtual companions. This quarter, we released three.
That was part of my presentation this morning, price. There are concrete examples, and they are available on the market. Basically, we expect to have some revenue coming in 2026 from there. In front of the processes, we have the generative experiences. Right now, we are focusing on the use cases, which are really the ones creating a real breakthrough. We are not looking for the small improvements in the processes if you want. We really want to create a moonshot. It is at least 5x , if not 10x improvements of the way to work currently. For this, it will be a token-based approach, a usage-based approach, which will facilitate the adoptions. In front of the solutions, you have what we call the Virtual Twin as a Service. It' s really, rather than to sell the tools, we are selling the output of what we are creating with it.
It is a combination between the technology we have with the services capabilities we have, either from us or from the ecosystem, in order to better serve our partners. If you mix all those three, we have new categories, new business models, complementing what we do, leveraging what we do. I think we have a clear roadmap. On a yearly basis, clearly, we can give you some data points on the projections of the revenue stream. We have a concrete roadmap for 2026 and 2027. Almost every quarter, we will come back with the new capabilities, the new category of solutions we are introducing on the market, just for you to understand the attach rate between the roles, between the process, and between the solutions with all the new categories I' ve just mentioned.
Thank you. Our next question comes from the line of Nicolas David at ODDO. Your line is open. Please, sir, go ahead.
Yes. Good afternoon, Pascal. Good afternoon, Rouven. Thank you for taking my question. I have two actually. The first , coming back on Centric, I would like to understand what really triggered this decision to move or to accelerate the move to SaaS, a s it was pretty clear from the beginning of the year that you would be facing super tough comps in H2. At the beginning of the year, if I'm not wrong, you seemed to be betting again on strong upfront revenue coming from renewals. Even at the time of Q2 earnings, it seems that you were still believing that you would have some upfront revenue. It seems that this decision was made pretty recently. This decision is linked to the fact that you acquired the remaining shares of the company, which freed you from any constraint to do that or something like that.
My second question is regarding the acceleration of the move to subscription, and beyond Centric this time. Is this time a move on which we have greater control, or does it remain in the hands of customers mainly, as they still have a choice? The fact that you still don't have total control on that, may explain the lower than expected license in Q3 and the very broad guidance rush in Q4. At a certain point of time, do you intend to get a better control of the phasing of the move to subscription, or do you still think that letting the client choose is still the best solution? Thank you.
Thank you, David, for your question. I will start with [crosstalk] add whatever you want. Centric, there are two reasons behind this decision to accelerate the transitions or the acceleration to the SaaS business model. One is the one you just said, which is we acquire the remaining piece. Obviously, you know when valuation is at stake, I think it's probably better to do it just after because we know that this could have an impact on the revenue stream, and it's right now in our full control. This is basically much more the technical reason, let's say. The fundamental reason is much more simple, of this. I had a chance to meet with many of Centric customers, and many of them are requesting to move to SaaS from a product standpoint, but also from a business model standpoint.
The reason is because the market they serve is still doing a lot of M&A. It's really an easier way to connect the systems together. It's much more fast in terms of adoption. Many of the Centric customers are also SAP customers, and many of them are moving to SAP Alert currently. They want to do this move in one single shot, if you want to have the complete SaaS architecture being the foundation for their information systems. Those are the two fundamental reasons why we are accelerating for Centric. Coming back to the subscription, there are two ways to answer your questions. One is, for sure, that we will continue to offer the upfront license model alongside the subscriptions? There are fundamental reasons for this. One is because there are certain industries which are CapEx-based, and we need to respect this.
The upfront system is much more convenient for them. There are certain industries like the defense, t hey do not want you to operate the software for them. It does not mean that at the end, the license will not be cloudified if you want, but it will definitely not be the one operating the system for them. You have certain countries like China where the subscription and the SaaS will not be widely adopted for a certain fraction of the market, for obvious reasons. If you do the sum of this, you know the upfront license is a little bit less than $1 billion over the year. For one year, it will progressively reduce to $0.5 billion, but I do not expect this will be below this. Now, why I'm saying this, is because for all the others, obviously we are pushing the subscriptions, especially for the large contracts.
