Good afternoon for the ones on the webcast. I'm François Bordonado, Dassault Systèmes from the investor relations team. From the company, we have for this first Life Sciences Day in New York, Bernard Charlès, our Vice Chairman and CEO. Pascal Daloz, Dassault Systèmes CFO and Chief Strategy Officer. Tarek Sherif and Glen de Vries, Medidata's Founders and Co-CEOs. Rouven Bergmann, Medidata's Chief Operating Officer. Claire Biot, VP, Life Sciences, and Jason Benedict, BIOVIA VP, R&D. I would like to welcome you to Dassault Systèmes Life Sciences Day. At the end of the presentation, as you see in the agenda on the screen, we'll take question from the audience. Please note some of the comments we'll make during today's presentation will contain forward-looking statements which could differ materially from actual results. Please refer to our risk factors in our 2018 document reference.
Let me now introduce Bernard Charlès, our CEO and Vice Chairman.
Thank you, François- José, good morning to everyone. We are very delighted to have you here for this morning with a rich program. I have to tell you that, over the last two days, I've been learning much more about life science and what our customers and partners are doing with Medidata. It's a good training class for me. We want to reveal to you a little bit more as committed about what we want to do in the world of life science, because I think this industry need to change, to say the least, in my mind. We are excited, I am very excited with that perspective. We have not done a light study about that sector. We started 14 years ago.
We started really in 2009 with secret projects, and we'll talk to you a little bit more about those in a moment, really to understand and socialize with the sector of both life science and healthcare. The first thing I want to tell you is at Dassault Systèmes, we make bets, and we formulate them in a very clear way, and we did that for the last five years. I'm mentioning that more for the newcomer who don't necessarily knows us well. The first bet we did is to say we're going to unflatten the world by moving drawing system to 3D design.
Today, we have built the world standard in across many industries, still a lot to do, but there is no question anymore about the fact that this is a must, especially in what we call the fab s phere, the industry at large. 10 years later, we did another bet, which is represented here. It's working. Which was to say, "Well, can we use this to do the digital mock-up?" We call it the digital mock-up of an entire highly complex product called an airplane. This was done with Boeing on the 777. You have to know it's a EUR 15 billion program, 40 countries, seven years. I'm not impressed with the numbers in Life Sciences at all because they talk about EUR 2 billion, EUR 3 billion. When I listen, they tell me it's very, very high number.
I said, "Yeah, it is in some way." After they tell me it's very complex, I say, "Yeah, in some way." They say, "Well, by the way, we have a lot of people involved." I say, "Yes, I agree, but I think things can change." There is a before and an after 1989 because at that point in time, the entire world industry understood that the digital world could replace physical prototyping in a big way. It has created an incredible wave of transformation of the world industry. It's established. There are proof points on it. The automakers started to do it slowly.
The nature of what we do, and I want to mention that now, since the beginning, is one, provide science-based infrastructure to manage complexity, enable to do highly complex collaborative process management of multi-discipline knowledge and know-how, and use that digital world to lifecycle things, to add the time to it. That was 1989. We made another bet with another company called Toyota, and we said, "We want to do digitalization of your entire production system." You notice that the Q3, we mentioned that they are going to the next generation. That was not a minor news. It was small announcement but big news.
We demonstrated to the world that we could do the digital twin, the Virtual Twin experience of entire global complex production systems. On February 9th, 2012, we published a two-page paper where we said the new equity of the company will address three s pheres. The fab sphere, the bio sphere, and the geo sphere. We put in those spheres a way to look at the world in a way where we said the innovation should be at the center of this understanding about the bio world, the fab world and the geo world, material science. That was another bet, I think we are walking the talk with all the moves we have done since then. This was February 9th, 2012. We said, well, focusing on product is not enough. We should put the things upside down.
We should create universe where we can formulate experiences, the value of what is used in the economy as opposed to what is sold in the economy. That's why we call the platform the 3DEXPERIENCE platform. Well, you can understand now with the little symbol we have that our goal is to do the virtual twin of an entire human body. That's what we're going to do. We believe we have the reasons, good reason and proof points to make it happen. We have started with organs . We are working on human cells. What you're going to see this morning is related to all the pieces coming together to make this possible.
If I was starting from that point, you would say, "Well, what's the track record from the past?" I believe that those kind of bets are very clear, very precise, and I want to go through them quickly. Before I go in the life science by itself, I just want to remind you we're there in June for the , 2018, that when we published, when Pascal presented the growth plan for the next five years, this was almost without the life science world. There was a little bit of it, Pascal will come back on it. This is not new. This is what you have seen in June. We think we can continue double-digit growth in, if I shortcut it, the fab sphere, the world of the make, design, simulation, creation, production of products and solutions.
That's a baseline for us, and what we're going to talk about today is really what is our ambition, midterm or long term, for the Life Sciences, basically a new core for Dassault Systèmes. Briefly said, we are in the Yellow Sphere with many industries. We call this the Fab Sphere. We have initiated activities with BIOVIA and of course, Medidata. We position that set of solution in the Bio Sphere on healthcare. The Geo Sphere is really related to territory, cities, how you build the world for citizens. Including energy is now being put in that world. The numbers here are roughly the GDP numbers in the world economy. Pascal will come on the trends from that standpoint. What I say is that the new core is not small.
The second thing it says is that the new core, from my observation in the last 10 years meeting customers, is far behind in understanding how the digital world can help them. These enterprises are document-based, grade PDFs on digital documents. They are prisoners of office documents, they don't see the capacity of modeling, simulation, and data science yet. Statistics a bit, but not much more. Eleven industries, 61 segments. You have seven of them on the fab side. On the bio side, you have one and a half, and then you have the other three in the world of what we call geo sphere. For those of you who are following us, you are very familiar with that. What I want to mention here is that we have a very precise map of solutions, what we call Industry Solution Experiences.
Industry solutions is measured by the outcome to the companies. Developing a car from 70 months to 15 months or 12 months, that's an outcome. This is a real number that has happened in the last 15, 20 years. Same for aerospace, marine and offshore, industrial equipment, et cetera. Outcome-based, we have three measures. Outcome-based, performance for teams to do collaboration. We call them process experience. Making the users champion in what they do on defining their future job, and we call this the workforce of the future with roles. We basically deliver solutions which are based on 3 things: roles, process experience, industry experience. If you want to develop an airplane today, you don't need to go anywhere else but contact Dassault Systèmes. Nine out of 10 planes in the world are done with Dassault Systèmes software. Eight out of 10 cars, I think we are expanding that scope.
I'm going to do analogy here. We have moved from functionalities to outcome, process, or role-based KPIs. That's for what we have done up to now. Now I'm moving to the new core, which is life science. I'm trying to do some analogy. First, I think it starts from the very strong elements of our equity system, the virtual world, extend and improve the real world, and I will show you why. That suite was published on February ninth, 2012, now we understand that many of the moves we have done are associated to the journey and the growth plan and the value map on those three spheres.
When I am asked a question about where is Dassault Systèmes going to invest, what we do, the framework has been established in 2012, we usually follow it with high precision. Under our purpose, because we are a purpose-driven company, harmonized product, nature, and life with the virtual universe, I think echoes a lot of what I've heard yesterday with the great customers that you have Tarek and Glen built excellent relationship with over the last 20 years since you created your company and then really improve. We have a common culture. You have developed your group. I have done the same with my team, we understand what it is to develop a company from a few people to what it is today.
I think that was very probably the first criteria for us to decide to come together, I'm sure you will talk about it in a moment. 2009, we start a biointelligence project. We move and do the Accelrys acquisition, which is an extremely powerful platform for material science and bioscience. By the way, I present to you something constituted of 28 elements myself. You are 28 elements of the Mendeleev table. Not 29, not 26, only 28 elements of the Mendeleev table makes a human and it's 58 elements here. 58 here, 28 here. Think about it. People don't think about it that way, that's an interesting perspective. Very interesting.
I mentioned Mendeleev because it's the 150 years anniversary of the Mendeleev table this year, as it is the 500 years of the deaths of Leonardo da Vinci, my good friend. I am a pupil of Leonardo da Vinci. Accelrys, bioscience, material science, understanding from the molecule how you create new things, whether it's living tissues or new material science or additive layer manufacturing and others. This is a serious platform which is already integrated. It was a core move for us. In 2018, we revealed the Living Heart program and many others that the team will talk to you about, now we are in 2019 with Medidata. What is the play here? We have built a platform which is experience-based.
The beauty of experience is when you see an experience, you understand things. You can capitalize on knowledge and know-how. As Einstein said, "The ultimate of knowledge is experience." I will add the ultimate of know-how is 3DEXPERIENCE. In any world, a virtual experience help people to understand the phenomenon. More importantly, if it's presented to a group of people with multiple disciplines, they discover it in a different way. The platform as a business model, you have heard about the marketplace, you have heard about how we want to do the Amazon of production for manufacturing, and it's moving toward that direction. One click away, you can send the digital design and get the physical part. Those are things that which are happening. How do we do that?
We have invested massively in science, what we call multidiscipline science, and that was for the fab world, the manufacturing world. We are doing the same now for human. You will see concrete example this morning about how is this being applied. Living Heart being one of them, and as you know, FDA loves the Living Heart program because they think it's going to change the world of surgery, especially vascular surgery. You will see a few things here. It's about dermatology, neurology, cardiology. You see here the wide spectrum, and we use some of those science-based modeling and simulation to serve those things. The interesting news, it works. We have proof points that it works. That's what we want to reveal.
If I look on why we connected with Medidata, is the way I look at the life science world on the healthcare system today is they are far too expensive as compared to what is the outcome. Those companies have been behaving like very rich companies. It has happened before in other sectors of the industry, and we have seen how it could change quickly. The way I would summarize it is the world has been based on small molecules, chemistry in the pharma. Relatively simple complexity in the development, gigantic complexity when it's applied to wide population. Things are coming upside down, and this is the blockbuster story on all this. You need to make money with 1 because you lose so much money with so many that fails. That's the model today.
That's the way I see it, and I don't think I'm so negative. If we were doing airplane this way, we would do 100 airplane, put them in the air and say, "This one is flying. I'm going to produce this one." All the other one have dropped. Of course, it doesn't work this way anymore. The point is, this is shifting with biotech, especially on the coupling of equipment on bio-biologics. It's shifting to higher complexity in the development for smaller targets of population, targeted population, to an extreme of individual personalized. Those companies are not used to manage the complexity upfront. They are used to manage the complexity or evaluate the result at the end, not at the beginning. It's upside down, and the process of development, research, manufacturing will change in a big way. That's my bet.
I'm taking it. If you are here, you have to decide if you take it or not, we are going to make it happen. If I look at it's even worse when I look at the stupid patent process. Of course, when you have a simple things, you want to protect it because it's so simple. That's the reality. If you do very complex things, you don't even need to go to patent them because in fact, I recommend you don't patent them because people will not be able to replicate them anyway. Things are changing from that standpoint. Then the patent thing, you know, 20 years, if you take 10 years to develop it and do the clinical trial, you have only 10 years of life cycle. It's really a race.
The numbers, those are typical numbers for us. We know how to work those companies manipulating those kind of numbers, and I believe we can help them go and look at their world in a different way. Here's four things that I will briefly, before I conclude, tell you about. There are four focus in what we do. On the north side is multidiscipline collaboration, making sure that all specialists of different disciplines who don't talk to each other, don't understand each other, can finally understand each other with high clarity. You will see illustration with Claire and the team and Jason on that topic. That's the collaborative process. I visited many companies, very large companies in the life sector, life science sector. I am astonished with the poor environment they have for multidiscipline collaboration. It's document-based.
They don't find the documents they need when they need them. They are difficult to read. Nobody reads them. It's very artisanal. On the west side, of course, you have the representation of things and phenomenon. When we started to say that we would be doing a digital airplane, every one of the top specialists said we will not succeed because it's far too complex. Aerodynamics is too complex. The problem is not to be perfect. The problem is to reach a level where a group of people can imagine new solutions just because they see in front of them something which is not far, not too far from the real phenomenon, and that's the power of collaborative process, so representation of the phenomenon. Big data analytics. We are a big data company for 35 years.
I don't know if you have an idea about the volume of data generated by our clients. They are just gigantic volumes of data, including life cycle of the data they do for the products they do. We know how to manage very gigantic, large data structure, highly complex and heterogeneous. It's not with documents. It's through representation, what we call modeling and simulation. On the south side, you have the confrontation between what you think is the representation of the phenomenon with the real phenomenon as it is observed. Welcome to Medidata. There is an arm we don't have, which is all the real data from clinical trial, on confronting this to the new representation of the world for life science is going to be exciting.
This is a video that represents all the process briefly as I'm going to conclude with timing just being before I give the floor to Tarek. I think I just need to push maybe once more. Here is a real situation. Cooperation we have on the Living Brain project. This is the brain. You see a cooperative environment, collaborative environment going on with all the specialists from the different disciplines. We represent the brain at the level we can represent it. The professors are discussing about this. They are preparing the process by which they are going to put electrodes in the brain. The problem with the brain is it's moving the head, it's quite complex. You connect this to a mass rows of data, the individual data profile.
Of course, you plan this, you connect this with other disease or other problems that might have happened before. You basically take a system approach to what you are going to do on the human. We take, of course, scans. We build automatically the 3D views, like in this case, a vascular with a coronary flows reconstitution. This can be done in a few minutes. Today, the cycle time for this job is several days. I've heard about a startup in Boston trying to do that and maybe wanting to do an IPO. We need to do it faster because I think our technology is far superior. We can rebuild the vascular in a few minutes, and they have a very accurate flow in vascular.
This is the DNA connection with the other data, and we connect those elements together. This is a real cooperation going on. In fact, the platform for Living Brain, I think, Patrick, if I Patrick Johnson, the Director of Research. Stand up. We started together the Biointelligence project. We are doing that, and we are starting clinical trial of the different platform itself for the virtual brain right now as we speak. You see the process here. You can understand that if you, some of you knows how it works today based on documents and publications, based on big tables of data, on graphs. This is a new world because you have the world of collaboration, the world of patient journey, the world of modeling and simulation, and the world of real data coming from clinical trial others.
Welcome to the new core industry for Dassault Systèmes. That's the summary on why we are making this. It has been well prepared. With that, Tarek, join me to tell us how great you have been doing things. I'm so pleased that you are here and that we are together.
Me too.
To make this happen.
We are so excited to be part of the Dassault family, thank you, Bernard and Pascal. I'm Tarek Sherif. I'm Medidata's co-founder and co-CEO. I think just as I thought, you know, after a year of 10 years of being a public company, I thought, "Okay, no more analyst presentations and investor presentations." Bernard and Pascal pulled me right back in. It really is a pleasure to be here and to join with all of you today. I'm gonna give you a little bit of an overview of Medidata, and I think, fill in some of the spaces that Bernard created in terms of talking about what was the rationale and what is it about Medidata? Who are we as an organization and why we're so excited to be joining forces?
I think in order to get a sense for us as a company, it's important to understand our mission. From day one, we cared about building great technology that would impact patients' lives. That's something that underlies our entire culture of our business. The people we attract, the customers that we work for, they all care deeply about impacting patients' lives. I think in a small way, we've been able to do that, and I'll walk you through a little bit of the history of the company. What brought the two companies together was a shared vision, and that vision is that there's a big transformation that's happening in the world of life sciences, in the world of the way therapies and drugs are discovered. We both, as organizations, have a passion for innovation. We care deeply about impacting society.
When we saw that that vision that we have, that we could accelerate it together and make it more of a reality sooner, that convinced both sides that we should come together. I think that's what we're so excited about. The words get used a lot these days. I assume, you know, in this community, you've heard of the idea of precision medicine. Precision medicine means that you're targeting smaller and smaller groups of patients with highly tailored therapies, ultimately getting to the individual as having a tailored therapy just for them. The entire industry of life sciences is built around an entirely different model, and that model is around coming up with an idea, manufacturing it, and giving it to millions and millions of people.
