Okay, let's go ahead and get started. Welcome, everyone. My name is Vikram Purohit. I'm one of the healthcare analysts with the research team, and happy to have with me on the stage Bill Feehery, CEO of Certara, and John Gallagher, CFO. Thanks for both for joining us. Appreciate it. All right. Thanks, Vikram. Yep. Before we get started, just need to read a brief disclosure statement. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. With that, happy to get into it. Bill and John, I guess to start with, could we talk a bit about the evolution of the company since the IPO, specifically in terms of Certara's business mix between software and services, how that's trended, and how that might evolve going forward? Yeah, thanks. Great question to start with. Well, during the IPO, so our IPO was in December of 2020. We were primarily a services company at the time, although we had a big biosimulation core. And we saw. You know, one of the things we did with some of the capital we raised in our IPO is we acquired a few software companies and created an engineering team around it to expand our software base. So if you fast-forward a few years, I don't remember what our percentage of software was back then, but right now we're about 60% services, 40% software, so that's been ticking up. Our investments have largely been in creating more software because it's core to the biosimulation mission that we have. It's valuable, it's sticky, it kind of lets us extend our arms into lots more therapeutic areas by having you know invested in software. And now the AI revolution has come by, and we're investing heavily in that. Our services are also quite important. They're very profitable, still 60% of our business. And you know we find that as a you know a really important way to magnify our impact, particularly in our smaller clients, who you know don't necessarily have the big groups that are necessary to buy the software and use it, so you know we really need to have the services side to serve them. Right. John, anything you might add? Yeah. I mean, just to follow on to what Bill said, I think this past quarter was the first quarter where we were less than 60% services revenue, so 59-41, so the 60-40 that Bill said. And that mix has trended from what was two-thirds services, and that's moved quickly based on the growth, the fantastic growth we've seen in the software business. Got it. Got it. Maybe that's a good segue point then to discuss the recent quarter in a bit more detail on some of the customer dynamics that impacted the quarter and that you think are gonna impact the year. So as a backdrop, I think with the quarter, you mentioned that, within services, kinda your Tier 3 customers were ahead of your expectations, Tier 1, a little bit maybe lower than your earlier expectations, and the software was steady. So kinda help us unpack all those dynamics and how you think the rest of the year is gonna unfold. Yeah. Yeah, so from a customer tiering perspective, we say that the Tier 1 customers are our largest customers. Those are customers with more than $5 billion of revenue. And Tier 3 are customers with $100 million or less or no revenue. And so what we did see in the quarter was that the Tier 3 customers, we saw some pickup there. We suspect that's on the heels of a strong Q1 funding environment for biotechs. And then, so we saw that start to come into our business, which we expected might be the case during Q1 when we saw the companies getting funded, and so it was a nice proof point to see that show up in some of our bookings during the quarter. Alternatively, as you mentioned, our Tier 1 customers, we saw some softness. I'd say it was in a handful of customers who are facing some challenges and some cautiousness around spending and budget, and that is causing some delays on our side as well, as far as achieving bookings, starting projects, and finishing projects. Got it. And, just to help us kinda get a sense of, I guess, how tight those tiering bands are, how frequently do customers move from Tier 1 to Tier 2 to Tier 3? How fluid is that? Those tiers are... They're relatively stable. I'd certainly say that within Tier 3, when a biotech company has revenue and then finds some success with their drug, then certainly you could see some movement from Tier 3s to Tier 2. But I'd say the top tier and the Tier 1 is relatively stable. As a reminder, too, the composition of our revenue, about 50% of our revenue is coming from Tier 1 big pharma customers, and about 30% of our revenue is coming from Tier 3 or biotech-type customers, and then the remainder is in Tier 2. Okay, so it's currently 50, 30, 20- Yes. Tier 1, 3, 2. Correct. Got it. Okay. So then looking ahead for, for the rest of the year, any, any guidance, any thoughts on how quarter over quarter things might progress, both from a software perspective, also from a services perspective? You know, software's been performing consistently well, even during some of the end market disruption that we've seen. When you look at software, it's been consistently performing mid-teens from both a bookings and a revenue perspective, and so we don't see that changing. We're happy to see that as a good proof point as it relates to biosimulation, adoption, and runway going forward. As far as, you know, how to think about how the services component is going to affect the company during the back half is we expect to see a Q3 that's relatively stable from the performance that we've been seeing, recently, and then an uptick in Q4, which is typical seasonality. Mm-hmm. So each year, we do have a strong Q4, and that seasonality, we saw it last year, the year before. So