There is one simple reason for this, m ost of the large contracts have a cloud component right now because usually, it's a minimum of five years commitment. The reality is, the system we are developing is extremely sticky. You do not make this investment only for five years. Usually, you do it for a minimum of 10 years. Clearly, for them, they need to integrate the cloud architecture in what they do. The cloud business model is coming with it, and t hat's an easy way for us to push along this way. After, you have the indirect channel. This is maybe where it's a little bit tricky. Let me explain to you why. For the partners, there is a big difference from a cash flow standpoint between the upfront and the subscriptions. The danger is what?
It would have been easy for me to put a lot of incentive for them to move to subscriptions. If I do that without having time control, the risk I'm taking is for them to repurpose the large support and maintenance revenue stream we have thanks to the large installed base into subscriptions, without expanding the scope of what we do. The entire purpose of the SaaS business model for us is to expand what we do. That's the reason why I do it basically sequentially, and we do it with some specific offer. Now, if we step back, I think if I link the previous conversation I had with Jay, about how the platform is used now in the cloud and how the supply chain is also endorsing the platform in order to unlock the AI capabilities, it goes with it.
For SolidWorks, I think the subscription is right now exceeding 35% of the total of the mix. We are gaining in the mix almost 5 points every year. We see this accelerating. Let's say, I guess ahead of 2026, probably we will be at 50% already. The net of this, and this is what Rouven was saying, at the end, if you look at the recurring revenue, the subscription is already being at almost half of it, if not surpassing that, right? It's more than 50, but i t's really more than 50 in 2026.
This is a triggering event because, as you could imagine, now the growth will be extremely visible in the recurrent part. When it was only 1/3 , obviously it was diluted by the rest. This is how we have let's say, a better control in a way, how we are pushing the subscription over the time, David. Long answer, but I think you have the framework of all the levers we are using in order to make it happen.
Yes, thank you for the very comprehensive answer. That's clear.
Thank you, [crosstalk] .
Our next question comes from the line of [Jason Celino] at KeyBank. Your line is open. Please go ahead.
Great. Thanks, Rouven. I think one of the components of your Q4 guide is leaving room for the possibility for longer sales cycles. I wanted to ask about sales cycles that you saw in Q3, particularly on the industrial innovation side. Did you see or are you seeing longer sales cycles, or is this just a component for Q4 to leave some cushion?
Yeah, thank you, Jason, for the question. Yes, the answer is yes. Yeah, we are seeing that. We always said coming into H2, we have a pipeline that is more weighted H2 versus H1. We, in fact, were confident to be online to meet our annual guidance. Really, besides what we discussed on lower than expected Medidata and also the situation on Centric, which was discussed, there's also an impact of decision cycles on large contracts that we need to take into account. I think we had a good quarter in Q3. with 9% growth in industrial innovation.
We also had of course more potential of deals that have moved into Q4 or are still in our pipeline to be closed, I would say. Now, that's a reality. I think what's important for you to know, is none of those deals we are losing to competition. We are building those and closing them at the right time with the right economics, and in the right model, as Pascal said, because we want those contracts to be not one-time. We want those contracts to be recurring, like a springboard for value creation and transformation. This is why we are taking a very focused and specific approach on closing those transactions.
Okay. Thank you for the clarification. On Medidata, it sounds like you're expecting similar trends in Q4, and your enterprise motion is not able to compensate for the declining volumes. What gives you confidence that clinical trial volumes in Q4 won't decline further?
I think the trend right now that we are seeing is the declining trend. It's fairly, I think a clear picture that we see in 2025. In our 2025 number, it's an outlook of being in line with Q3 for Q4. This is baked in. The real question will be, what's going to be in 2026, t he starting point? As it relates to the clinical trial volume, I don't perceive for Q4, it's similar to Q3. I think we have that level of visibility at this point.