The business model, the infrastructure, all the processes are geared to mass distribution, not to the concept of precision medicine. The reality of the science, the kinds of improvements that we're starting to see with things like CAR T therapies are that when you do focus on small groups of patients with specific biomarkers or even on an individual, the results are remarkable. You start saving people's lives. The transformation that we together want to enable, and I think as you hear more of the presentation, and you get to know us a bit more as a combined organization, you'll understand why we are best positioned to help that transformation to happen in life sciences because it is going to happen.
It needs the kind of company that does so along with Medidata and BIOVIA and all the other solutions that we have. We need to come together in order to make that transformation a reality. Just a little bit about Medidata's, from the very beginning, we were a cloud-based company, a subscription-based company, and that's something that we're bringing into the Dassault family. Give you some stats on the business. Currently, we have about 5 million patients worth of data. Those patients are some of the sickest people in the world. They have rare diseases. They have cancer. They've gone through clinical trials, and I'll talk a little bit later in the presentation about why that's so important. We're about a 3,000 person organization. We're global in nature.
Clinical trials are run around the world, we have been the software infrastructure that has run clinical trials in over 140 countries around the world. Interesting stat, in 2018, 13 of the top 15 revenue-generating drugs in the world have been developed on Medidata software. We are a very key supplier to the life sciences industry, to pharma, to biotech, and to device manufacturers. In total, we've run about 19,000 trials, I'll give you a little bit of perspective on that later on. We have about 1,400 customers, that number has been growing fairly rapidly. Just to give you a little sense of the business since we began.
We started with a fairly simple idea back in 1999, and that idea was to harness the power of the Internet to take what was a very manual, slow process and bring the digital world to it. What the core of our business was and continues to be is something called electronic data capture. It's the idea of bringing data in using the Internet, helping to manage that data. That was a paper-based process back in the nineties and before then when you were developing new drugs in the clinical development process. The area of drug development that we focus on is when the work is done in the research lab and you begin to do testing on patients. That's the clinical trial phase. I'll get a little deeper into that.
We saw an opportunity to take something that was a very manual process, very error-prone, very lengthy, and very expensive, and harness the power of the Internet and software to make it much more efficient. Over time, we saw other areas in the clinical development process that were equally inefficient and manual, and we started to apply technology to those. I'm not gonna walk you through all the different solutions that we came up with, but I think you can see that over time, we started to build out our footprint within our customers, and that's very important. When Rouven Bergmann walks you through our business model, you'll understand why we had such great growth over 20 years and why we maintain sort of the strong customer relationships that we have today.
Over time, what that led to was us developing a platform that you can view as the system of operations within life sciences. On the left-hand side, what you'll see is all the various inputs of data that are required when you're running a clinical trial. It may be data from the clinician who's running the trial. It may be sensor data, more increasingly from patients today. It may be images. It may be genomic data. It may be consenting to being in a clinical trial or lab data. All the things that are inputs to making a decision on whether you should move forward with developing a drug or not. On the right-hand side, you see some of the new data flows that are coming, maybe from EMRs, maybe from social media or claims data.
That's the kind of data that you need to enrich the decision-making process in clinical development. Obviously, you need advanced analytics to make those decisions. Some of the strategic pillars of our business are to be that operating system at the core of the decision-making process for our customers. We are the technology that they rely on. Much like you would rely on the phone service for your communication, we are the core infrastructure that pharma companies and biotechs rely on to help them manage the data and to make the decisions on whether they should move forward and how to move forward with developing new drugs, which, as you know, is critical to their success. It's obviously a very strategic role that we have. We surround the technology that we develop with best-in-class services.
We live in a very domain-specific vertical where domain expertise is very, very important. We've spent 20 years building the knowledge to be able to serve our customers effectively because providing technology is not enough. You have to make sure that you deliver the value based on the service that you deliver with that technology. I think we have a very strong reputation for that. When we win customers, we never lose them. Our customer attrition rate, we used to report it publicly, was less than 1%, much less than 1%. Our turnover in customers was less than 1% for most of the history of our business. We're very sticky, but part of it was delivering great technology, but part of it was the delivery itself. It's the service people in the front line who our customers trust.
Increasingly, an aspect of drug development that I think is a little bit less appreciated by broader audiences is that the patients are getting much more involved in drug development. They are inputs into the development process now. They wear sensors, they provide objective data back through diaries that they fill out. There's more data that allows you to make qualitative decisions, but that are objective about how efficacious a drug is, and that becomes more and more important as you're targeting patients in smaller populations. You're gonna hear a lot about this throughout the presentation. We're in a data business, as Bernard pointed out, maybe drug development and the entire process hasn't been as efficient as it should have been.
Well, for a process that should be based on data, the analytics have not been very advanced in our industry. That's something that's changing right now. The advent of artificial intelligence, much more rich data sources, that's bringing about a renaissance in how people think about drug discovery and drug development. I think we're at a very interesting time coming together to enable more of that to happen. I would say that, in terms of thinking about the life sciences industry overall, it's probably one of the best times in a century to be in the drug development business. It's also a great time to be in the business of providing technology to the companies that are developing these drugs. Obviously it's a huge market, right?
$1.2 trillion or more gets spent on therapies and drugs. That number has been growing very rapidly. The process of developing drugs is about a $100 billion business plus. One of the interesting facts about that is technology plays a very small role right now. I think I saw a statistic, a couple of years ago, actually, that said behind the U.S. government, pharma was the least advanced in adopting cloud technology. All other industries were further along, yet they spend so much money on developing drugs. There are a lot of compounds that are in development. 16,000 different drugs that are currently being developed. We'll go into some of the issues in a minute.
You look at the number of drugs that actually come to market in a given year, the FDA approves something under 100 drugs a year. Think about that. 16,000 drugs being developed, in any given year, in a good year, it's 100 drugs that are making it to market. That tells you something about some of the inefficiencies and the problems involved in the drug development process. Before I go into that, I do wanna reiterate something. Pharma R&D has very long cycle innovation waves. We came off of one in the 2000s, where there had been a series of blockbusters that came out in the 1990s that drove growth for the pharma industry, that cycle ended. There wasn't as much innovation.
There was much more focus on cutting costs, on M&A, on bringing the industry together and consolidating. Over the last five years, the next wave of innovation has started, and these tend to be 10, 15, 20-year cycles. You're starting to see a lot of innovation coming out of the life sciences industry. That innovation is very good for patients, it also means there's a lot of opportunity for companies that can help pharma companies transform. That's what we get so excited about. There's a very, very long opportunity that's starting to open up in front of us. Let's look at some of the stats. You have a one in 10 chance of having your drug that you went through multiple years bringing through research to get to a phase I study.
There are three phases that you go through before you can commercialize a drug, typically. Sometimes it's a bit more than that, but just to leave it very simple. You have a one in 10 chance of getting it right and getting this drug to market. As Bernard said, you know what? Can you imagine if you design a plane and you have like a 10% chance that it's gonna fly? Those are not good odds. They spend a lot of money to bring a drug to market. $ 2.6 billion is not sustainable because, while it's not as much as an airplane program, there are people who run a lot of these programs, there's a lot of money going into the development process.
If you're targeting smaller and smaller groups of patients, it means the revenue on the other side is going to be smaller. If you develop a blockbuster and you can sell it to millions of patients and it drives $10 billion of revenue, well, then a $2.5 billion investment at the front end is easy. If you know that the maximum revenue you're ever gonna get because it's a small group of patients that you've targeted that have the right biomarker profile, maybe you can only spend $ 250 million or $ 400 million. The industry is not currently equipped to spend less on smaller targeted groups. The timelines are because You would expect and as it should be.
Regulations have gone up, and I think that's been a good thing for the most part because it means we're bringing safer drugs to market. I think one of the other annoying or difficult facts is that even when you get your drug right, even if you've gotten all, you know, you've managed the risk, you've gotten it to market in a timely way, only half the drugs that come to market ever achieve the revenue potential that was forecast for them. There are a myriad of reasons having to do with reimbursement, having to do with marketing, et cetera, that may cause that. Overall, what you're hearing is it's an industry that is, as Bernard likes to say, very wealthy. It's cash flow rich. It drives a lot of revenues.
It drives a lot of profitability, it has some systemic problems, especially in an era where the underlying science is shifting in a way that they have to transform. I think that, you know, to us, that means nothing but great opportunity. The opportunity comes in a couple of different ways. Our value prop, both singular as Medidata, as a standalone company, and now together with Dassault, is actually pretty easy. The first one is we're gonna improve productivity. That means we're gonna help you to get your drugs to market faster. We're gonna help you to do it at a lower cost, which obviously you have to be able to do. We're gonna help you maximize your ROI. You get the return, you get your drugs to market faster.
I think one of the other things that we have as a value proposition is using data in a meaningful way to help you make better decisions, to reduce the overall risk when you are bringing a drug to market or when you are bringing a drug out of research into the development process. Ultimately, we wanna improve outcomes. We wanna make sure that the right drug goes to the right patient at the right time. Couple of things about Medidata specifically that make us unique is I mentioned earlier that we have 5 million patients worth of data. That data is incredibly valuable. About a decade ago, we started to ask our customers for the right to use their data on an anonymized basis, both their operational data and their scientific data.
It's an amount of data that cannot easily be replicated by anyone else on the globe. We have data that is global in nature. It comes from every therapeutic area. It's cross-industry, most importantly, it's very rich and deep. We know everything about a patient in a clinical trial for the period that they're in that clinical trial. We know every measurement. You can use that data to drive real meaningful insights. Again, I won't go through all of these, you can help to demonstrate the value of a therapy that you have in development by comparing it to the standard of care.
You can make better decisions about where to target, which doctors to target, or which patients to target in your clinical development process, which means that you'll be able to get your clinical trial up and running faster and done faster. The leading cause for delay in clinical trials and higher costs is that you can't accrue enough patients. We can help you make better decisions about where to find patients. We can help you make better decisions about whether you should proceed with your drug development process or not. We can help you to explain to regulators and to the payer community and to the providers why your drug that's in development currently is better than the standard of care or anything else out there.
Those are very valuable insights that our customers have never been able to get before anywhere else, and they can get them from Medidata today. I'm gonna ask Rouven Bergmann to come up for a minute and just spend some time explaining our business model to you.
Thank you, Tarek. Good morning, everybody, from my side. You saw Tarek talking about our strategic pillars, what differentiates us as a company. What I would like to spend some time on is to explain to you what are our strategic components of our business model. How does Medidata actually work? One of the really important concepts we have is the land and expand model, and that's built for long-term and durable growth. There are five vectors, what I would call it, that are very important to understand about our business. The first one, it's always good to start with customers. You heard about the 1,400 clients that we have.
When we started 20 years ago, we started with the first one. Over time, we saw also the chart that Tarek showed, where the curve was going. There's an acceleration of customer growth in the last two to three years. What's so important with this customer base is that it captures all segments of the market. We have customers in the top segments of the largest pharma companies in the world that run the most complex portfolios of clinical trials across many therapeutic areas, med devices, pharma, everything together under one roof. Also, we are able to cover the biotechs, the companies that just started and have one or two trials. What's really important to understand is that we serve this market with the same standard solutions. Our cloud scales across this vector. Right.
They're not different sets of Medidata versions that these companies are using. No, they are starting with that system, and they're scaling their growth with our platform over time. That's very important to understand. The ecosystem, we say here 10 of the top 10 CROs, very important because those clinical research organizations, they give us access to our biotech market. They serve the industries in operationally running clinical trials. It's very important to work with them and enable them to be more efficient. For us, it's also a very efficient way to get to market and addressing the biotech firms without having to have a direct sales force that covers all these small companies. The second vector, revenue retention. Tarek talked about the revenue retention and the record levels of revenue retention that we've been enjoying.
Really what I think is important for you to take away from today is that when we have a customer, the customer stays with us. That's reflected in the numbers. Even if we have a customer, we have a track record of expanding our relationships with these clients. On average, over the last two years, we've been able to expand revenue when we renew with a customer, the revenue commitment on an annual basis. On an annual basis, like for like, by over 25%. Going through a renewal cycle is, to us, a source of growth because we can work with our clients to expand our share of wallet, our relationships. Our product is very sticky. You saw also the journey of innovation that we have.
We have multiple products that we can offer as part of a renewal cycle. Long-term relationships. Typically, our subscription contracts have a duration of two to five years. We renew our contracts, that gives us a lot of visibility in terms of our revenue model, they expand with us over time. One point that I don't want to lose mentioning is the differentiated service offerings that we have, which is very important, not just from an implementation perspective, but these customers stay with us. They have ongoing support services that they buy from us. They really rely on our capabilities to help transform the clinical data management and operations. Revenue mix. Something we are very proud of. We've been, over time, able to build a very stable model.
About 85% of our revenue is cloud subscription-based, with a high subscription growth margin, and 15% is services revenue, of which about half of it is recurring support services. Also for those type of offering, we have very good visibility, because they go co-terminus with our subscription revenue. The last point, I think very important and a source of our success, and I think we, as Medidata team are very proud of our pricing model because the way we've designed it is transactional consumption-based.
Essentially what this results in, as our customers are growing and are running more trials, we are able to grow our business and our revenue because how we price our service offerings or our product offerings is based on the number of trials that they have a subscription to use our platform for, as well as the number of patients they can enroll or the number of sites they actually enable to enroll patients. It's very granular defined, and it gives an ability as we grow our ecosystem, we grow our revenue. I think that's five very important points to understand. When we put those five vectors into action, what does it result in? First, we are growing our customer base.
You see that in 2013 to today, it's at 3.5 times acceleration in number of customers. Today, Medidata is the standard, the gold standard in life sciences for clinical development. Most of the companies are using us across the world. That's what's reflected here in the number. The other important point is, so when we have a customer, we are very focused on expanding our relationship with these clients. You see this here on the, on your left side, that's cohort slide, which is three segments that we chose to kind of want to demonstrate that to you. There's the green one, which is the biggest one. These are customers that we acquired prior to our IPO.
When you look at the revenue we generated with them in 2013, a $1 in 2013 equals about $2 in 2019. We've doubled that revenue with these clients over time. It comes because they're using more of our products, but they're also running more trials. We have done the same thing with clients that we acquired between the IPO in 2013. For those that joined us from 2014 onwards, we are now 3.5x Up. Every time we have a customer, we are able to expand relationships and grow our business. I think that's very important to take away as well. How do we actually execute this land and expand strategy?
Our opportunity to grow within our customer base. Bernard talked about the framework that Dassault Systèmes has built to grow. That is kind of our framework and how we've designed our business and our growth strategy, is to start sometimes small with our customers. Typically, they start with core data capture, the core data capture platform. Over time, as we've been innovating and investing and make it more seamless to work and expand into the platform, we have an opportunity to attach more product. Attaching more product gives us an uplift of two to five times over time.
When we get to our customers to the next level, where we really are embarking into a transformation journey through data science, analytics, and our unique assets that Tarek walked you through, then we have an ability to go even above that 5 level. We have a number of customers we have done this, and I think when we later listen to the panel, Glen and the team will actually show you how this has worked for some of our customers and how successfully we've been able to put this strategy into action. With that, I hand it back over to Tarek. Thank you so much.
I think the big takeaway here is we have an industry that's in the midst of or in the beginning of a transformation. By coming together, I think we can both drive that transformation, but also create an enormous growth opportunity for our combined organizations. Because once we're in an account, we tend to grow our revenue profile in that account. I think the various solutions that Dassault brings to the table, as well as those that we'll be developing over time together, give us an opportunity to really transform our customers, create a lot of value, but also generate a lot of growth for our combined businesses. Just wanna take a second on our culture because I think it's so essential to understand a bit more about Medidata.
The alignment between Dassault Systèmes and Medidata has been so strong. I think in a year of getting to know each other and in now starting to work together, we feel really good about where we are together. As you know, you can have deals that make or acquisitions that make a lot of sense strategically in one way or another, but the thing that has to align for it to work is the culture. There has to be that shared passion. I think we see that. You know, we, our folks could not be more excited to be part of the Dassault family. The way we've been welcomed in as an organization has really been amazing. Just, you know, corporate social responsibility, very important at Medidata as it is at Dassault Systèmes.