historically, we see that, we are anticipating, and it's contemplated in our guide, that that repeats again this year, too. Got it. Got it, great. On your guidance, I think you've guided for revenues of $385 million-$400 million for the year, right? What drives the bookends of the guidance range? Yeah. So, to be at the high end, we would need to see some significant acceleration across the Tier 1. All of the customer tiers and services, we'd need to see some significant acceleration really to get to the high end. And that's why on our last call, we said that, based on current trending and the dynamics that I just described, that we're really focused or trending toward the lower half of that guidance range at this point in time, based on the dynamics that we just talked through, particularly with the Tier 1 customer softness. Got it. Got it, okay, and on the topic of visibility, what do you see internally, and kind of what do you track internally to get visibility over the coming, call it, two to four quarters of your business? Yeah. We look at the data that's out there and available, so we look at clinical trial starts. We're looking at capital markets funding for biotech companies, then we're also looking at our own pipeline and what that pipeline, not only the size of that pipeline, but what our recent conversion rates are of pipeline to get to revenues, and those are some of the key measures that we're looking at to get an understanding of, you know, what the near-term few quarters would look like. Understood. Understood, okay. I guess more broadly, when you think about biosimulation and the overall addressable market here, I think one common question that comes up is, As you pointed out, I think in your materials as well, there is the potential for, I think, multibillion dollars in biosimulation sales, just kind of broadly. What are the, I guess, industry-level factors and the company-specific factors and your, kind of your customer-specific factors that need to happen and need to kind of unfold for Certara's top line to work towards that, that broader TAM that's been laid out? I think. You know, we've done our best to measure the TAM, but it's a, it's a complicated exercise because, you know, we're probably the largest company that's really providing biosimulation software. We have a lot of custom. We're, we're kind of in an ecosystem, particularly with our larger customers, who have internal groups that are using the software and advancing the science with people leaving and joining the company from them all the time. Right. And, you know, the reason why the TAM is sometimes hard to measure is because we haven't modeled everything in pharma that could be modeled, and so there's opportunity to keep expanding into new modalities, new therapeutic areas, things like that. You know, generally, we've seen a broad encouragement from the regulators, which provides some tailwind as to what they'll accept and what has been accepted as drugs have moved through. So we need to see that continuing. You know, a lot of our investment is on expanding the biosimulation software, expanding, you know. A number of years ago, we started expanding into biologics because that became, you know, even back in 2014 or 2012, that was becoming a bigger part of our business. But as we move forward, you know, we've expanded into, you know, new modalities like ADCs or CAR T, some of the new therapies, new technologies that our customers are having, and we need to continue to see broad regulatory acceptance of using those models there. And I think the other piece of it is just around the user base. So, you know, over time, we're seeing more people that come out of their PhD programs and know how to use this software. We're seeing more, you know, acceptance of the number of drugs that have been accepted, where there's label claims. That means that there's basically been avoidance of some degree of trials, so that builds upon each other. And so that generally, those tailwinds we expect to see continue. There's plenty more to go. We've invested a lot in what we call quantitative systems pharmacology, QSP, which is kind of the cutting-edge area that's kind of pushing our modeling forward. And that's brought in a whole bunch of earlier stage modeling activities, for example. So, you know, we're kind of in this issue where we're advancing the science, but we're also kind of advancing the biosimulation ecosystem that exists throughout pharma. And we've benefited for a while, and we don't really see that we're anywhere near kind of final innings on that right now. There's a long way to go. Got it. Are there kind of ready-made use cases in your perspective across your customer base, where there's just been maybe inertia or just a lack of kind of cultural acceptance that you think are kind of low-hanging fruit for Certara's biosimulation software and services that aren't fully utilized yet? It really varies. I mean, there we do a lot of different biosimulation. There's a lot of parts of biosimulation software and services, and some parts have become quite standard. Other ones, it really depends on, you know, whether it's applicable and what the customer thinks and, you know, whether they think that it will help with their regulatory or, you know, discussion. You know, if you look at things like, you know, PK modeling, it's pretty standard nowadays in the industry. Mm-hmm. We have software which is very widely used because it does the modeling well. It's tied into the workflow. Everybody's standard reports, people know how to use it, but it's very. I mean, you're really not gonna go forward with a drug development today and not use PK