What was for us more difficult to assess after the first six months is, are we able to see that swing back to growth? Is volume stabilizing? It was under pressure in the first six months, but still i f there is a balancing effect throughout the year, it definitely would have stabilized our ability to deliver the full year. Looking into the first months of Q3 gives us a picture for Q4, because this is the ongoing assumption. That's why we are assuming Q4 in line with Q3 trend for the volume part. May I ask [crosstalk]?
Yes. Jason, one thing, which is, I want just to be sure we are clear on this. In fact, we are offsetting the decrease of Medidata with the PLM activities, except it's not reported in the line of life sciences. It's in the industrial innovation, b ecause if you do the math, -3 % for Medidata, believe me, we do much more than $10 million with the PLM on the rest of what we do.
Overall, we are growing in the life sciences. I will not want you to jump to the wrong conclusions. Nevertheless, to echo what you say, Rouven, we have not yet balanced the effects of the termination of the studies and not being able to compensate with the new staff. The volume is still at stake, but the enterprise-wide is growing more than double digits, and it's offsetting largely this, except it's not reported in the same line.
Okay, [crosstalk].
When you do your comparison with our peers, it's an important thing you should take into account.
Thank you. Our next question comes from the line of Michael Briest at UBS. Your line is open. Please sir, go ahead.
Thanks. Good afternoon. A couple from me as well. Just firstly, it hasn't come up in a while, but what does the M&A pipeline look like? Given some of the operational challenges and organizational changes you're making, do you have the willingness to take on large-scale M&A at the moment? Also, if M&A is not really on the agenda, what about buybacks to capitalize on the weak share price?
[crosstalk]. I think we have a pipeline for M&A. Again, the management channels you are talking about are extremely, by the way, well prepared. If you are referring to Centric , [Fablice] is with the company for the last 20 years. He w as the CEO and the President for more than five years. Clearly, the team is prepared for this. The topic is much more for us. It has to be relevant for what we want to do. All the acquisitions we are considering right now should be projected in the concept of, in the framework of the 3D Universes we are building. You need to have a cloud and an AI component in it, to justify the fact that we will basically invest a significant check. Otherwise, we will continue to do some natural extension of what we do.
The recent acquisition of Contentserv, for example, is a good example of how we could expand Centric . We have these also pipelines across many domains we do. No, it's still on the agenda, Michael. As usual, to do the big move from a transformation standpoint, it has to fit with the strategy of Dassault Systèmes and where we want to go. That's always been the guideline, and we will continue to do along this way.
Okay. Just on outscaling cloud, I think you have a proposition in France for sovereign cloud. There is a lot of sort of interest in that topic currently. There's AI gigafactories being developed with EU support. Can you talk about your ambitions in cloud, and maybe whether you would look to use the hyperscalers more or less and the CapEx consequences of that?
Okay. 3DS upscale is two things for us. One is to address the sovereign cloud, which is a market. Now it's becoming a real market, given what is happening around the world. By the way, it's not only true for France. It's true for many countries around the world. The fact that we comply with some specific French standard, does not mean we are not complying with others. Just keep this in mind. It's not one or the other, w e should do both. We should continue to develop our scale accordingly, to ramp up with the hyperscaler. Since day one, we had this strategy. I mean, AWS is the other part of what we do. We are relying on them on many things we do. The fundamental reason is very simple.
When we are addressing either industry or geo where we do not have a critical mass, it's much better to start with AWS because for the one who knows, the business model of a data center is very simple. The day you open it, it has to be fully loaded. If it's not fully loaded, you will never have your payback. The way to do it with the hyperscaler is, we are using the hyperscaler to create, if you want, the footprint and the base.
When the base is enough, we open the data center in order to have the full benefit of it and keep the margin for us. That's the strategy. This is very simple. The more we expand with the cloud in a way, the more we're going to use also indirectly, the hyperscaler in conjunction with 3DS upscale. From a CapEx standpoint, I think we did a good job until now to keep it in the norm. Just because we have this flexibility, we will not be catched by this if it is basically directly your question.