We've both done a lot of things proactively internally, and we've gotten external validation around that. One other thing that's quite important, we work in a highly regulated industry as does Dassault Systèmes, in at least in some of the industries that they're focused on. We have put a lot of time and investment and thought into making sure that we meet all the data privacy requirements, that we have data that's secure as, at least as secure as it can be in today's environment. That's important to our customers. It's something else that they focus on. I wanna leave you with just a thought here, which is that I've talked a lot about the opportunity, but there's also a purpose behind it.
I think we share this caring for having an impact on society, having a broad impact. As we come together, we can transform an industry. I think over the longer term, what we really wanna do is impact the entire society by how we interact with the overall healthcare ecosystem, and that is very, very exciting, I have to say. With that, I think we're gonna open it up to Q&A. Is that right? I'd ask Bernard-
We will.
Pascal
May I ask you, Bernard, Pascal, Rouven to join us on the stage. I'm seeing the first question. Yeah, we do. Let's go.
Great. Thank you. It's Mohammed Moawalla at Goldman Sachs. A couple of questions. First just for Tarek and Rouven on Medidata. Can you just clarify up front in terms of the data that you hold, who owns it? What are the sort of the security arrangements you have in place? Is that all fully in the cloud, or is some of it sit on premise?
Yeah. Our customers, our end customers, pharma, own the data, as do the sites, obviously the ones who generated it. We have secondary use rights to it. We anonymize the data. We do hold it in the cloud. As I said, we take a lot of precautions around the security. Clinical data is a little bit different from the data that you typically see around consumers in the sense that it starts off in a de-identified way because most of that data is when you enroll a patient into a trial, they become a unique identifier rather than having the phone number and the patient name, et cetera. It's got multiple layers of de-identification in it.
There are some use cases where we are collecting data directly from patients now, but obviously we're very sensitive to all the various privacy requirements around that and the consents that are required. In a typical clinical trial, you get consent from the patient right up front to use their data.
Okay. Secondly, I mean, you talked a lot about how much money the pharma industry is spending on drug development, yet also how extremely inefficient it is. It reminds me about banks and their spending on IT. Can you give us a sense of what the IT spend is among the kind of pharma and biotech companies as a percentage of your budget if you know that? As we think about that shift, is that gonna grow in absolute terms? Within that, is there any specific shares of wallet?
Yes
that you are sort of targeting both independently but now with Dassault?
Yeah, absolutely. I'm gonna get a little bit on thin ice because the numbers are hard to come by. Sort of mid-single digits is historically the number that's been thrown around for IT related to clinical development. That number's absolutely going up. Cloud adoption is starting to pick up, and that plays directly into our hands. The focus on AI and how you can get more value from the data, and that's one of the areas where really, I think the industry has a lot of opportunity going forward because the data was historically used just in, from the perspective of let's run statistical analysis on it, right? Not using the rich data and not looking for deeper insights. That's something that's very changing. It's slow right now, but I think the it's going to pick up the pace.
The opportunity for us is to turn, you know, that spend that's in the mid-single digits into a double-digit spend. Obviously, that's a huge opportunity. I'll leave it also, if you wanna add on anything.
No, I think, when we have developed the plan, the bet is exactly what you said. Today, it's in average between 4% to 5% of spend, in the IT related to the total spending. The goal is at least to reach up to 10% in the next five years.
Right.
This market is really under-penetrated, and you will see in my presentation that it's highly fragmented with many niche players. Just because the offer is not well-structured, there is limits right now of what you can do.
Last one for Bernard Charlès. As we think about the platform you're looking to build here, you've obviously got now the data. You've got a lot of the kind of tools within Dassault Systèmes. Obviously, you talk about analytics, Veeva is sort of closely aligned with Salesforce, who have also made some analytic acquisitions. Is there something you need to further augment in terms of your analytic capabilities, or do you think you have that sort of end-to-end platform complete now?
I think we saw the due diligence and through our recent discussion with, especially with Glen here, I think we find out that Medidata has an incredibly powerful data science with Acorn. Great presentation yesterday with customers. If you add what is in Medidata and what we have on the front end side, I believe that we can quickly show the difference as compared to everything that exists today in the industry. We need to showcase and connect, but I think this should be a very big differentiator. Related to the competitive landscape, you know, I will do a lot of analogy with the competitor you mentioned on some of our past competitors in the other sectors, if I may. It's very thin. We are very deep.
I don't think documents will win.
Hi. Hello. As you bring the two companies together, would you envision as the processes, products, and systems and approach come together, that you would be sharing with us over the coming years, the types of long-term partnerships that you've announced in aerospace, whether it's with Airbus or mining with BHP in the life sciences sector? In your initial strategic thinking, can you sort of share with us what you think that opportunity looks like, above and beyond the clients that you already have independently together as the company becomes one?
Yeah.
Thank you.
Yeah. Well, I think there is If I give you some adjacent inputs to the way I see the decision process evolving in these industries. First, a few years ago, when we initiated the contacts, thanks to BIOVIA, Accelrys and BIOVIA, we discovered that in these kind of companies, the decisions are taken at a very relatively low level, at least when it comes to the R&D or manufacturing. Not speak for clinical trial, but for the side that we know. This is elevating now. When we meet now with those companies, the CEO wants to be involved, and it's not long ago. They want to understand. They are trying to better understand their transformation roadmap. It happened the same way in the other industries we have been serving that you just referred to.
Of course, this is a very critical factor. Who owns the decision process? Fragmentation in what they have today is based on the fact that they have left the decisions to the specialist. You know, the specialists don't care about what the other guys are doing. They just want their nice toolkit for what they do. This is why the process is broken. There is no digital continuity in those companies. Almost zero. Now, they talk about data lake, just thinking that putting things in a big tank will solve the problem. It does not. It does not solve the problem, because if the data don't understand each other, you do not have the proper semantic. You just have a collection of data, and you cannot do anything with it.
In short, yes, we see the evolution of ownership of the decision-makers, and I think this is a condition for those long-term big contract to be set up. I'm convinced that it's going to happen. Too early to tell you how fast. I'm getting an insider experience from something else I'm doing as a board member of a pharma company, so I'm seeing it from the inside, which is eye-opener for me as being a member of the board of Sanofi. I think that syncs really well from that standpoint.
Just if I may add on, Bernard, I think, what we're starting to see is that the CEOs of some of the pharma, this isn't that broad yet, and the boards are they're basically charging someone in the organization with thinking about building or building a strategy around digital transformation. How long that takes and how quickly it's adopted is a different question, but the conversation has been elevated to the board level. There are organizations that are being more aggressive, some are being less aggressive, but it is a conversation that's happening at that level now, whereas in the past, the decision-making had always been much lower, you know, head of clinical management maybe deciding which vendor to use for their infrastructure. Now that's, now that's being elevated because these are, these are major transformational programs.
Thank you. Good morning. Jay Vleeschhouwer. A structural question and a practical question for Bernard and everyone else. The practical question is, have you developed the product integration roadmap? Can you talk about some of those details just at the core software architectural level in terms of taking, let's say, DNA from BIOVIA or SIMULIA or even the manufacturing side of the company and integrating that in some timeframe that you can talk about with Medidata? Then the structural question is, Bernard, you've talked for years that one of your highest priorities is increasing the number of DS users. You are getting a good customer base, of course, with Medidata, but the number of customers, as it was with Accelrys, is not particularly large, although they've grown it.
How do you think about the growth dynamics or vectors in the context of a fairly concentrated, relatively small number of customers to drive revenues from the acquisition?
Well, I think later on, there will be discussion on Glen. Glen would be well-positioned to tell you how much he fall in love with the 3DEXPERIENCE platform. I think that Medidata is a very powerful platform for clinical trial. It's very well done, cloud-based, good technology. We have the front-end platform for R&D and also manufacturing. ONE Lab from BIOVIA is being extremely well architected now on the experience platform. Jason will show that this morning. I think it's going to be relatively easy to connect on because it's not a rewrite of the code. It's a connection of two great platforms. On the number of customers, you know, I think it's related to the previous questions. Who owns the decision in those companies?
The reality today, it's a very fragmented set of systems that they have. Very disciplined, fragmented. We saw that 30 years ago in other industries. There are so many limits with that. I think our responsibility is to showcase that the transformation can happen. It's now, it's possible, and the niche players should be replaced by a consistent digital pipeline for the industry. It's up to us to work with customers to do that. I think.
Maybe I should add a few numbers because I'm not in agreement with what you said. You have close to 5,000 pharma. Today, only 1,400 has been served by Medidata, almost 800 by BIOVIA, and we have an overlap of about 600. You still have most of half to be conquest. That's point number one. Every year, you have almost 1,000 startup coming with active molecule. They are an active pipeline. Those guy are coming on a regular basis. If you look at the medical devices, we are talking about 50,000 people-- customers, sorry. Today, we are reaching 10,000. Not taking into account the hospital, the practitioners, those are the supply chain we want to connect with the platform.
Coming back to your question, I think if you look at the footprint of this sector, it's as large as the industry is today.
yes, thank you.
Yeah. Stefan Slowinski from Exane BNP Paribas. A question for Tarek, kind of along the similar lines of the last couple questions on customers. You talked about the land and expand strategy. You have 12 of the largest 17 pharma companies. I guess if we focus on the 5 that you don't have, why do you think you haven't been able to win them as customers? What do you think it will take to penetrate those accounts? Also just wondering if you have any kind of initial feedback from your clients and customers on the acquisition by Dassault Systèmes.
Let me start with the last question first, and I'll get to the other two. The response has been incredibly enthusiastic and positive. A lot of our customers know Dassault Systèmes, and they trust them and they value their software. They're really happy with us coming together. In fact, it's kind of been an interesting challenge because they're like, "Okay, when are we gonna see, you know, the fruits of this?" They wanna get going. They wanna see everything come together because I think their view of how the industry needs to transform aligns with our view, and they just wanna get going with that. It has really been universally positive, and that's gratifying. It means we weren't, you know, we were right when we made this decision.
I just wanna add something as it relates to revenues, by the way, to what Pascal was saying, which is, our model has always been, yes, it's important to add new customers, but the stickiness with the customer and driving increased revenue is how the model has been so successful for us. I would say in any given year, 80% of our revenue, 85% of our revenue increase comes from our existing customer base.
If you're talking about the kinds of numbers that are out there in terms of new customers that we can acquire and the stickiness that we have at, you know, with no turnover in customers, negligible turnover, and now a second, you know, a larger solution set that we can sell into them, and then that doesn't even come for the things that we're going to be developing. I'm, you know, I suppose I wouldn't be here if I wasn't as enthusiastic, I can't even speak anymore, got too excited, about the idea of what we can do together.
The low IT spend today.
Yeah, the low IT spend. As it relates to the specific customers, I probably spent a decade answering this question. It's always kind of been we have over time added more and more of the top 25 pharma to our list of customers. We're gonna get them over time. Some are slow to move. Change is very difficult at pharma, especially when technology's built into a process. It can take years to make that decision. It's a double-edged sword. When you're the incumbent vendor, it's very difficult to displace you, even if the newcomer has great technology, because you have so much process built around it.
Change management is so difficult, so expensive, and takes so long that it's in a risky part of the business because if you get clinical development wrong, you're talking about the future revenue generation. We always point this out, but we were a billion-dollar, almost a billion-dollar revenue company. I'll give us some credit. We were running probably what the future of $1 trillion of drug development on our software and in our database. You are core to their future, they're very slow to change you. I think we have the best technology in the industry. I think we have the best services folks in the industry and domain knowledge, we're innovating like hell, we're gonna be moving even faster.
I think over time, those organizations are gonna come to us. Some of it is entrenchments, any number of things. I'm not saying we're perfect. I'm sure there are places where we could have done a better job of selling into them. Some of the customers that we acquired in the last few years, like Bristol Myers and Novartis. Novartis took us a decade to have them move. Once they move, they stay with you for decades. It's worth taking the, you know, it's a long sales cycle, but it's worth it.
Thank you. Stacy Pollard with JP Morgan. Bernard has provided a vision in life sciences. Tarek, I'd like to ask you, in terms of synergies with Dassault in the life sciences business, what do you think some of the best opportunities are for your business intermingling with Dassault? Perhaps some specific examples of what you think you'd be doing, which technologies could come together. I know, of course, on the technology side, you've got Dassault's data platform. You've got perhaps OUTSCALE as an opportunity, but I'm more interested in the science-specific examples.
Can I ask you to hold that question till we do the second part of our presentation? We can help you with it visually too. We'll give you a more in-depth answer than I can give you in a, in a, you know, in a short period of time. What I would say is that they are what Dassault Systèmes has today and the direction that they're moving is so complementary to the direction the industry overall is moving. Right now, research is siloed from development, which is siloed from commercialization. Those silos are starting to break down, not because of some great epiphany, but because the economics of what regulators and what the providers and because of the payers, they're pushing pharma to change the way they develop drugs and even conceptualize drugs.
That is all going upstream, all the way up to research, and that's where that combination of knowledge. Researchers right now typically don't get any feedback on what's happened in the development process and even less so what happened in the commercialization process. If we can bring sort of these, this disjointed data back to the researchers, you're going to get a feedback loop that produces much better drugs, which again, will reduce sort of the timelines that they have for getting drugs out. It will improve the efficiency. We'll talk a little bit more about that in the next presentation in a few minutes.
We take one last question for this session.
Yes. Thanks very much. I was wondering if you can comment on trends in vendor consolidation in the clinical trials IT space, as well as do you foresee major share shifts, given, you know, you know, some of your key competitors wanting to gain share of voice with some of the clients that you have?
Sorry, I didn't catch the first part of the question.
Vendor consolidation in.
Vendor consolidation.
You mentioned-
Vendor consolidation.
Yeah. Yeah.
Yeah. That's an interesting one. There is some consolidation that happened when we were doing some smaller acquisitions around the core of our space just to bring some additional technology in. Roll-up strategy does not work well in our industry. We always stayed away from it, and I think what you'd see is that for the most part, it doesn't happen because clinical trials are very discrete units. Unless something's going terribly wrong, nobody wants to move from one technology to another in the middle of a clinical trial. It just is a disaster to try to do. We've actually been given some trials because using other technology has been such a disaster. It's not easy to do. We are, as Rouven mentioned, and we are the standard in the industry.
We run more than 50% of all clinical trials. There are some competitors that have tried to come into this space with very, very limited success. On the periphery, if you're doing things like document management, it's a process, it's a lot easier to do. If you're dealing with core clinical data, it's the market share displacement or gains are very difficult to come at. I think we're a provider that's viewed with a high level of integrity, high level of innovation, good services organization. Very few of our customers are unsatisfied with us. I mean, if you walk around the halls outside, we've got 1,000 people basically talking about how excited they are to use us and, you know, to see the innovation that comes out of it. I don't see major market share shifts happening.
I think that we are gonna continue to be very aggressive. We've been taking market share for since we started the business, and I don't see anything changing in that regard.
Maybe one aspect to add, actually, the need for simplification and consolidation also has been a catalyst, where IT departments look to really consolidate and look for partners who can cover the process end to end and really have that vision to move into the future. That's where they come to us. We have examples where we have customers that move to our platform. They were able to get from over 40 disparate systems to below 20 and further down, right? They did this because we were their partner. That is something that, of course, that's part of our strategy, simplification, consolidation, and take it all on.
We do the break? I propose a very short break, 10 minutes. If you could be back at 10:30 A.M. sharp, that would be fantastic.
I was just have one comment. If you know Pascal, he cannot smile. Seriously, he got an accident 2 days ago, so I speak for him. We suspect a competitor wanted to clap the door on his. We cannot find the competitor yet. If he's not smiling, it's because he has 10 stitches on his, on his, lips.
Yes, I'm done.
Are you okay?
I'm okay. I'm okay.
Very good. Just for you to know.
10:30 A.M. Thank you. Ladies and gentlemen, may I ask you to be back to your seats in the room? Please go on. Yeah.