modeling. If you get into, you know, things a little bit more complicated, like population PK modeling, a little bit more complicated. If you get into QSP, it gets quite bespoke very quickly. Mm-hmm. And, you know, over time, those things, you know, someone takes it forward, uses it, you know, and then drugs tend to follow on, and those things become more standard. So, you know, it's kind of just a gradual progression that happens over time. I think Certara has been around a long enough time now that this isn't that new. So there's a pretty good population of people that are that understand what we can do, that know how to do it, and that understand that there's a huge benefit in terms of changing the probability of success in drug development and changing the cost of drug development that can be obtained here. So we're starting to get that kind of acceptance. But, you know, it's hard to estimate, but I would say, you know, we're still only a fraction of the total opportunities where modeling could be used or actually used. So there's still a long way for us to go. Got it. Got it. And then looking internally into the company, what are some of the areas where you're keen on making further investments to kind of build out your own capabilities and build out your product suite and offerings? Yeah, that's a good question. So we're investing very heavily in AI. Yes, I get it, a lot of other companies are doing it, too. You know, we look at this as specific to what we're doing in drug development modeling. You know, AI lets us do a couple of things. One is it lets us process unstructured data in a way we never could before. Most of our modeling is buried somewhere in scientific literature, which is really just unstructured data. So the ability to, you know, to basically leverage our scientists to process through that and create meaningful models in a fast way, changes, can, has the opportunity to change our business. Most of our work results in some kind of a document that has to get created, written. So we've invested in one of the products we launched recently, we call CoAuthor. It has to do with, basically using AI to write regulatory documents. Takes a lot of the time and cost out of that, and then, you know, it's a big market in general. So those are two of the things. We're also investing a lot in just the flow of data in pharma. So our Pinnacle 21 product is a validation product. So what it does is it really says, if you have a data standard, does this data meet that standard? We've added to that over, you know, through acquisitions and through our own internal investments, metadata repositories. So basically, you let pharma companies define their own standards. They can enforce those standards to their EDCs, to their clinical trials. Then they can use Pinnacle to determine whether the data that comes from their vendors meets the standard, and you can cut out a lot of cost and a lot of time from this massive data collection process, which pharma companies need and we need for biosimulation, for example. Mm-hmm. You know, the last thing is we're investing. We announced that we're buying a company called ChemAxon, and that's an investment that we think is really important because we want to move biosimulation much earlier in the drug development process into the discovery and the lead optimization phase, and investments in sort of integrating that with our biosimulation products are gonna be pretty important as we move forward. Got it. Got it. Great, that's helpful. I guess on the topic of ChemAxon, since you brought it up, how do you weigh the pros and cons of developing capabilities internally versus looking outside? And I guess, what is your filter for, BD, M&A? Yeah, it's a good question. We're in an area where there's just a lot of little companies doing interesting things. And our view is we just have, you know, we track a lot of people and, you know, sometimes the stars align where you have, you know, the way we tend to think about it is, we have technology that we can integrate, we have a team that we think wants to stay and grow with us, and a price that we think is fair for our shareholders. They don't always align, so we're always looking at a lot of things, and when they do, we move forward. Pretty much all of our recent acquisitions have been companies that we worked with for quite a while. ChemAxon is not an exception. We've worked with them for years on one of our products, so we know the team well. We've done co-marketing with them. A lot of our customers buy their products and ours already, so we kinda see the possibilities to integrate. And on top of that, their technology, which revolves really around searching for chemical entities, is quite complex, and it would take us a lot of time and a lot of money, and frankly, I'd have to hire a whole set of different people to even attack that. So I think, and then to answer your question, I think that's one case where, realistically speaking, I don't think we were gonna do that internally. Other things, you know, our bread and butter is biosimulation. It's not that we don't look at other biosimulation companies, but we feel pretty good about our internal capabilities. Our internal capabilities are quite good, and we're, you know, on a path, so integrating something might be a little bit harder on that. You know, I think it's a balance. You know, every CEO has got to balance in, you know, spending dollars internally versus maybe you can get some acceleration externally. I think in our case there's a lot of small companies. We have a very good track record. I forget, the last eight or so companies we bought have been relatively small, but