Okay, that's helpful.
We'll take our last question from the line of [crosstalk ]. Your line is open. Please sir, go ahead.
Thank you. Good afternoon, Rouven. Good afternoon, Pascal. Two questions, if I may. The first one is on the guidance for new license or upfront license revenue, so r evised onwards today. Can you help us to bridge previous guidance versus today's guidance? Would a decline be the norm for the coming years instead of the kind of stability you were presenting before?
I'm also referring here to Pascal's comment about dividing by two this €1 billion revenue line. Is it for the next three, four years, or was it a comment with a more long-term horizon? The second question is on the large deal you signed in Q3 for the [crosstalk]. I'm just trying to understand if it's a multi-hybrid contract, as we saw with other car makers. If it's the case, why haven't we seen the benefit of that in the upfront license revenue line, the famous 15%- 20% revenue booked upfront with this kind of contract? Thank you.
Thank you, Derek. Let me clarify the first question on the upfront license. [crosstalk] all the three questions you're asking are very connected, right? Of course, and so I will try to connect the dots for you here to clarify this. Again, when you look at year-to-date and Q3, the growth is in recurring and subscription. It's a subscription of 16%. You reflect on what Pascal said before when we talked about the topic of CapEx versus OpEx, and whether we give customers the choice. The reality is, the large contracts we' re signing all have cloud. With that, we have a model that allows us to create a more rentable revenue line than what the traditional multi-year subscription with upfront license was before. Now you consider that, that all of the flexible hybrids that are more rentable because of the cloud component are reflected in our subscription revenue.
That's of course i s a catalyst for the subscription revenue growth. As more and more of the large contracts are in the subscription line in a recurring way, year-to-year recurring, of course it has an impact on the upfront license part, which is less. You see that. The difficulty in terms of the guidance is, it's very hard to assess coming into a quarter, coming into a year, how much of this is going to be recurring versus upfront. This is why we want to give you a narrower range on the subscription part, because there is a strong contribution from deals that we have signed and that are ramping and that are adding to the subscription growth.
By definition, the upfront license is always at a point in time, and there can be variability. That's what you see reflected in the Q4 guide, with a larger range on the upfront license part. That's simply what's behind this. I think I've answered also the third question, because for the large deals that we are signing, that are cloud and subscription-based, you see them reflected in subscription. That's the model going forward. Does it answer your question, Derek?
Yeah. For the, let's say, medium-term view, long-term view, the license revenue decline that we should expect, is the flattish aspiration is still valid or is it today no longer the case?
For upfront license?
Yes, just for license [crosstalk] .
No, it's not going to stay flat. We just said it's going to approach half a billion over time. It cannot be flat if it's $1 billion today. It will decline over time, but t he growth will be reflected over proportionally in the subscription part on a recurring basis. With Centric, that's another element of it. It's going to add to this model. As you rightfully said before, multi-year contracts that we signed with Centric, the real revenue growth was in the upfront license part, which is, as you're changing the model, will become recurring. That's what we are focused on, and that's our priority.
Very clear. Thank you.
Thank you, Derek.
This was our last question. I'll hand the conference back to the speakers for any closing comments.
I want to take the time to thank all of you for the time you spent with us, and the rightfully key questions you raised. More than ever, you know I'm convinced we are doing the right things. If you look at the market conquest and the dynamic with the large customers we are winning over the last 12 months, we are building the foundation for the growth for the coming years. The fact that we are willing to accelerate the transition to the cloud and to the subscription, I think it's the right things to do. You should see it positively rather than to challenge us on this way, because I think at the end, we are doing the right things.
Now, Rouven and I will be on the road in the coming weeks, and I hope we'll have a chance to see some of you, if not all of you, in the coming days. I hope we will follow up these discussions. See you no later than early next year. For the one that Rouven and I will have a chance to see in the coming weeks, o kay, i t's only a [crosstalk]. Thank you so much. Bye-bye. Thank you.