All right. Hello, everyone. Can you see me well? Yes? It's my great pleasure to introduce the next session this morning. I will share the stage with Glen de Vries, who is the Medidata Co-CEO and Co-founder, and David Geddes, who is the Vice President for BIOVIA R&D. I am Claire Biot Vice President for Life Sciences Industry, and we are thrilled to be with you today to present Medidata combined with Dassault Systèmes joint value proposition. As you've heard from Bernard earlier today, we have a big belief at Dassault Systèmes, which is that the virtual world extends and improves the real world. What does this mean for life sciences? Well, of course, we will still need real therapeutics to treat real patients, but we're convinced that these therapeutics can be optimized thanks to the virtual world, thanks to modeling and simulation.
As many of you know already, and Bernard covered that earlier today, it is a major challenge to develop a new drug, moving from early stage discovery all the way to commercialization. It can take up on average 10 years, $ 2.6 billion, and it is a major challenge for life sciences companies to move to accelerate time to market. Why? Because they want to address patient unmet needs faster, and they also want to increase the duration of market exclusivity. Down that path, of course, there is a scientific challenge moving from hundreds of thousands compounds to a single substance. Also, when this substance has been discovered, you want to validate it, and it becomes a challenge if you want to move towards precision medicine. What are the right populations that you want to treat? Which clinical trials should you run?
After that, you want to manufacture this product into reforms, dosages, packaging, depending on the market and countries. Obviously, there is a massive challenge in terms of science, but there is also one of managing complexity. If you look at top 20 companies annual reports and industry analyst assessments, what are the internal and external forces that drive the need for change in life sciences companies? We identify five of them that I will briefly describe now. The first one is personalized health. What we mean by that is that life sciences companies aim at developing a holistic approach to care, leveraging human data, genomics, but also behavior and the environment together with technological breakthroughs such as IoT, AI to achieve precision medicine.
Tarek already covered that and Donna as well, but it's very important for us, I will insist a bit more. This is not about developing a different treatment for each and every individual, but here it's about considering individual differences over the course of prevention, diagnosis, and treatment. The second challenge is knowledge capitalization. This is a need for transformation that we know well at Dassault Systèmes because it's well shared across industries. Now, importantly, over the past decade, life sciences companies have grown a lot by merger and acquisition, and you have witnessed that happen. As a result, they are often divided into numerous isolated divisions. To manage this, most of the companies have organized complex matrix-based organizations to enhance communication and data exchange, but much more is required.
Knowledge capitalization is requiring connecting people, systems, and data in a virtual ecosystem across the entire innovation continuum. The first need for change is total quality. Here it's about ensuring quality and achieving regulatory compliance. The goal is to create a compliance framework for innovation, ensuring that you embed quality and regulatory best practices early in the development cycle and ensure interoperability of the product throughout its life cycle. The fourth need for transformation is to achieve development and manufacturing excellence. Here, why is that important? Well, if you want to move towards personalized products and if you want to drive the cost down, you want to have adaptive and predictive development and manufacturing. What does it mean for clinical trials?
Well, you want to have smart clinical trial design, you want to have agility in the way you run clinical operations. Moving down to manufacturing, the companies want to leverage the technologies coming from what we call Industry Renaissance to be able to manage production processes in real-time, while continuous scale-up and tech transfer should ensure that the product is manufactured as it has been designed and as registered. The fifth need for transformation is about reinventing the value chain. As Tarek already mentioned, life sciences companies have understood that they need to shift their focus from the product to the patient and to the outcome for the patient, and they have to move away from a one-size-fits-all approach to supply chain.
They are seeking to demonstrate value to payers, to regulators, to patients, to physicians, and that pushes them to reinvent their business models and has an impact on the entire value chain. As a consequence of this need for change, we are convinced that the next frontier in drug development is about transforming the patient experience. What is our joint value proposition to meet these challenges? Well, joining forces allows us to create the first end-to-end scientific and business platform, going from early research and discovery all the way to commercialization, including pre-clinical development, clinical testing, and manufacturing. Jason and Glen are going to take you along all of these stages and highlight our game-changing values and synergies along the innovation continuum. If I summarize briefly, in research and discovery, we unify predictive science and experimental results to accelerate therapeutic innovation design.
With ONE Lab, we transform laboratory compliance, efficiency, and collaboration. In the field of clinical testing, we come with a broad range of solutions encompassing data management and clinical operation to accelerate value, reduce risk, and optimize patient outcome. With License to Cure, we connect quality across the biopharma enterprise and provide data-driven approach to regulatory processes. With Made to Cure, we help organizations reimagine engineering, operations, and planning to achieve manufacturing excellence with a special focus on biological processes. In the field of commercialization, we help our customers demonstrate real-world value to payers, regulators, patients, and physicians. Now, before turning to Jason, let me make an important point.
We are convinced that Medidata's core asset in patient data, you have heard a little bit about that already, so coming from clinical trials, but also real-world evidence and advanced analytics will strongly combine with the power of modeling and simulation to catalyze the next generation of patient-inclusive therapeutics. That is a perfect illustration of our belief, which is that the virtual world can extend and improve the real world. With that, let me turn to Jason, who is going to take you through the innovation continuum.
Thank you, Claire. I'm gonna start with a little demonstration of some of the capabilities in Design to Cure. All of our industry solution experiences in life science, we refer to them as V+ R, Virtual+ Real. This complements the virtual world extending the real world. We start by characterizing what we wanna do in our drug discovery program. We refer to that as a target product profile. From there, we're able to structure the drug discovery and development process, ask questions along the continuum, look at analytics, and take decisions. You can do this through the 3DEXPERIENCE platform in a social context, also using the ideation funnel here, for example, to exchange with your colleagues. Not only is this social, but there is deep science here.
This is not a thin platform, but a very deep scientific platform that adds social on top of the science. This helps us break down the traditional barriers where scientists hoard their information to themselves. We can take that information and publish it into a knowledge graph. We heard about knowledge graphs earlier. This is a knowledge graph for biological research data, bringing together multiple omics disciplines into a solution that we call the Living Map. On the Living Map, it's not just a knowledge model. You can simulate. With that simulation, we can find new drug targets or multiple targets for a drug candidate, helping us improve how we target a therapeutic for a patient population.
Once we have that target, in the 3DEXPERIENCE platform, we're able to exploit multiple modalities of drug de-design and development, from small molecule to biologics, and now with emerging technologies in immunotherapies, cellular therapies, CAR T therapies. All of those modalities can be addressed, connected, and designed. We use the word design very purposefully here, because we're making this an intentional design process. All of that knowledge, all of that capability, the know-how can be expressed through the 3DEXPERIENCE platform here in a common data model, in a common user experience that combines this together. Now, at its best, we're able to combine the power of modeling and simulation and knowledge-based models expressed here through our AI-based drug design and development capability. What you see here is a solution we call Generative Therapeutics Design.
Our first target is small molecule, and we're going to extend that to multiple modalities as well. With that, you see multiple modalities are coming out of research. This is putting a lot of pressure on the drug development organization to change their game, too. All of the value that you see here are derived from real value engagements with our customers. We are looking to shave years off of the drug development, design, and development processes. We're looking to help target the right therapy at the right target, and increase the time to market so that we can cure patients faster. Now, picking that up in drug development, this is an area that's undergoing a lot of digital transformation right now, and the potential gains are huge.
I'm going to take you on a small tour of our ONE Lab Industry Solution Experience. The first thing is we start to see a decomposition of the traditional barriers between discovery and development. This has to happen because of the new modalities that are emerging. With solutions that we have here, like Scientific Intelligence, we're able to connect the dots. This is not a simple search engine. This is a knowledge-based data lake solution with specific scientific capability built in, searching the chemistry, searching the biology, aware of the biology, and able to connect the dots from all that information that we created in discovery and now also in development. We need to develop procedures. We need to develop the drug process. With ONE Lab Industry Solution Experience, we've been able to replace multiple disparate solutions.
This comes back to the n minus one question. Do we see a consolidation? Yes. We need to have a consolidation because the knowledge is in too many disparate systems. We see multiple ELNs, multiple LIMS systems, and we're able to replace those with ONE Lab. ONE Lab is not just a replacement for these systems. It is better. It is fully digital. The methods and the drug development process that we're designing here are supported by an ISA-88, ISA-95 structure, and that eliminates the technology transfer between the different silos. What we're starting to see is our customers are adopting ONE Lab not only in process development, but also in quality control and manufacturing.
As they make that adoption in both places, they're able to see the synergy between them, where they no longer have this recoding barrier for the technology moving from process development into quality control. With that, we're able to eliminate further systems, we're able to realize efficiencies, and we're able to get better quality out of the system. We see a big land and expand opportunity there, just like Medidata has seen with their solutions. Okay. Here, we're completing a study. This is a drug formulation and stability study. You can see that ONE Lab is connected to the 3DEXPERIENCE platform, that we can capitalize the drug development knowledge into the Scientific Intelligence capability along with the drug design capability, and all the way up to the clinical batch release. That's not all we can do.
We can also develop the device, the drug delivery device. Here you see a small vignette from our IASO demonstrator. You can go look at IASO. We produced this. It's real. Biologics, we see an increase in the biologics investment. These things are hard to deliver to the patient in a comfortable form, right? They're viscous. They need to be injected or absorbed. With IASO, we were able to design a device that the patient can wear at home or on the road, and it slowly diffuses the large molecule therapy to them. They don't have to get an injection. They don't have to get an IV bag. Overall, you're able to create a better molecule to target their disease. You're able to create a better delivery experience for that disease all in one platform.
With that clinical batch release, we need to test it. I'm gonna hand it to my good friend, Glen, to tell you about that.
Thanks, Jason. Now we've got the right target, we've got the right drug. How do we give it to the right patient at the right time? That is where we come in with Medidata's approach to being the system of operations for a clinical trial. We need to bring multiple modalities of data together about patients in the real world and match it up to our virtual expectations that we've established prior to these tests. What you see here is an illustration of all of those types of data going into the Medidata platform, and let me give you a sense of what we then do with them. Running a clinical trial is an extraordinary complicated business process. It is dependent on doctors and nurses operationally that are not part of the pharmaceutical company or the medical device company that is running that study.
You saw earlier today how unpredictable timing is in clinical trials. It is because of this dependency, in many cases, between the planned experiment and what is actually happening from a patient availability perspective in the real world. On the Medidata platform, we help people plan and execute and then analyze every step that is necessary along the spectrum of executing a clinical trial. We do that in a way that provides context to data, both in the real world and apropos to one of the questions about product synergies and roadmap, actually gives us a context for looking at that data in a more sophisticated way in the virtual world. I'll give you an example of patient data. In this case, it's coming from an iPad that the patient is using to enter data.
This process that I'm illustrating is exactly the same, whether the source of data is a nurse recording something in a specific application for a clinical trial, whether it's a piece of data that's coming from an EMR record, whether it's a piece of genomic data. It goes in and gets categorized from a clinical data management perspective. If you're not an expert on life sciences, which is okay, Claire used two terms which are incredibly important, clinical data management, which is what we're looking at here, and clinical operations. Those are illustrated as Rave EDC and CTMS on this chart. It used to be two separate systems. You heard Rouven Bergmann talking about our clients liking the fact that we bring this together at Medidata.
The data manager is looking at the quality of data, and the clinical operations team here is looking at any operational issues, any flags that are on that data to make sure that it's all collected in a compliant manner that is going to meet regulatory expectations. We bring that together in our platform for every type of data, and we also give it data context. In terms of our platform and the Dassault Systèmes 3DEXPERIENCE platform, we actually think about semantic layers, how we categorize data in very compatible ways. You just saw Jason talking about that at a molecular level. Well, that's what we do at a clinical level, so we can have that data available for reporting from a scientific perspective or from an operational perspective. Operationally, this is terrific.
We actually can help our clients do clinical trials in a more efficient way than they're doing them with disparate systems and older processes. That is one of the things that we have 1,000 clients in this building in dozens of tracks simultaneously going through, learning about all the different benefits that you get from our platform in today's world. The world of clinical trials and the world of life sciences, as you heard Claire talking about, is also changing. People are realizing that they need to be much more patient-centric than they were in the past. The old days of developing a small molecule, a relatively simple drug, and putting it into this complex, giant market, which had its own challenges, are changing. That was a population biology problem. It was an epidemiological problem.
Today, as we start thinking about treating individual patients, as we think about precision medicine, as we think about individualized therapies, it's now a problem that we need to think about on a very personal level. That has implications for clinical trials. One of the things that we think about at Medidata is not a light switch, but a dimmer, a dial. As patients are being thought of in a more central way as individuals, not just as parts of a large denominator, we need to start to be able to shift the way we think about clinical trials from that population basis or from something that you would do in a clinic to something that is done in homes.
Part of what Medidata is trying to do is not just facilitate that in the clinical trial context, but also set ourselves up to operate in a world where patients are the center of a therapy instead of the physician who is prescribing it being the center of that therapy. I just want to take an opportunity to also answer a question about users. One of the announcements that we made yesterday at Medidata was that we are extending what we do in terms of individual apps for patients to actually create a central platform location for patients who are participating in clinical trials to come and participate in every facet of those clinical trials, how they learn about them, how they're consented, how they provide data, how they get data back, so they can understand what's happening with them.
In fact, the life sciences industry hasn't done a great job of thinking about the fact that patients are not just central to one clinical trial, but we should think about a patient in the context of every clinical trial that they go into. A lot of cancer patients go into a clinical trial for one company and then wind up in another trial with another company. We are not thinking of as an industry around how to create that transition. Well, that's what Medidata is doing now with our patient platform.
I think one of the exciting opportunities, I'll come back to this when we get to the commercial part of the spiral that we were showing, is that that can extend into a platform for patients to be connected to the therapeutic companies that are developing the things that are helping them in a much more holistic and commercial way. That means that the 1 million Medidata users that we're adding to the 3DEXPERIENCE platform as of today will actually turn into not just clinical trial physicians, but physicians who are treating patients in the real world and the patients themselves. That goes from 1 million to tens of millions. I think, honestly, literally, as we incorporate patients into the platform more, billions. That brings me to analytics.
When you start to think about these numbers, and in some cases, the numbers are incredibly small. I will give you some examples of that. In some cases, the numbers are incredibly large. There is an opportunity, as Tarek highlighted, for the life sciences industry to do a better job of thinking about generating information from data. This is a Medidata slide, which you will probably never see these layers on again. Well, you will probably see these three layers, but it is missing a really important layer here, which is the virtual layer.
What you will see, again, from a product perspective, is how, if you look at the virtual world and the real world, all of these steps along the process of what Jason was illustrating, what we're talking about with clinical trials, and what we will now talk about in the future of commercial, actually come together with beautiful parallel tracks. You've got a data fabric, again, inclusive of research data on the BIOVIA side, inclusive of all the patient data on the Medidata side, illustrated at the bottom. At the top, you've got the life cycle of all the activities that are being done today in terms of running clinical trials. In the middle, Acorn AI is our brand name within Medidata for our most advanced analytics, some of the most rocket science type things that we're doing.
I wanna give you an example of a couple of them to illustrate what I mean about this analytic future and large and small sets of data. As we think about the world of clinical development changing, as we go from these small molecules going into big, complex markets to large molecules and very specific therapies that are targeted for very small groups of patients who we can categorize and understand very well at the start of the research project, we need to think about a reality, which is that the smaller and smaller sets of patients who get more and more precise therapies may not always provide enough statistical evidence for a regulator, for a prescription writer, for a patient, for a payer to make decisions around.
What we've embarked on with Medidata and with our unique data assets, as Tarek and Rouven were talking about, is the ability to take a single patient and reuse their data over and over again. I know not everybody is a life scientist in the room. This is a survival graph. Everybody at the beginning on this one corner is still okay, and as you go down, you see people are either having their tumor progress or they're having a cardiac event. Some endpoint, could be death, is happening in the study. What we wanna show is that the curve for the new therapy looks shallower than the curve for an existing therapy. Usually, patients are used once and once only in these analyses. They appear in one of those two curves in one submission to the FDA.