if you look at both what we paid and what the result's been, it's been a very, very good deal for shareholders, and it's enabled us to build out this software platform that we kind of see as a vision for the long-term growth of the biosimulation market. Got it. Got it. Would you say the company's now in a bit more of a, more of an integration mode, given the recent acquisition? Or do you think you're still open for future BD if the opportunity arises in the near to midterm? Well, I mean, like I said, there's always things out there, and when the stars align, you know, you have to be able to move. We have a good balance sheet, so- Mm-hmm ... you know, we have some capability if we have to. But having said that, you know, we're acquiring ChemAxon in the fourth quarter, and we've recently bought two other companies in the last year, both of which are doing really well. But you know, getting those right and integrating them has been a big piece of our focus this year. Got it. Got it. Okay. You talked about AI a little bit in your kind of opening remarks, so we can maybe double-click on that. What have been some interesting products or product suites where you've woven in AI, and it's been, from your customer feedback, highly, highly beneficial to your customers? And I guess, what do you think could be the near-term revenue lift from integrating AI into a couple of your product suites? Right. Well, I'll start, and you can talk about that part. But the one we're highlighting right now is we just launched CoAuthor, which is our regulatory writing product. It's more complicated than you think. You can't. This is not something. Regulatory writing is not something that lends itself very easily to just, say, using Microsoft. No, no problem, Microsoft Copilot. But if you were to look at the specifics of what you have to produce and the environment and the data you have to tap into, there's a lot more to that product. We launched that this summer. We have it in the hands of paying customers already, and our internal regulatory group has been both trained, and frankly, they've been highly involved in the design of that product. So, we're pretty sure that the regulatory writing market's gonna be highly disrupted by AI because you're producing a lot of unstructured data. Some of these documents are millions of pages. It's really expensive. Everybody, you know, everybody in pharma is paying a lot of money to produce this. And, you know, our early estimates are it's not unreasonable to take out 30% of the cost using AI. Do we take out all the people? No, we don't. You're still gonna you need a human in the loop, maybe a couple of humans in the loop for this. But you can take a lot of the less interesting, the sort of, you know, lower-level parts of this- So that would be 30% of the cost of producing that specific document? Correct. Right. Right. So that's a pretty big number, right? Right. And, you know, we think, we think there'll be... This is one of the markets, you know, I think there are a lot of markets like this, where people are writing lots of documents, where you're gonna see some disruption from AI. We think this is an important one. We think we know this one, and we're getting a lot of interest from that particular product. We're also using AI to produce custom AI tools for companies. So the technology that we've bought from Vyasa is what's called a data fabric, and that's kind of a fancy name for, you know, companies have lots of internal and external databases that they would like to train an AI on. They don't necessarily want to give that data to anybody, even us, and they don't want to pull it all out of a database and restructure it in a central one, which goes, it'll cost a lot of money. So what this lets us do is leave the data where it is, index across it, train an AI, and you can have your own custom one based on the data that you own. As you know, across pharma, you know, some of these companies have a hundred years of data stored up, which, you know, it's great to have it stored up, but how do you use it? Right. And so, you know, that's become a bigger piece of it. We're interested in ChemAxon because they, we're gonna create products around basically predicting which leads, which lead optimization, which leads in the optimization stage you can bring forward. So there's a lot of things we can do. And we're interested also in using it in modeling, right? So, you know, modeling is a big exercise of processing massive amounts of scientific literature to create a model. And so, you know, using AI to pull that out in a more efficient way makes our people way more efficient. If we can create a model faster... One of the problems we often have is someone will say, "Well, I'd like a model for whatever," but it might maybe it'll take us a year. Well, they don't have a year. Right. They're gonna go off, and they're gonna do something else. So if we can shrink that time, we can expand our market, for example. Right. Right. And when you say, like, they could go do something else, in those situations, what do customers do? Do they- More trials. -build their own in-house? Oh, they just do more trials, brute force, more trials. Well, if they want to sit around... I mean, if we can't make it in a year, it's unlikely they can. Right. So, you know, yes, you can invest in- Or some other- Creating it. Right. Company could, right. Right. But a lot of times what you'll see is they'll just run off and do trials because, you know, you got to keep the project moving, right? Right. You know, we don't lose tons of business that way, but generally, if you can produce a model much faster, you know, we can meet the