By categorizing the data on our platform as we have, we're able to use patients in multiple curves, in multiple pieces of evidence generation. In fact, if you think about the virtual world supplement to what we do here, the obvious, I think, not simple idea, this is incredibly complicated simulation we have to do, but is to actually add something to this, which is inclusive of a virtual twin in terms of being able to predict from a simulation perspective, another survival curve. This is work that we are doing not on paper, not theoretically. We at Medidata present this at scientific conferences. We work with groups like Friends of Cancer Research. We are at the FDA presenting this, and we are working with clients to use this kind of evidence generation for the submission of new drug packages.
Again, there was the question about competitive differentiation. I think that we're making the rubber meet the road in ways that other people are putting out press releases about. I'll give you another example. This was a big data example. Let's reuse the millions of patients on our platform. I'll give you, hopefully, an incredibly compelling small data example. It also fits perfectly into everything that Jason was presenting. Yesterday, we had a presentation from a physician. His name is David Fajgenbaum. Not only is he a physician scientist, he also suffers from a rare disease. It's a disease called Castleman disease. I guess for 20% of the people with Castleman disease, the good news is there's a drug called siltuximab, which can actually keep the disease under control.
It's autoimmune. You wind up going into organ failure, and most patients die two years after diagnosis. For one out of five of those patients, this on-market drug is a great answer. Nobody knows or knew how to figure out what patients would be that one out of five. What we were able to do is by taking data in the exact same process that you saw. Remember we started with data from that iPad? Well, imagine the other pads for data from clinics, for data about the patient's proteomic profiles, so what genes are on versus off. We were able to take data from multiple clinical trials, not on thousands or millions of patients, on fewer than 100.
Bring that data together, stack it up, do an analysis on it in a way that allowed us to identify not how one out of five patients, so a less than 20% chance of being right about this drug working, to a 69% chance of being able to identify patients for whom that drug would work. This is literal precision medicine, doing a better job of finding the right treatment for the right patient at the right time. You may not be a life scientist, but if you're a financial analyst, you're probably good at math. You're probably sitting there saying, "Well, what about the other four patients who we're not helping?" That is where the connectivity between what Jason was talking about and what we have at Medidata comes into play.
One of the exciting things that David was also presenting about yesterday is if you start to look at the commonality of the patients who aren't helped by siltuximab, which is related to looking at Interleukin 6, a particular protein. Well, there happens to be some other commonality. There's another target called mTOR, and there happens to be a drug already on the market that was used in kidney transplantation that can actually suppress that particular protein. And that drug is now keeping David alive and has created a whole path for development of new therapies for patients with Castleman disease. Again, in example one, I was showing you how with data reuse and big data, we could make things that created more evidence per patient who is enrolled in the real world by creating what effectively is a new virtual representation of that data.
In this example, we're figuring out how to take these theoretical views of pathways and molecules that can then be applied to real-world scenarios to create effective, incredibly valuable medications in rare disease, which is regarded as the hardest part of life sciences to get involved in and move the needle on. With that, now that we've got something that we can give to the right patients at the right time, I'll go back to Jason.
All right. We have to deal with some regulatory and quality affairs, I think, as well. This is what License to Cure does. I'm gonna start with the new stuff, okay? This has traditionally been a document-dominated world. It's an arts and crafts project. Filing a CMC report can take 7,000 hours of labor. It's risky. You get audited, you have to show all the evidence. What you see here is a combination of content and data-driven automation. The world of regulatory, the world of quality is gonna shift. On the regulatory side, we're able to, through the semantic data layer, grab the data that we need with full traceability and automate large sections, if not all of these critical assets the companies have to produce.
You can imagine the savings of going from 7,000 hours per filing per year per product, down to maybe 100 hours, okay. It's reasonable. We can achieve it. The objects that show up in these are not simple charts. They're live business objects, lifecycle managed. You can click through, you can see the audit trail. You can go all the way back to the source system. When the auditor shows up, you go to the artifact in question, and you can look at the entire provenance. It reduces the risk for our customers. This is where we're going. Everything's gonna be automated, data-driven, scientific in nature, okay. On the quality topic, if we do our job well in process development, we're doing quality by design. The risk in the quality phase goes down.
We still need to have a quality system. You'll see it in a minute. We have both the quality, enterprise quality management system, and we have the enterprise document management system. I started with the next generation up front. We're putting all this work into building a semantic data fabric so that we can leverage it to replace these thin line arts and crafts projects with real data-driven evidence. It also enables us to do new things like continuous submission. We can do this for both the drug and the device. This is what License to Cure will allow us to do. Then we need to manufacture. Just a few facts here for License to Cure. Again, these are from real customer value engineering engagements. Into manufacturing.
In process development today, most companies are in this phase where they develop the process. It's a little bit ad hoc. They don't have a process model that they're actually building toward. They're building a body of knowledge in a document again. They have to go to manufacturing, and they have to recode this for their manufacturing execution system. We can do better. How do we do better? When we're developing the process, we're able to put the knowledge into a model. Instead of building a document, you're building a model of the manufacturing process, both the device and then actually the plant. You can see here, you can navigate on the plant. You can do virtual commissioning. This is a huge, a huge value to our customers.
You can simulate large or all elements of the plant. You can see the ergonomics of the plant. How is the assembly of the device going to work? Okay. You can plan your production, and you can optimize your production. You can optimize the clinical supply. You can optimize the production supply. You can even optimize the delivery, the optimal delivery to the marketplace, which we're gonna get to in a little bit. Okay. You can go in and you can run your day-to-day operations in the 3DEXPERIENCE platform and manufacturing, okay? This is developed in partnership with multiple brands. Bernard referred to the power of the brands. This is developed in combination with DELMIA brand.
BIOVIA brings to the table a deep understanding of bioprocessing and the ability to monitor in real time the control strategy and to keep your process running, save you batches. Here you can also manage your bill of materials against the manufacturing process to make sure that you're compliant and you're able to transfer the technology from development into manufacturing digitally, not through documentation. Again, this is a huge savings, both in time, error, and optimization of the process. Now that we've produced the product, we have to go to market, I think.
All right. We've got a treatment that works. We've got all the regulatory documentation to prove it, and we're ready to actually get it to that right patient at the right time. What next? This is where to just put a personal spin on it, I love Lego. As Rouven and Tarek and I got to know the Dassault Systèmes team and the platform, it was amazing how these pieces fit together. We can now take the other side of the way Medidata thought of the world before a week ago from Monday. That is not just the clinical trial data, but the real world data. Data from practices around the world, data from EMRs, data around prescriptions, data around reimbursements.
We bring those into our platform in a way that allows our customers to make commercial decisions. Commercial decisions span a couple different things today. It is certainly the traditional view of what markets should I go into? What countries should I launch my drug in? How should I deploy my sales force? Increasingly, the way we think about commercialization of therapeutics, apropos to what I was talking about in terms of patients, is in a very much more patient-centric way. That's not just because the therapeutics are more specific themselves.
That's because we also have a commercial need in life sciences in a value-based care, in a value-based contracting environment to actually show that this therapeutic was deployed effectively and produced the kind of therapeutic value, had the level of efficacy and safety that we expected when it was given to a patient for a payment to be made back or to not give a rebate. What you see here is data coming into the Medidata platform from real-world data sources. It's data that we can put into our semantic layer and actually treat the same way we would think about clinical research data and assemble it into reports that show safety, efficacy, value, commercial success, as well as scientific success, both at a population level and at an individual patient level.
This is actually an example of a project where for robotic surgeries, this is a medical device example, we are not just showing the outcomes of a particular case, a particular surgery for a particular patient, but looking at, from a physician's perspective, from the surgeon's perspective, how well they are doing, as well as providing a dashboard back to the life sciences company, so they can see at a population level how well the treatment is working. This is the final step in going from the initial concept of a therapy all the way to, as we're showing here, its successful delivery. With that, I will hand it back to Claire.
Thank you very much. Let me wrap up this session by focusing on the key benefits that we provide to our customers. Well, important, we are basing these KPIs on value engagements that we have been conducting with our customers, and we are benchmarking this against publications in the field. Importantly, the value we bring to our customer concerns both the top line and the bottom line. As far as the top line is concerned, we help our customers accelerate time to market by 20-35 months, and at the same time, we help them increase the probability of success by 30%, which means that in a given period of time, they can bring more drugs to target unmet medical needs to market.
Now, as far as the bottom line is concerned, we help our customers drive the cost down by 5%-10%, and at the same time, we help them improve quality by reducing the risk by about 25%. Importantly, we cannot achieve these results without joining forces together because, as we have shown you today, you need an integrated platform to connect people, ideas, and data. The 3DEXPERIENCE platform is the catalyst and enabler to make this happen. Thank you for your attention.
Thank you very much. Claire and Jason. Please, Glen, stay on stage. Marisa and Jason, can you join us on stage?
Ladies and gentlemen, this is Jason Raines and Marisa Co, two Medidata clients, long-term clients, people who we know very well. Actually, why don't I'll let you guys introduce yourselves, and then we'll kind of talk about our histories working together. Jason, you wanna go first?
Sure. I'm Jason Raines. I'm the Vice President of Data and Digital Technologies at Apellis Pharmaceuticals. I've been in the industry about 25 years now, and I cut my teeth as a research coordinator years and years ago at a, an obesity clinic in East Texas, and fell in love with research and worked my way into the CRO landscape and then quickly into pharma. I've been heading up functions in the data management and the data science space for, I guess, 12 or 13 years now. Been partners with Medidata for going back since 2007.
We'll get back to that.
Hi. My name is Marisa. Can you hear me? Yeah. My name is Marisa Co. I've been in this industry 34 years, and I currently run the R&D business insights and analytics at BMS. What that does is what that means is we use analytics for three purposes. One is to accelerate the asset strategy, and most of those curves that Glen de Vries showed, my team, in collaboration with the clinical scientists and biostatisticians, actually run so that we can develop the right strategy for the assets. I also run the clinical trial analytics, who actually uses that information on clinical trial design to figure out where countries and sites we should run the clinical trial.
I run the real world data team that takes all of that Glen was showing and understands the use of real world data to actually help reimbursement and value access teams as well as HEOR teams and regulatory teams to actually prove the value of our medicines to payers, regulators, healthcare physicians, and patients overall.
Thank you both for being here. Actually, Jason has worked at some extremely large companies as well, but we have an interesting contrast. We have a relatively small versus a very large pharmaceutical company. Somebody who has come to the business from the more transactional side versus Marisa from the analytics side. I hope that what you'll take away is this, we're all now in the middle in a very interesting place. Let's start with the transactional side. Jason and I met about 10 years ago. As you heard from Tarek, the origin of Medidata was doing this thing called electronic data capture, making sure that we were replacing paper in clinical trials. Maybe tell us a little bit about your Medidata experience.
When I became the head of data management at Alcon, this was back in 2007, 2008 timeframe, we were using a very old way of capturing clinical data from the sites, the investigators. This method was basically they would write this information into the medical chart, and then they would have to transcribe it onto a piece of paper as a 3-party NCR paper. This would then be sent in to the pharmaceutical companies, and it was hand-entered into clinical databases. Double data entry to make sure we've got quality and there are no mistakes between source of capture at the site to the point at which we actually had it in a clinical database, and we can actually do the analytics on it, determine if we have a safe and effective product.
This system was fraught with quality issues, and it was very slow. Electronic data capture was actually created, and I think Medidata product was a very viable solution to transform the industry back then and get us away from this legacy and slow and poor quality processes. We implemented Medidata EDC. We did some piloting at first and proved out the concept. We had some change management. I think any time you have a remarkable change in the industry like this, you have some generational type of pushback. You know, I think a lot of innovation is related to generations changing. There was some pushback that we didn't believe that this actually worked.
It's a regulated industry. How are we going to prove that things that are going into the Medidata cloud are safe and that they're not going to do something bad with the data? There was a lot of these questions at the table. Eventually, we implemented. We really did some remarkable things in terms of the speed at which we could execute clinical trials, and we could make decisions by months, if not halves of years, faster clinical trials. We were actually able to decrease the total cost of delivery of clinical trials by, in some cases, in some of our areas, up to 20%. The quality was actually improved. We would get to quality faster.
In terms of quick win, fast fail, different types of clinical trials, we could actually decide to kill a project faster and accelerate projects that we believe were actually working faster, and this ultimately benefits patients. I can attest to multiple through Alcon, Novartis, and Biogen, and now at Apellis. I have confidence that I have a partner and have a capability that enables me to help patients faster. There's no question.
Alcon, Novartis, Biogen, Apellis, you're a four-time implementer of the Medidata platform
That's correct.
I think you're definitely on the upper end of that curve, so thank you.
That's correct.
Jason and I met in a, I think, like a software evaluation originally, for Medidata. Marisa and I think, were sitting together at a medical conference because everybody thought we were crazy, and nobody wanted to talk to us.
That's right.
Mari, tell us about the history you have with Medidata.
Yeah. We will go before BMS, I worked for Amgen for a number of years and that's when our kind of paths first crossed. Back in early 2000, we were not only changing our EDC system but also implementing a very sophisticated risk-based monitoring environment. Which means when you think about how we conduct clinical trials for the FDA, all that data that we collect, all the data that the sites collect from patients and so on and so forth, back in the days, and probably still today, has to be checked that everything is complete, that there are no deviations from the actual data that is expected from a protocol.
All of that was done, by hand, by folks that actually looked at the patient charts and the data that we collect, through many ways, and actually make sure that all the data was there, was clean, to be able to do the analytics to send to the FDA. Well, today, with technology and with AI, all of that could be done in almost no time, constantly, shaving months and months and months between the end of a clinical trial and when we have the data, what we call the database lock, the data lock, so then, biostatisticians can proceed with the analysis.
As you can imagine, not doing that correctly could yield to a false submission, which for us is probably one of the worst nightmare. That's kind of the one implementation with Medidata. The second interaction that Glen and I had was almost like a geeky date kind of thing. We were in Florida at one of the Medidata conferences. When I retired from pharma the first time, one of the things that I did is I joined a group of eight volunteers that actually had the thought that we could do a virtual clinical trial, and we created Mytrus.
I run clinical trial sites, and one of my biggest passion was how do we make this very easy for patients, especially in the informed consent. Because I saw firsthand what giving a consent to a patient was like, working with the patient to try to helping them understand the 45 patients, sometimes pages, sometimes 60 pages that an informed consent entails. You know, I said to Glen, "Have you ever worked with Mytrus or heard of Mytrus, and is it possible to add the informed consent to, you know, the Medidata platform and continue to, you know, create a seamless platform within all the steps of the clinical trials? I'm thrilled to see that part of Medidata.
Very recently, we finished a project together where, we used the Medidata, the data housed in the systems as well as the sophistication of the analytics team, the Acorn AI team, to actually save, a clinical trial that was, that ran astray, that for those investors, that are here, represented, you know, significant value to the top line. We could not have done it had it not been for the Medidata platform and the Acorn AI team, so.
Thank you. But to connect the dots, that patient platform, the CEO of Mytrus, which we wound up acquiring, was the person, Anthony Costello, our SVP of patient, who made the announcement about our patient platform yesterday. That was a great conversation we had. I can ask both of you, but I'll start based on what you just said, Marisa. There's the processes around running a clinical trial today. There's the extension of platforms with things like e-consent and thinking about the patient. There's this idea of using data to do a better job of, or when we have to rescue a clinical trial. What is, what is your vision, let's keep it to clinical trials for the moment, of a platform of the future?
What does the future look like in your perfect world?
Well, to me, and, you know, I keep it I think I would extend it a little bit beyond clinical trials because folks in life science, including the way in which most large pharmaceutical companies work, is with the traditional approach of drug development is linear. Drug development is anything but linear. What I see the potential for, the, the clinical trial of the future, if you wish, is you really need to start in the discovery space, in the translational space.
Because most of where we select the patients, we stratify the patients, we decide what might be the right drug for the right patient comes really from an ecosystem that starts with creating the drug, taking the drug to the clinical process, then into the market, and then identifying that for some patients, that drug doesn't work. For the patients that the drug does work, you know, we are thrilled. More often than not, especially in oncology, for 75 of the patients, that drug doesn't work. We take that information, and we have to go back to the discovery, back in the translational medicine, and back into what are the patient characteristics, what is the phenotypic and genotypic characteristics that may uncover why the drug doesn't work.
That really is going back to the early stages of translational medicine and discovery to figure out, you know, is there anything else that we get a change in the molecular structure, a change in a biomarker, a change or a new biomarker that we can pursue. That starts the process all over again with clinical trials and so on and so forth. The idea of a truly connected platform that allows me to understand, right, at the end of the day, what is that particular patient population? How do I identify it? How do we put it in a clinical trial? How do we actually identify where those patients are? Here we absolutely need the help of the providers who actually have the patients.
The idea of linking to the real world data and having a direct access to those patients, to those physicians who have the patients, so we can alert them, that a drug or a clinical trial might be beneficial to the patient and actually them, you know, enroll that patient in a clinical trial without necessarily having to jump through hoops, for, you know, informed consent, patient identification, the site survey, and a whole host of things that today with technology could be done, you know, seamlessly. The pharmaceutical companies still are, you know, working in, some of them, in the 1950s.
What I love about what Medidata is doing is forcing all of us to actually run faster because they're 10 years ahead of any pharmaceutical company in terms of how they're thinking about drug development, drug discovery and development, and commercialization in a seamless, unison way versus in the fractured way that we think about it.
The fractured way, we worked together, Jason and I, on when it was really a best of breed idea of let's take as a life sciences company, all these different systems and tie together. That's how we'll be successful. We kind of went into the platform. What are your thoughts about the future of trials?
Well, I think that was extremely well said. I think that, you know, when I think about the future of clinical development, it's going to be the accumulation of a lot of different disparate data assets that are coming from a lot of different places. In some cases, I think, you know, when we think about best of breed and we think about a holistic suite of products, I think that you're on the right path to actually provide a platform by which we can integrate all these different data assets and hit it with advanced analytics to find the needle in the haystack. Because precision medicine is truly about a patient's response that is really unique to them only.
Historically, it's all been about, you know, a brute force kind of approach, and you test a lot of people and you have, as an average, maybe not a significant siG&Al to say that I have a safe and effective drug. If I hit the right patients, this thing can work amazingly well. When I think about the engine that has to gather all that information, 10 years ago, there was sort of a best of breed approach because the maturity of the technology and the technology companies wasn't really there. Perhaps I had to pick this one, and I had to pick this one and pick this one and make them work. Today, there's not a lot of companies that actually provide this holistic suite of products that actually enables clinical development.
Medidata, in my opinion, is the leading company that can partner with pharma and biotech to actually deliver all of these data assets in a way that actually accelerates clinical development and precision medicine. I can defend that by all of my experiences with other companies and the processes by which I actually have to implement in order to get this to work. The questions often are of to me that, Jason, how were you able to get a lot of these metrics that you were able to obtain with Medidata? I answer it usually with an analogy of cars or racing. I wanna pick the best car, but everyone else can maybe pick the best car to actually get the fastest record, if you will, or to win the race.
Medidata is the car, but I actually need professional services. I need people that know how to drive that car and can teach me how to drive that car in a way that makes me win, right? Winning for me is faster and precision medicine is that are getting this to patients. I also have to have a humility and be open to thinking differently about this as well. This is what I have gotten from Medidata from the years, Glen. I can come to you. I've come to you dozens of times, and I'll say, "Look, I have this problem." "I'll give you my opinion," is what you say, "but I know who you need to talk to." This is about a partnership, too.
It's the technical solution, which I believe is a better car than what's else out there in terms of the competition, because it is a holistic integrated suite from consent to lock, post-lock, management of the data in an efficient way, in a compliant way, but also get an enormous amount of confidence and experience within Medidata that I can leverage to help me drive the car better. What happens in the future is just expanding, I think, that capability and integrating data in a more seamless way from electronic medical records, claims, digital. The digital footprint is gonna continue to go up. I think patients are gonna be empowered with digital. They're gonna be measuring their own symptoms. There's non-invasive hemoglobin apps. There's all sorts of this is exploding.
In a clinical setting, this needs to be connected through your eCOA solutions, etc . You have the capabilities to really actually push us more in that direction. With the data sciences and advanced analytics, we can find the patients that need the therapy, so.
You were on a panel yesterday, Marisa, on data science analytics, and you were talking about success begetting success.
Yeah.
Is this a jumping off point where we could discuss that?
Yeah, sure. Again, what I told you yesterday, which is, you know, you're forcing most of us to think differently, to think more now in, you know, in a more integrated way, which is very difficult in an industry that has run in the same way for years and years and years and is fragmented by nature. The concept of, you know, at BMS three years ago, we decided to consolidate analytics for commercial, for R&D, for GPS, for global production and supply. You know, one of the toughest thing was to really bring all of these disciplines together, like real-world data and clinical trial analytics and asset strategy, and start making sense of how do we deploy analytics to make really faster decisions.
You know, there was a lot of resistance because that's not the way we think. That's, you know, typically in R&D, we're extremely good at understanding the science behind the data, but not necessarily in applying kind of the business concept, using the science to apply or apply to business concepts and make decisions. I think the early successes that we had, like, you know, accelerating a clinical trial, greatly accelerating a clinical trial with data or using real-world data to actually help a payer understand that our drug, although it appears a little bit more expensive, it actually causes a lot less adverse events and a lot less hospitalizations that are very costly.
All of that data and insights that we've generated, when to, whether it be to support the better reimbursement or regulatory decisions, whether to accelerate a clinical trial or, you know, using analytics to actually bypass the competition, you know, in a, in a trial design and filing early. All of those are now expected. You know, the idea of partnering with, you know, a company like Medidata, where all that wealth of information, apply, with, or combine with a very sophisticated analytics team and Acorn AI, it only can make us even more successful.
Part of the success was when we started three years ago, my team was about 50, 55 folks. Then, you know, after that, with the integration of Celgene, we will be about 160 people doing analytics for R&D.
A question about the project that we were talking about in terms of rescuing a study. Not specifically that project, but it was one where we looked at data from outside the walls of your company.
Yeah. Completely.
is that a trend that Well, clearly, we believe is gonna continue, but how do you think about that?
We have to. I mean, we do not have enough information to actually make the decisions that we make. That's first of all. The second is for us to make certain determinations and certain and build models that actually are trustworthy to regulators or payers and so on and so forth, we need millions and millions and millions of data points. That will not come from the four walls of a pharmaceutical company. We're constantly partnering with others who have data.
One of the things that attracts me the most, there's so many things that attract me about what Dassault and Medidata are doing, but the idea of using, utilizing and deploying real-world data in certain analysis has allowed us to identify those patient population for whom the drugs are not beneficial. You know, we try to build models for, you know, how do we stratify those patients even more to understand the molecular dynamics of those patient of that patient cohort. The reality is, in order for those for those models to actually have any validity in clinical practice, or even for a regulator, we need to validate those.
The, you know, typically, the most trustworthy way to validate a real-world data model is with clinical data, because that's kind of the standard. The only company that I know of that has the ability to kind of bridge those worlds is Medidata. I'm, you know, thrilled on that standpoint. On the other standpoint, because I have been a Medidata fan for years, you know, talking, listening to Bernard today and what Dassault bring to the table, one of the most difficult things in drug development that we haven't quite figured it out yet, is our manufacturing folks need to figure out what how much production they need very, very early, even two years before we even know the dose of the drugs that we're developing.
Why is it important? Well, as the former head of finance R&D, I know that how much it costs to actually build a manufacturing facility or, you know, create production plan if you're wrong in certain assumptions about dosing, you might be spending, you know, $ 5 billion for, you know, sometimes you need like half of that. Really early in the drug development in translational medicine, we do clinical studies to actually find what is the appropriate dose to use with the patient. By that time, it's way too late to tell clinical manufacturing how much we need.
Using the data that Dassault has as it relates to manufacturing production with the data that Medidata has as it relates to all the trials, the early trials for dose finding, today, we could potentially simulate how much drug might we need, and, you know, help the our clinical manufacturing folks actually, you know, estimate what kind of lot and what kind of manufacturing infrastructure you're going to need.
Okay. As we now supersize our view, I think in a good way in terms of the systems and the data and the connectivity, if you wanna take a shot at what the structural changes of a life sciences company will be. You know, we used to have our little departments for data management and clin ops. What do you think the company that you're at or the companies that you'll be at in the future are gonna look like and how that will be different than they are today, Jason?
That's a good question. I think the evolution, I think, will require different competencies. There will be a different talent, I think, pool within the organization. I think there'll be talent that will be ingrained and competent within data sciences for sure. To the use case that was just mentioned regarding the ability to actually look at what is needed from a production manufacturing perspective and being able to take all that information and leverage that in such a way that you can create these simulations and predictions. We need support technically for that and just to be able to understand it, I think. I think that's one thing.
The other is, from an operational aspect, I think that you can actually federate and decentralize to some degree some of these things, which actually may help out because the data can be, what's the word? democratized in such a way that you can share it freely and in a confident way. and empower organizations to have access to it so that they were more informed about what's going on, either in front of them or behind them. This creates a culture of accountability within the organization, which I think is gonna be important. I think those types of, operational changes will come naturally as the technology and these, data science competencies start to evolve. The problem statements will shift, to some degree as a result of that.
I remember lots of times that we actually introduced new KPIs into projects that we were working on together, and that the introduction of that visibility and transparency was what changed some of the behaviors, right?
Exactly. I think, that, you know, we quote a bunch of different authors around this, but when you measure it, people change their behaviors based on just the fact that it's being measured. One of the benefits of Medidata, I think, is not only the scientific data that's being managed, but also the operational data that's being exposed related to site performance, the quality of the data that's coming out of the system. You change your behaviors, and you become very competitive. If you actually have the industry, if you have over 50% of the clinical trials for the last 10 years, I can actually benchmark my performance relative to my competition or others. Just exposing that to the leadership of the company, we're naturally competitive. We wanna be better.
Right.
That alone, I think, helps, even with the sites. If you expose it to the sites and the investigator sees that he's the slowest or the underperforming relative to his peers, his or her peers, then their performance naturally changes.
Actually, I'm curious about both of your opinions about this, that a patient is equally part of that equation. One of the things, and again, it was what part of your motivation around Mytrus and informed consent, putting things in front of the patient will actually make the patient behave differently. That could have real therapeutic impact on the outcomes of the therapies you're developing, right?
Without a doubt. To me, back to your question, I don't think we will see massive changes in, you know, organizational structure. I think companies have, you know, structures just to make sense of, you know, the chaos of drug development, right? One of the things, at least, you know, at Bristol Myers Squibb, one of the things that we have noticed is, you know, through the our digital health effort, which was a company-wide effort where leaders from many functions actually came together to figure out what are the questions that we need to resolve, and how do we go about resolving them.
Then out of that came the operating model that now allows a translational medicine person with the biostatistician, with the regulatory person, with the analytics person, actually to work together to solve those problems. That mindset of there is no single group that can actually solve the problem. You know, you need, you know, you need a village to raise a child. I think you need a village to raise one of our challenge, our child, or children who that is a drug that we need to get to market. To me, A, the talent that we're bringing into the company is one.
They have a very different approach to how they work and an expectation that they will be part of an ecosystem, not part of an organizational structure chain.
More of a collaborative change.
Yeah.
A difference in the way, actually, Jason was talking about before, people communicate. When you're doing science at certain scales and with certain models, you need to have the communications integrated into the actual innovative platform.
Indeed.
Yeah. A lot of what Medidata has done over the years, has come from our clients, not from us. Actually, Jason and Marisa have, as you've already heard, been involved in many of those conversations. I like a little pressure. I didn't tell them I was gonna ask them this question. Now you've seen a bunch of things. Do you have any requests for what you'd like us to work on next at Medidata?
Well, how many hours do we have?
I have the rest of my life.
There you go. I'll take you up on the offer.
Five minutes? We have five minutes now, but we'll keep working on it.
Well, to me is, again, in the spirit of precision medicine. Precision medicine requires not only data and analytics and the physicians, but precision medicine requires that we convince the physician that an algorithm could be the solution to a diagnosis and a treatment. For that, to convince the physician or to convince the regulators that is the case, we're going to need to validate all the model and all the simulation, all the models that we put forth as a treatment pattern.
One of the things that I was talking to Glen about it is-In order to do that, we need to now bring the power of real world data, the power of clinical data, combined with our kind of regulatory experts and our biostatisticians, and start really developing tools that will be deployed to physicians to ascertain and to decide which patient requires which drug with a lot of degree of confidence that that and validation that that is actually the right treatment pattern. Because of two things. One is the patient privacy laws will suggest that there's a clause that says the right to an explanation.
Physicians that don't understand the black box of any AI model and so on and so forth, will never use an AI-driven decision to actually tell the patient what or even what treatment they need to pursue. I think for me, if we want to make precision medicine in reality, especially in therapeutics, I think we're going to have to work together across the aisle between the real world data space and the clinical space to actually validate those algorithms and convince the regulators that we could do, you know, we can stratify patients and decide treatment algorithms through AI.
Right. Patrick and I are on that one. Jason?
To that point, I think the FDA and other agencies are really thinking a lot about the black box and the AI. They're creating guidances and they're modernizing their thinking about how to actually create software as a medical device and an algorithm as a treatment recommender, if you will, or a therapeutic recommender. I think one of the gaps that I would like to see Medidata also work on would be, I think one of the issues we have with patients is that they don't really control their medical record. How can we technically solve that problem as an industry would be something I would love to see you guys work with.
You have the clinical data, you have access to real world data through claims and pharmacy and different mechanisms of extracting this from the insurers and, you know, EMRs, et cetera. The patient, if they wanna really subscribe to research, how do they get their retrospective pool? How do they get their genome and all their phenotype and all the treatments and the polypharma that probably exists that actually can help with stratification, can help with actually marrying that up with the clinical studies to do this profiling that's really needed to decide which patients are gonna be more susceptible to a certain treatment or an adverse event pattern.
That, to me, might be something that you could actually supplement or augment your suite of products with that enables a patient the opportunity to just go in and say, "Look, I want all my information to be housed in my medical record. I want to participate in clinical trials. I have everything under my control and in one environment." Just imagine the power of that if it gets big to pharma and biotech to actually exploit.
Yeah.
That, to me, is something that would be exciting to think about.
It's actually kind of a interesting set of things together. You've got the black box. Even giving people, the patient, the visibility of what the inputs to that black box are is an improvement of what we have today, and probably a prerequisite to getting people to understand, trust, and have that validation.
Yeah. If I can have one more wish when I close my eyes and dream, now that you guys partnered with a European company, right? As much as we have availability of data in the U.S., we have very, very little as it relates to, you know, access to patients in Europe. I was talking to the European Commissioner not long ago, and she said to me, "You know, Marisa, we have shown that we can reduce by 50% the time to disease diagnosis, you know, from 5.2 to 2.5 years, just by using data. Yet only 9% of our patients, you know, have access to EMRs.
most of them, limited access to EMR. to me, that there's a lot of momentum in Europe to drive digital Europe, to drive the standardization and interoperability of platforms. I think what they need is really somebody who understands how to do it. there is a lot of momentum and there is a lot of investments in this area by the European commissioner. you know, we're talking about EUR 5 billion, which is not a minor undertaking. I think we need solutions for our European counterparts.
All right. It's on the list. Jason, Marisa, thank you so much. You guys are very busy. It means a lot to us that you take the time. Hopefully, you all think this is worthy and join me in thanking them.
Thank you.
Thanks a lot, Marisa and Jason. Pascal and Rouven, may I ask you to go on the stage. You will have a Q&A session, after Pascal and Rouven presentation.
Okay. Good morning to all of you. Good afternoon for the one attending the sessions through the webcast. It's my pleasure to conclude this session, and I hope you saw many examples, concrete illustrations of what I shared with you almost since June, since the announcement of the acquisitions. Before, I'm gonna make a quick disclaimer, not the forward-looking statement disclaimer, because, you know, it's due to my articulation. Please do not tell me at the end of my session that it's almost the same as usual. I'm not sure I will take it positively. What I want to do today is really to focus on, almost on the three questions you raised since we announced the acquisition of Medidata.