needs of the customers a lot better. Got it. Got it. On the revenue piece of your question- Yeah ... too, just to address that part. You know, we're generating revenue dollars from AI right now, so we have customers. ... that are paying us, and so we're gaining traction. As Bill mentioned, you know, the CoAuthor product is front and center for us at the moment 'cause it's live on the market now. So I think near term, that's our biggest revenue opportunity in the nearest term. Hmm. And then over the long term, we're, of course, investing in the platforms, as Bill had described. That's gonna help us maintain, sustain the strong revenue growth we've experienced in software and hopefully to help accelerate that as well. Got it. Got it. And when you say you're making revenue from AI, you mean specifically you're generating incremental dollars from weaving in AI into your current product suite, right? Like CoAuthor. Yes, that's correct. Got it. Yeah. Got it. Okay, great. So we've touched on kind of the evolution of the business, product development, BD, M&A, and just the broad topic of AI. I guess the next best place to go maybe would be competition. I'd be curious to hear your perspective on how the competitive landscape might have changed since, I guess, the topic of AI, ML technology has become a bit more pervasive in drug development. Do you think that more technology in the hands of more people, whether that's in-house at biopharma companies, whether that's with kind of third-party technology companies, does that in any way lower the bar for competition for your biosimulation offerings? And are you seeing any sort of increased level of competition since kind of this wave of AI and ML has seemed to have seemingly become a more common discussion topic across the biopharma industry? Or has it just been completely just- Yeah ... irrelevant for you? I think for what we do, that hasn't really changed, so the core offering we have is these are mechanistic models. That means that we're modeling and we're translating into mathematical equations, the known science, and we're doing that in a way that you can explain to a regulator what you did, where you got the data, where you got the parameters, maybe what you looked at, and you didn't include because you think it was wrong, but it's a whole documented process, and AI doesn't really change that, and you know, the fact that really the regulators are looking for the scientific understanding. What AI does is it lets you maybe expand that, right, so there's a chunk of things we don't know how to model, but we might have statistical data. We create, you know... So finally, we create a statistical model, which lets you create more complicated ones. So you know, I would say, it's probably gonna, you know, a lot of people in pharma are interested in solving the drug selection problem. So there's a lot of companies out there doing AI drug discovery. It's very exciting. A little bit different than what we do, so I wouldn't say they're really competitive. To some extent, if they succeed, we think that's great 'cause we tend to model what happens in the clinic. So if they come up with a great molecule and discovery, we're probably gonna, you know, wanna get involved in that at some point. Sure. But I don't think it's really changed the core, you know, the core dynamics of the industry, which is, you know, if you wanna avoid doing a trial, then you need to really focus on what's the underlying science and get that right in a way that you can prove it. And yeah, I would say there are. You know, we're certainly not the only company out there. There's some other. For one thing, a lot of our customers have significant internal capabilities around doing this. But I'd say it's the competitive dynamic is kind of proceeding the way it was before, so in our core markets. It's opening up new possibilities like our, you know, our the product we talked about, Copilot. Mm-hmm. Yeah, sure, that's a different market for us that AI has opened up. We'll probably see different competitors as we go forward. But, you know, I think that's all pretty exciting. Got it. Got it. And then in terms of kind of maintaining, like, a durable business, right, what are you seeing as kind of the key considerations that impact your ability just kind of to maintain your customers? And has that kind of changed at all through some of the dynamics the biotech industry has gone through recently, right? I don't know if the question makes sense, but it's more like if through the different biotech funding dynamics that you've faced, kind of across a portion of your customer base recently, has your ability to maintain customers changed at all? And has your, I guess, your commercial infrastructure or your marketing capabilities, have those had to adapt to be able to maintain your customer base? Yeah, I think that the last couple of years, there have been several disruptions in the pharma markets that I'm sure other companies are talking about. So you have the IRA, you have everybody coming out of the pandemic, and there were multiple effects there, and then there's been some dynamics specifically in biotechs over the last couple of years there... Right ... as well, and I would say, you know, it's been hard for us to predict when they were coming along or what all those were, but, you know, still through that, I would say, you know, we still have, you know, a customer base of, you know, almost 2,000 customers. Right. We're pretty broad-based. We have experienced some disruption as, you know, pharma companies have changed their portfolios as a result of some of these factors. You know, I guess the other point I'd