Does not mean that what you have seen today will not contribute for the growth for the next 10- 20 years, but I want to focus on the year 2018- 2023 plan. That's the purpose of my presentations. The three question I want to address are the following. The first one is, Why you made such a big investment in the healthcare? I remember doing the roadshow at the time of the bonds, and many of you asked these questions, and I will give you some answer. The second questions is, It's a domain, per se, with a lot of players, and we are not sure we understand the competitive landscape and how you are planning to differentiate yourself.
I hope you have seen some concrete things today, but I will come back on this. The last point will be, finally, what will be the contribution of Medidata to the year 2018- 2 023 plan. 2023, sorry. For that, I will share the presentation with Rouven Bergmann, because Rouven is a former CFO of Medidata, but now he's the COO. He's a guy being accountable to make the post-merger plan a reality and a success. That's the reason why I want him to be on stage with me, because it's a way for him to be committed. I'm not be the only one. Okay? This is for the introductions. Let's start for the first questions. Why are we investing so much money in healthcare? You know, we are long-term company. Being a long-term means many things.
It means that we have to prepare the growth at least 10 years in advance. The proof of that is not only we are sharing the 20 years visions with you, and this is what Bernard did this morning, we are committing our plan on a 5 years periods, but we have also to think this way. I took the framework, the purpose, you have seen it, Bernard presented it. I put the percentage of the GDP, and you see that the traditional manufacturing industry represent 20%. The, what we call infrastructure and territory is close to 50%. To a certain extent, you could ask me the questions, why you are not reinforcing the position you have in the manufacturing industry, or why you are not trying to expand your footprint in infrastructure and territories. The answer is here.
If you look at the growth, because at the end, the only purpose is to fulfill the growth, the organic growth on the long term. The GDP growth for the healthcare is twice than the other. At least this is what we are expecting for the next 20 years. The reason to believe are the following. There are two. You know, today, only half of the population have access to the health services. Only half. 7 billion people, only 3.5. It took, for many countries, more than 40 years for them to be able to cover 100% of the populations. You know, this is what happened mainly in Europe and in the U.S. More recently, you have country like Spain or others, Korea, it took 20 years.
If you look at the last trend, China, for example, it's only 10 years for them to have access to a coverage of 100%. And all the nations, they took the commitment that in year 2013, 100% of the worldwide population will get access to the basic healthcare services. This is, at a high level, a big trend. The second trend is the following. There is a strong correlation between the life expectancy at birth with the spending. You'll notice on this graph, you have a few bubbles. On average, you know, it's around $4,000 per capita. The life expectancy at birth is exceeding 80 years old. You have an exception, the U.S. They spend almost twice, and the life expectancy is a little bit lower.
It means that health is not only the healthcare and how you treat it, but it's also how you behave, how you take care of yourself. I do not want you to restrict the definition of healthcare to only the pharma and the med device sector. It's much more broader than that. We are engaged to bring also this on the long term. You have a bunch of countries where usually the coverage is not at 100% of the populations, and we see this acceleration trends. If you combine those two things, we had good reasons to believe that the GDP growth for the next 20 years will be twice the traditional manufacturing sectors of the infrastructure and territories. That's the basic reasons why we took these decisions now to invest to fulfill the organic growth for the next 20 years.
There is a but. Because there is no way you can, you know, give access to the health services to the worldwide populations the same way we did for half of the populations. The economic equations need to change. I think Tarek highlighted in a very good manner in his presentation when he told you that there is a new innovation cycle starting. I think this is a reason why we have the legitimacy to be in this space. Because you could also question, "Why, guys? What would be your value added to come to this?" We are convinced that the innovation cycle in this industry is science-based. Everything we do since day one is based on the fact that we put science into an IT system. That's the reasons.
The second reason is because, you know, you have seen this number. There is a lot of inefficiency. You know, this industry cannot continue to spend as much, and the health services has to be affordable for the worldwide populations. The economic equations needs to change. This is where we play a role with this. Again, it's everything you have seen it. I'm just summarizing just for you to be sure you have the story. What we are bringing is unique because not only it's a way to cover all the different discipline to foster the innovation cycle, and you have seen it in actions with Jason, Glen, as well as Claire. You also have listened carefully the testimony of the two customers.
It's because, you know, as Bernard said, we have a different way to promote this value. The first one is through the solutions we are maximizing the outcome, and the outcome is at stake for this industry. The second thing is, if you want this to be, you know, master with a high level of quality, with a lot of efficiency, with reducing the time cycle, sorry, you need to also foster the collaborations between all the different disciplines. We have defined the processes to make it happen. Last but not least, with all the roles we have, you know, to cover all the different disciplines, we know how to equip all the people being involved in this innovation cycle. Everything is relying on one single platform.
Again, remember, this platform is unique because it's the only way, it's the only platform being able to manage the knowledge and the know-how in a collaborative manner, combining the two approach, one which is the data science and the other one which is the power of the imagination through, you know, modeling and simulations. That's the core of what we do. If you look at the landscape, and I briefly discuss it through the Q&A, it's highly fragmented. None of the competitors, you know, mentioned in these slides could do what we are doing. Let take, I mean, the several categories. If you look at the highly specialized one, you know, especially the guy being upstream, they have the domain expertise, but they do not have the platform, they do not have the go-to-market, they do not have the full industry expertise.
To a certain extent, they are very good point solutions. As many customers say during this event, you know, their knowledge and know-how is stuck in one single systems, and there is no way you can flow this across the barrier of the organizations. You take the big one, you know, the one you know because they cover many, many different industries. They have an IT platform, but frankly speaking, they do not have industry expertise. This is probably the reason why Medidata won many market share against one of the big names mentioned into it. It's also true for, you know, the manufacturing part. You know, this industry is science-based, and if you don't know how to mix the science with the industry expertise, you are not in the game. Last but not least, you have a newcomer.
You heard the name, twice at least. You know, if you look at the reality, they are doing two-thirds of their revenue with CRM systems, sales force automations. This is where the revenue is coming from. The revenue is not coming from the product life cycle or the drug life cycle management, even if they claim that they have an ambition. If we zoom, they only have a point, which is quality and compliance management. Through the demo you have seen from Jason's, I hope you have understood through this demo that the game is over to be document-based. Because we have to be data-driven and generate automatically all the reports for the regulators. That's the way we are planning to do the change. That's the way we want to be game changer. My point is, no one in town can replicate what we have.
If you sum up the industry expertise, the platform, the knowledge, and the domain expertise, and this is combined, you know, with two teams willing to be together. Tarek explained it very clearly that the culture is the same, the DNA is the same, and the will to transform this industry is here. The third question is related to the contribution of our five years plan. The first point is this one. You remember we committed to double the addressable market over the time, going from $16 billion to $32 billion with 3DEXPERIENCE platform. We announced it in year 2012. Where are we? Today you will see. There is animation on the slides. It's a $38 billion addressable markets, of which $8 billion is coming from the healthcare.
Against this market sizing, it's a pure software revenue. It's the compilation, the addition of all the existing revenue coming from all the point solutions you have seen previously. This does not take into account that you still have homegrown systems running in many, many pharmaceutical company. My point is this quantifications of the market is probably the minimum quantifications. The second message is the growth expected is 10% growth on an annual basis. The main drivers for now is on the slides, is the growing pipelines in terms of drugs. Today, it's more than 16,000 new drugs in the pipelines. It's almost twice compared to year 2001. You see an inflection point, because in year 2013, the trend changed dramatically. Which is a proof that there is a new innovation cycles going on.
this is a proof also that the complexity of developing a new drugs is higher because you need more options in your pipeline to make it happen. That's what is driving the market now. Knowing that my belief, and I think it's also the belief of the Medidata teams, the coverage of this market is not well done. You know, you only have few companies being served, and the penetration of all the solutions is quite limited. It's almost a third. If you combine all those levers, you know, I'm pretty convinced that the 10% growth will be sustainable over the next 10 years. Now, how do we split this market? Again, it's for you, just to have a sense.
You have the three big existing markets, the lab, ONE Lab, and you have seen it with Jason's demonstrations, the clinical trials, and also the production side with what we call Made to Cure. It does not mean that the others are small. It means that the others are not well deployed. That's the point. Okay, that's for the market growth. Now from a market share standpoint, where are we? With a combination of Medidata and Dassault Systèmes, we are number one with 9% market share, almost twice than Oracle and SAP, and slightly ahead compared to Veeva. I just want to remind you that in Veeva's numbers, two-thirds of the revenue is coming from the CRM. It's not coming from the market we want to tackle. That's very important for you to understand.
The second message is if you assume that the market growth will be at 10%, and we will continue to gain at least half a point in term of market share per year, we will be able to continue to grow at more than 15%. That's the point. Now, how this combination will work effectively? Here is a way we want to do it. Medidata is becoming a business unit as part of Dassault Systèmes, and you have seen the management team with Tarek, Glen, and Rouven. They are the leaders, and they will continue to run this business and continue to grow this business, leveraging the rest of what we do. The life science industry is becoming the second-largest industry for us, slightly behind the auto sectors, but bigger than the aerospace sectors.
I remember you, pushing me, punching me, I should say. It's not because I have stitches in my mouth, but punching me about that we had too much dependency on the auto sectors. To a certain extent, you have an answer now that we are balancing the revenue and the footprint across multiple industries. Last but not least, we will combine the go-to market of BIOVIA with Medidata because there are so many companies we are not serving. You know, I gave already the numbers, but if you think about it, we are touching only a third of the market. We are touching a third with almost a third of the portfolio we have. That's the point. If we structure the go-to market, and it will be a dedicated one, it does not mean we want to leverage the rest of what we do.
It has to be dedicated because coming back to the question we had during the Q&A, it's a different way to engage. It's a different way because, you know, they are at a different stage of maturity. We need to be able to start in a different manner compared to other industries. That's the reason. The second reason is because you need to have a high degree of industry expertise across the field ops, operations. It's not easy to build such a team. I think if you combine the two teams, we have by far the largest field operations to serve this market in a proper way. In addition, you know, Medidata and I and Dassault Systèmes, we started to build a complete ecosystem. You have seen during Tarek's presentations, the CROs, companies like Parexel or IQVIA.
I remember some of you during the roadshow telling me, "But they are your competitors." I said, "No, they are our partners." Why so? Because you know that we have an indirect model for the traditional manufacturing industries. SOLIDWORKS is promote and market through an indirect model. We could do the same. the valves are not the valves. They are a different nature of partner, and the CROs is a good example. They could become our partner to at least expand the reach of this market and serve them in a proper way with, again, a high level degree of expertise and maybe some new business model because we could be much more linked to the outcome, what will be produced with our software.
Rouven, I think it's the time for you to come because I think you have a few things to say to the crowd.
Thank you. Thanks, Pascal. Yes, Pascal, as you already introduced, I think what's top of mind for you is to understand how does our growth plan going to look like, for the next five years. Before I go through the components, what I want to start with first, we think that this is a very attractive plan for investors. Secondly, we believe and we are very confident that this is achievable for us. The 13%-15% is what we are targeting, now for the next five years, every year. There are a few things. I mean, This is a very simplistic summary, but I think it is a good framework for you to think about how the, what the different levels and vectors of growth are going to be for us.
The first one, our core business. You heard a lot from Glen earlier today in the product presentations and our really strong presence in electronic data capture and how we serve the life sciences industry and operate clinical trials. At the same point in time, Pascal just walked you through the overall market economics, right? It's a very strong growing market. We have a very good track record of taking market share from the competition. Pascal alluded to this. We feel very confident that we continue over the next five years to take market share from the competition, number one. Number two, that there's continuous growth in the market. There will be more clinical trials started because that's how the industry, as Marisa said, that's how they're changing. There will be more precision medicine, more trials.
This will be more opportunities for us to monetize. The third point, that's why I was so particular earlier in the morning in my presentation, was around our pricing model. Because the way we've defined our pricing model, we are capturing at the level of the clinical trial metric, the pricing. The way we are not contracting with our customers that don't think that they have an all you can eat type arrangement, they don't have that. When they start more trials, we will be able to charge more. That will be the reason why we can monetize this growth with our core capability. We believe that six points of the 13%-15% over time will come from our core product offering. That's also what we have shown over time.
We have a good track record of at least delivering these growth numbers. Again, you saw the capabilities that we have built through continuous innovation. We serve really from the enrollment of patients all the way to the submission of data to the authorities. We cover all aspects of the platform. We have a lot of products and capabilities to attach to increase our share of wallet, and that adds another four points, we believe, conservatively. If we put that together for our core competencies, that's about 10 points of growth that we will add over the years by continue to do what we do. I will also mention one additional point.
I think Pascal alluded quickly to this as well as Marisa mentioned the difference between the U.S. and the European market. When you look at Medidata, we are very focused on the U.S. market. We have about 75% of our revenue is U.S.-based. Now, we will be part of a European headquartered company, and I think our presence in the market will be much stronger, and that gives us better access to clients and will also strengthen our core business. I think these are all arguments to believe that the 10 points are realistic. For data and analytics, you see this is really one of the big opportunities that we have to help transform the industry. Conservatively, we are assuming this to have very marginal contribution to our plan.
Of course, we are much broader in the way we pursue the opportunity. I leave it there. Professional services. We talked a lot about the importance of it. They are an enabler for transformation. The domain knowledge is very critical, and it's differentiating us in the market, so that will continue to add growth to our company as we grow. This is also nothing changing compared to what we've been doing. I think we also want to make sure that, you know, as a combined company, we have to really be thoughtful around how we address our synergies in the market and how we build the framework to go after a much much bigger opportunity. As we heard, our customers need to transform, and we can only do this together.
Bringing these capabilities together should at least add two points of growth over time. You see, we don't consider this to be immediate, but over time, to contribute marginally to our growth. Of course, with the goal to be much bolder and drive more growth from that. That's the top-line picture. Now let me transition to the bottom-line picture. Here, also the same time frame, the next five years, our plan is to achieve 27% profitability. That is about 2 percentage points increase year-over-year or 200 basis points year-over-year increase. We believe that we have a real opportunity to improve our margins, while at the same point in time, we continue to invest in growth and in innovation. We have done this as a standalone company.
We always had about 80- 100 basis points organic margin improvements year-over-year, so we have a real track record of driving operating leverage in our core business, and that's one thing that we will continue to do as well. You see this here, operating leverage, we estimate that to continue with about a contribution of six points to this plan. That's about half of the 200 basis points per year that we will achieve, a little bit more. This is, of course, a framework and, I think maybe conceptually to add to this, there are multiple ways for us to achieve these numbers, of course, right? There's not just one way. There are multiple levers that I'm introducing here.
Of course, now we have an opportunity to really, really go after synergies, and accelerate that path to increasing profitability by combining the organization. For us, when we think about a number of areas in R&D, we've been becoming, you know, as a cloud company, more and more exposed, to some of the increasing cost trends, that, infrastructure-as-a-service company like, companies like AWS has introduced to us. We now have an opportunity to reverse this trend by leveraging Dassault Systèmes' OUTSCALE capabilities over time and really become more efficient in providing our services, cloud power to our customers everywhere in the world, right? That's so important because that's also where the regulation is changing. We have to be closer to our customers.
This would otherwise to be, as a standalone company, massive investments we would have to take. resources and engineering resources. As we are continuing to grow our portfolio, we have a real opportunity to optimize our resource mix and location mix. This is something that would have been an opportunity as a standalone company we would have to do. Now we are having an infrastructure that we can work with and can tap into, so that's, gives us an advantage. These are just two very high-level points. G&A synergies, as every one of you would expect, are going to contribute to that. I give you a couple of sound bites. We will have much higher buying power in the future.
We are going to be working with an organization that already has a large presence across the globe, right? We have 16 offices today. I think Dassault Systèmes has 180 offices. We don't need to invest into real estate, right? We are going to be part of Dassault Systèmes, who's already present, and we can mitigate a lot of those, reduce a lot of those investments in the G&A area. Yeah, that's the high-level outline. I hand it back over, back to Pascal, thank you.
Thank you, Rouven. Stay with me because the last slide is the result of what I just explained, is the contributions to the EUR 6 EPS target for 2023. You remember we sliced the EUR 6 EPS in different pieces, the one coming from the adoption of the 3DEXPERIENCE platform by the largest surveys we have. Second one is the continuum of the industry diversifications. We had this KPI, or I would say target objectives, to have EUR 0.80 coming from acquisitions and new business model.
Among the EUR 0.80, EUR 0.70 will come from, you know, what Rouven just presented to you, assuming the plan, which is almost doubling the revenue of Medidata in the next five years and gaining almost 10 points EBIT margin over five years. The rest of the assumptions stay the same. I mean, I didn't change the currency exchange rate. The tax rate is the same. When you're gonna do your model, you can basically rely on the same assumptions. There is only one, tow things. You remember now we have debt and the financial interest and contribution will not be the same. Clearly, we expect to have a deleverage over the investment cycle. To come back to 1x EBITDA, net debt over EBITDA ratio in the time of the investment cycle.
Today, after the transaction, we had EUR 2.5. Last but not least, you know, there is the EBIT margin of Dassault Systèmes, including Medidata, will not be below 30%. You have my commitment that we will stay at 30% plus. I think what Rouven shared with you, it's not only a realistic plan, but we know how to do it. We have been able to demonstrate on both sides that when it's time to put in place the right set of actions, we know how to do it. That's it. I think it's time now to take some questions. To do so, maybe I will ask all the different presenters to come on stage for the second part of the session.
Thank you. Thank you, Jay Vleeschhouwer , Griffin Securities. Two questions. First, I'd like to return to the subject of integration, specifically product or technical integration. We saw some examples earlier of the life sciences industry solutions you've already introduced, by DS. Those were the culmination of what I think were one of your most important internal initiatives for cross-brand integrations across DELMIA, SIMULIA, BIOVIA. Those did take time. The question is, how are you thinking for the intermediate term of the initial batch of co-packaging, co-integration between Medidata and what we just saw on the life sciences side from DS? Who, in fact, would manage that? Is that Rouven or who's gonna manage that process? Secondly, we've heard the term outcomes quite a bit today, as in patient outcomes, for example.
What is the potential for employing outcomes-based pricing model and/or marketplaces to any or all of life sciences development, commercialization, and/or even patient therapeutics?
Rouven, you take the first one.
Sure. Oh, I need the mic. Thank you. Actually, as you heard from Marisa earlier, a lot of the projects that we do that are innovative, they're using use cases with clients. I expect that there will be projects where we will collaborate with the other teams at Dassault Systèmes in terms of delivering the first iteration of certain use cases. Obviously, it's gonna be use case dependent, but that's where we'll bring the elements of data together, the elements of the ontologies together. I don't know if I'm gonna scratch the itch on, well, you know, what day is something gonna be done, but that's 'cause we're in the process of identifying those exact use cases and those first clients. That will proliferate into more and more platform integration.
Maybe I could add a few things. From a timing standpoint, the first phase is, again, as Rouven say, is to leverage the cloud capability we have. The number one priority for 2020 is to substitute progressively, you know, the cloud infrastructure with 3DS OUTSCALE, point number one. Point number two, we have two platform. The two platform needs to be connected. We have a strategy, so-called POWER'BY, to make it happen. We will develop the POWER'BY strategy in 2020. It will be probably on the market in 2021. We will start to develop, as Claire and Glen and Jason demonstrated, some, I would say, combined products. Again, we will start the development in 2020. You notice that in the revenue synergies, Rouven was highlighting, the most of the impact will be 2022, 2023.
This is, in term of sequence, what we are planning to do. We expect to have the completeness, I would say, before five years, right, my friend?
Yes.
The second question related to the outcomes may be for you, Bernard?
Yes. Yeah, happy to answer. Customers are, at times, quite interested in risk-based contracting, one kind of outcomes-based contracting. What we've seen typically happen in the past is that when we go down that path, as they start to think about what the upside or the downside for them, but the upside for us is they come back to a much more traditional contracting model. The one space where I see that there's a divergence early days is with Acorn. We see that some of the smaller biotechs who are so tied to outcomes, and there's a lot of potential variability. They wanna either get to the finish of a study so that they can submit to the FDA or have a monetization event, maybe more funding or sell the company. They're willing to put more of the contract at risk.
Quite frankly, we do have an interest in doing that. It's going to be, I would say, over the next five years, a relatively small part of our overall strategy. We're certainly open to it, there's not, you know, that much industry momentum behind it.
Claire, you maybe want to add a few things?
Yes, sure. In the past, I've been involved with pricing and reimbursement at the Ministry of Health in France, and I've seen, you know, a growing interest in outcome-based payments. What I want to say is even if we remain in a, you know, a traditional pricing model, we can, there is already a lot we can do to help life sciences companies moving towards these outcome-based contracts because they need to be able to demonstrate outcome to payers. All the data science that we do are going to help them do that. If you think about the 3DEXPERIENCE platform, there is really two sides of the coin, right? There is the platform as a system of operations, and we could already help our life sciences customers move towards outcome-based pricing.
now we should think about the 3DEXPERIENCE platform. As a business model, there is much more we can do to connect the extended value network. Here, you might want to think about more innovative ways to price outcome-based payments being one of these ways.
Thank you. James from Barclays. Market estimates weren't out there for Medidata all the way to 2023, but there was already some pretty bullish expectations for 2021. If we extrapolate those, clearly reconciles with your helpful presentations around what you're expecting for the business. Notwithstanding your comments around the product roadmap and how that comes together later, given the compelling logic we've heard today and the encouraging response also that you spoke to among customers, is there not an opportunity for nearer term synergies purely from the coming together of the two sales organizations? Linked to that, I wanted just to ask you for a bit more detail around the go-to market. You mentioned maybe an opportunity later to pursue the indirect channel. At the moment, what's the sales process like for Medidata?
I presume it's a very complex and long sales cycle. You know, is there some overlap and benefit there to the coming together of the businesses? Thank you.
Let me take the second question, if that's okay. Our go-to market model, it's a mixed model. Let me explain. We have a direct model where we sell to pharma, life sciences, med device, and we have a sales force, field sales force, as you would imagine, geographically based. They sell the breadth of our solutions. There is an indirect channel, but it's not in the sense of the Dassault Systèmes indirect channel. There are CROs and some SIs who also are our partners in sales. We view them as our indirect channel, but they are really implementers. The CRO industry has multiple functions for pharmaceutical companies. As it relates to clinical development, they are outsourced data management.
They are outsourced monitoring, so going to the clinical sites and making sure that everything's happening. There are some other functions that they provide. Historically, we've worked with them as a channel partner. What they would do is they would co-bid. Let's say there's a study that's being started and the CRO wants to do the implementation work. They would co-bid Medidata's EDC product, for instance. When we won, it would be either on both of our paper or on their paper, and we would be a passthrough, always a passthrough to the customer. That relationship with the CROs has changed over time.
It's become much more strategic, where they are consuming our product both to make their own operations more efficient, and they're committing to use our technology and then reselling it as they resell their services to the pharmaceutical companies. That's the model that we've had, and we continue to see the CROs play a more, an increasingly larger role in our sales efforts. We will have, you know, maybe I'd say at this point, about 70% plus of our revenue is coming from our direct channel, and that's gonna continue to be a significant effort from on our part.
In fact, as we come together as organizations, the breadth of our, the breadth of what we can do on a global basis from a direct perspective actually increases because of some of the, as Rouven Bergmann mentioned, the presence of Dassault Systèmes in Europe. They're stronger than we are there. Also they have a much longer, larger presence in Asia Pacific than we do, even though we have a certain amount of revenue that comes from there. China, for example, is a very high-growth market for us, but I think we can accelerate there.
To complement what you just said, the CRO is only a category of partner for the research. You know you have the equivalents for the manufacturing side. Clearly it's an entire new ecosystem we are able to build to expand the footprint and the reach. Please keep this in mind because it's only the starting point. The second thing, coming back to the questions, what can we do in 2020 from a go-to-market standpoint? We have to be realistic. You know, the two teams will be together for the first time, and I do not want to take the risk to defocus them because, you know, it has to be well prepared. The plan is the following. We are working on a name list of accounts.
It will be a limited number of accounts, and we will approach them together with a combined value proposal. For the rest, we will continue in 2020 to promote independently. It's only for 2020.
Great. Thank you. Pascal, just on the financial plan, I was surprised that the G&A optimization is sort of limited to two points, given I think Medidata's G&A to sales ratio is actually quite quite high. I think it's in the double digits. Are there additional investments you're sort of baking in? also, when it comes to sales and marketing, is there any optimization there or even acceleration of spend baked in the plan? lastly, on cash flow, I know that the cash conversion of Medidata relative to Dassault is quite different. Can you shed some light on any plans to improve cash conversion?
Okay. I will start with the last one and maybe the G&A you can add what you want. Rouven, feel free. The cash flow is you do the math. You know, you have a revenue growing between 13%-15%, and you have a leverage of almost two points per year. Automatically, the cash flow will almost triple over the next five years. Going from a little bit less than EUR 100 million to a little bit less than EUR 300 million. This is the plan. I'm sensitive to that because, as you know, we have debt, and we took the commitment to deleverage rapidly. You remember for the first nine months, we have generated EUR 1 billion.
You could expect us to continue to be on the same trend for the next few years, and half of the cash flow will be used to reimburse the debt. That's the plan. On the sales and marketing, you know, could we accelerate and spend more? Yes. Before, you know, I think Bernard stated very clearly, the customer needs to be ready also. This market is shifting now. I do not want to be in a position whereby we over-invest in sales and marketing and discover that it's taking more time than expected to ramp up the sales. You almost blame me for the last five years because I was too pushy on the 3DEXPERIENCE platform, and now you start to see the light with big deals coming.
I want to be almost in the same positions, manage it collectively because we are on the same boat, and we will decide together how far we go from an investment standpoint on the sales and marketing sides. I think there is so much we can leverage by, you know, having our salespeople promoting the combined solutions that it's probably the first priority for us. On G&A?
Maybe before I come to G&A, one maybe clarifying aspect to cash conversion on the Medidata side is cash conversion is always also a function of the timing of invoicing. I think our invoicing timing is very different to the way Dassault Systèmes are currently operating it. Of course, it remains for us to see if we are going to adopt this and change. We typically invoice quarterly in advance. We have some contracts where we also invoice in arrears based on a consumption basis, so that delays cash generation. That's just to keep in mind. You know, we have a track record of increasing cash flow over time, so it's never been an issue for us.
On the G&A side, I think to answer it very simply, no, there are no incremental investments planned that would be a headwind for us to be able to get more leverage out of G&A. In more detailed terms, I mentioned buying power. I think we have to still quantify what that is going to look like. We want it to be prudent. Secondly, we will be, as Medidata, we haven't taken the step yet to think about global shared services, so we pretty much operate with a template operating out of the United States, which will give us an opportunity to tap into certain infrastructure of Dassault Systèmes that I think will give us definitely incremental leverage.
I don't see that we will have to invest into G&A to continue to grow our business, as we have been had to do in the past. All of these factors will contribute.
Last but not least, this story is not about cost saving. It's about growth. There is something Medidata is doing we do not have yet the full experience. It's a pure SaaS business model. All the G&A are dedicated to support a pure SaaS model, and I think it's our common interest, with the rest of what we do to leverage this expertise and to some extent diffuse, put the right leaders at the right place, and we want to capitalize on the current organizations to make it happen.
Just a follow-up question on share-based compensation, and how you expect that to evolve over time on the Medidata side. I think there was a nine-point difference between non-IFRS or non-GAAP and GAAP margins. Do you expect that delta to remain the same over the next four years?
You want to take this one, Bernard?
If I understand it correctly. Now, there are two questions, I think. You have the accounting treatment. Clearly, we apply the rules of what we do- currently. it will be, you know, it will be part of the non-IFRS. there is related to the policy, I guess. Are we continuing to do the same things and what we do because we have a different practice on the way we compensate the people and the way we, to a certain extent, we associate the people. there is one common pieces, if I may. We are rewarding the success. if we stick to the plan I just highlight to you and with Rouven , the practice will be almost the same than what you did.
One last question.
I mean, it's. You know, I think we have a track record of attracting very good people and retaining them. Our regrettable attrition rate globally for a software company was under 4%, which is highly unusual. That's because of the benefits package we have. It's because of our pay practices, et cetera. Obviously, we can't be a special snowflake in Dassault Systèmes. We have to be part of the family. That's what we signed up for. I think there is definitely room for us to continue the practices and benefits that we have in a way that makes sense within the larger family that we're now part of.
We'll continue to I think the big focus is hire the best talent and execute well, and I think that's the focus for everybody, up on the stage right now.
We will take one last question.
Yeah. It's Charles Brennan here from Credit Suisse. Just three really quick ones. I'm conscious of time, so it will be quick. Firstly, just to follow up on that share-based payment comment. Can you just help us from a modeling perspective, what's your expectation of the 2020 charge for Dassault and share-based payments, including Medidata? Secondly, we've heard about OUTSCALE a few times as a source of synergies. I can understand how this solves a short-term utilization a problem, but over a five-year view, can you just remind us why you're competing against the hyperscalers? Quite a few of the software companies are going the other way and embracing them. Lastly, from a software company business model, I don't often hear companies talk about wanting more services.
most people want to embrace the partners, whether it's a Cognizant or an Accenture more. Can you just remind us of the business model? What's the margin profile of services for you, and why not just give it to the SIs?
You want to take the services piece?
Yeah, let me address the services piece. Actually, services in the business that we're in are atypical for most SaaS companies. As you know, the model for a lot of SaaS companies is that they effectively give their services away. They have zero or negative margins on it. We drive between 35% and 40% gross margins on our services. They are an integral part of how the software is delivered and maintained by our customers, and they value it. We have customers who many years ago, over a decade ago, implemented Medidata's core EDC platform and who maintain services relationships with us over that entire time. If you look at about half of our services revenue, it's repeat business. It comes every single year from our customers. In fact, it's grown over time. It's not a commodity service.
Yes, it can be outsourced. CROs can be an outsourcing partner, some of the SIs can be. We bring, I think, a depth of knowledge and understanding that our customers value, and they're willing to pay for that, and it drives a very, I think a very profitable, very well-margined business for us. We continue to see that growing. In fact, as we look at what's happening with Acorn AI, we see that, being able to deliver the technology is very important, but having the right service wrapper so that you get the value from the technology is something that our customers are looking at. That's not gonna go away overnight.
For the stock-based compensations planned for 2020, my recommendation is wait February next year because we are here to discuss the long term. It's a capital market day. It's not, you know, an annual objective definitions meeting. Clearly, I will give you the details at that time. The last question was related to the sales synergy, right?
Just OUTSCALE moreover the previous points. I was looking for example, for Google or Amazon sort of integration.
Maybe you can take this one, Rouven.
Yeah.
you are the guy been negotiating with them.
Yeah. I think what's underestimated and underappreciated is you create quite some dependency. Having the ability to offer a highly scalable and a service that is regional and not just global, but also local to where your customers are, is extremely critical. I think it's going to be a key differentiator. For us, you can look at us as a small company and a big company, right? Our spend with AWS was somewhere between $10 million and $20 million. That's how precise I can be. I also have to say that it is very little compared to in terms of business volume. When you look at our overall business and customers that we serve, it's only a small part of it what goes through AWS.
When you think about how we are going to scale this out into the future, one of the concerns that we would have is how much money would we have to invest into an AWS in order to bring all of Medidata to these hyperscale companies? Of course, they're all coming and want our business. I think having options and choices and being able to leverage them is a significant advantage.
A few Well, I think my conclusion is we have the best team to serve the market. Really, I do think this is the most important criteria. I think with the best team, we can truly become game changer for an industry where we think the needs are there in every countries, and the timing is right for it. That's what we have been trying to communicate to you today, and I hope that you have seen things that you have never seen anywhere else in a real demonstration. Everything you have seen are not prototypes. They are real existing solutions as we speak. I think we are going to have fun to do it. Congratulations to all of you, and thank you very much for participating to this event.