make is the company's kind of evolved over this time. Our sales and marketing, we've been actively developing that over the last couple of years as the company got bigger. It's a little bit hard to separate that from what's going on in the industry. If we hadn't done that, where would we be? Sure. But I'd say overall, you know, we're being much more purposeful about making sure that we've got the entire world market covered, that we have the right trained person, you know, hopefully hitting the right opportunity at the right time. I'd say our sales and marketing investments have been, you know, reasonably heavy over the last couple of years. They're probably not gonna accelerate as fast as they have in the past, since we're feeling pretty good about where we've gotten to, and you know, the products have changed as well. So, you know, even if you compare with our IPO, you know, the product suite we have is much more sophisticated and much more integrated than it was, right? So- Got it. Got it. So, I mean, it seems like net-net from your comments that whether it's been the proliferation of AI, ML, or whether it's been kind of the biotech funding market ups and downs and pipeline prioritization that biopharma companies, competitive intensity doesn't seem to have changed too much for Certara over the past couple of years. I would say, you know, it's probably pretty steady. The biggest issue for us has not been... I mean, to be honest, it's not that we don't pay attention to our competition, but there's so much opportunity to make this market bigger, that really, the focus has been on making sure that we find all the opportunities where biosimulation could be used, 'cause it's way bigger than what we actually find. Right. And so that, that's what I'm trying to do with filling out the products, you know, creating an integrated product suite where we can follow a drug through its development, where we've got a professional sales and marketing team, where we're... You know, wherever the pharma industry is- Yeah ... whatever size company you're talking to, we've got an offering, or we've got somebody knowledgeable that you can talk to. And so that's probably been the bigger opportunity for us than worrying. And like I said, I do take the competition seriously, but you can grow this market so much faster than you know, by just finding all the white space opportunities, really. Got it. Got it. And is ex-U.S. a focus at all for the business right now? Or growing ex-U.S. as a percentage of kind of global sales, is that a focus for you? Yeah. You wanna- Yeah. Yeah. Ex-U.S. continues to be about 25% of our total revenues, with the majority of that coming from Europe. It is an area of growth and focus for us, especially as some of these newer markets we're able to penetrate for the first time and get as new customers on biosimulation. So I wouldn't say it's like the key growth driver, 'cause you haven't heard us mention it in that way, but it is a meaningful contribution to our overall revenue achievement, and we do think there's a good growth runway there into the future. Are all the software and services offerings available in the U.S.? Are they effectively all available ex-U.S. as well based on whatever the customer needs are ex-U.S.? Yes, to a large extent, that's true. In fact, you know, one of our businesses, the Simcyp business- Mm-hmm ... is based outside of the U.S., it's based in the U.K. So, it is a very global nature to not only the company but the products and services that we offer. Many of our consulting base are outside of the U.S. as well. Got it. The subject matter expert scientists that provide the work for regulatory services or biosim services are outside of the U.S. Got it. Got it. Okay. I guess maybe we have 35 seconds. I'll ask you a final question. Five to 10 years from now, how do you envision kind of Certara's portfolio of software services evolving over time? Broad-based question for you. Look, I think the way we think about this is the pharma industry fails, like, 90% of the time when it's developing a drug, and that, that's killing the industry, especially with the IRA, which effectively takes some of the profits out. Maybe that'll get worse, maybe not. Either way, it's a problem, and if we do this right, biosimulation has a possibility to meaningfully change that, and if you just change it by a little bit, it's worth a ton for the global pharmaceutical industry, so our vision of this is, you know, we're creating a software platform around biosimulation, where we can start very early in a drug development lifecycle. We can help with discovery, we can help with lead optimization, we can... All this is gonna, the data is gonna move seamlessly 'cause pharma companies bring in different groups of experts, but as you're moving along, you know, which molecule did you pick? What was the dose? How are you gonna design the trial? Which trials maybe do you not need to do because we know certain things, and then even beyond that, you know, our vision is, as you get into medical affairs and you're talking about, okay, now the drug has been approved, but how are we gonna prove its value to payers and the medics? Right. We can do the same thing. Creating that overall platform with basically world experts who are really, really good at this and get your drug- Mm-hmm ... you know, really moving. That's kind of the vision here. We've gotten a long way towards it. There's still plenty of opportunity left. Got it. Got it. I think that's a great note to end on. Bill, John, thanks so much for your time. Really appreciate it. Thank you all for listening in. Great. Thanks a lot. Thank you. Appreciate it.