Waters Corporation (WAT) Earnings Call Transcript & Summary

Good afternoon. Welcome. Dan Brennan here at Cowen's 42nd Annual Healthcare Conference, Day 3, finishing strong with the CEO of Waters Corp. Udit Batra, who I've known from back in his Merck days. Really looking forward to this conversation. Hopefully, it's been a great conference for all of you. Feel free anyone who's watching or listening to the webcast to shoot me a question through the video link. I think you've been using it, and I'll see it pop up and we'll try to integrate it with our flow of questions. But I just wanted to say to Udit, welcome, and thanks for being here.

Thank you, Dan. Happy to be here.

Excellent. I thought maybe a couple of kind of high-level question to kick it off. I mean, one of which is just recognizing the state of the world with what's happened in Russia and Ukraine, the awful tragedy there and some of the fallout from that, the oil price is up 30%, risk of recession probably higher and consumer confidence, we assuming business confidence probably being negatively impacted. Just wondering, Udit, like, to what extent from some of those metrics, anything there that kind of creates a risk either as we sit here today for first quarter or for your '22 guidance?

So Dan, firstly, I think the most important thing here is all the people were impacted by what's happening right now. As you know, I have lots of friends in different parts of Europe who are physically much closer to the crisis. The first thing we did is we got our employees out of that zone. In fact, the morning of the invasion, I think it was 5:28 a.m., I sent a message to my HR head and said, get everybody out and we were successfully able to. And there were [ couple guys ] in Russia. That's the first order of business. And I mean, our heart -- at least my heart goes out to everyone who's there and all my friends and colleagues who are in Europe as well. Overall, from a business standpoint, it's not that impactful. I mean, Russia, Ukraine constitute less than 0.5% of our sales, so it's not that substantive at all. From a supply chain perspective, there's not much impact either. We don't source much out of Russia or out of Ukraine or that part of the world. So really from a business standpoint, not much impact. And then I think you talked about inflation and commodity prices. Our TA business does sell into that area overall, as you know, the industrial space. And that -- I mean, the business has done super well over the last year. There is no more sort of abating of demand that we see on that front in the TA business. And secondly, we've been pivoting that business more towards even faster-growing areas with secular demand like batteries and recyclable polymers. So all in all, really no meaningful impact. And also from a pricing perspective, I think that does not add to our challenges, at least not right now.

That was great. So it's been about 1.5 years since you joined Waters. The company has seen a material improvement in growth from the low single, mid-single up to the mid-single plus, including this year, you're at 5% to 7%. You've got 5 initiatives, probably many more than that, but I believe 5 that you've been discussing as a key enabler to add 100 basis points a year, plus a revamped management commercial team on top of that. So maybe the question is, of the 5 initiatives, like which of those represent, you think, the biggest benefit as you're thinking about your 2022 guidance? And like, do you endorse that 5% to 7% growth? Is that the right way to think about Waters as we get past '22 into '23 as well?

Dan, a lot to unpack there, right? But overall, first thing and the most important thing you need to know is Waters is back on track. I mean that's what we have said that we would do first. Commercially, for sure, and the metrics suggest that, and I'll get into that in a minute. And when you said more than 5 initiatives, no. In fact, it's 2 more than I usually would have had. I usually think in 3s, and I said, okay, 5 is fine. We had 50 ideas and we said, no, 5. And that's it, right? And that gets implemented through and through the organization. So it's not 20. It's only 5 commercial initiatives that are really driving the growth. They all add -- I mean if you just look at the overlaps, et cetera, about 1% for the growth. And that's why the algorithm is, look, mid-single-digit market as it stabilizes, 4% to 6%. And so we should have a 20% lead over the market. If the market's midpoint is 5%, we'll grow at 6%, if it grows at 7%, we grow at 8%. And we think these initiatives add on. And all of these initiatives should help us at least over the next 2 to 3 years and some even longer, right? So now to sort of pick through them, one of them is adding in its own right. I mean, I don't think you can pick one and say, well, this one is higher and this one is lower, this year and versus next year. I mean the instrument replacement is self-explanatory. We added $30 million in 2021. We'll add $40 million, so $10 million incremental. The service attachment rate, super exciting initiative. We're already in the mid-40s in the service attachment rates from plan to warranty conversion. That's a pretty respectable number as I look at industry peers, but we think we can take it much higher than that, especially with what we have done in the U.S. and in Western Europe, and we can do that in China. I think it's quite a nice runway, that was 200 basis points over the last 2 years. Contract manufacturing as a segment, again, something that was ignored for many years. I mean we've just started, right? So we've gone from about 15% penetration to 20% penetration. We think it should be around 30% or so. That's going to take a couple of years as well. E-commerce, you know my history from MilliporeSigma. And I mean, there, we had 70% to 80% of our business in -- with a significantly larger consumables business going through e-commerce. And here, we had about 20%. So we have a nice runway to get when the market median is around 50%. And so the next 2 to 3 years, maybe 4 years, we get to 50%. So you can see a nice runway there. And then finally, launch excellence. Waters has a history of finding difficult problems to solve, go back to the UPLC time frame. Now with our premier technology, with our MRT and with our Arc HPLC, we find difficult problems to solve. We solve them. We were just sort of stumbling on launch excellence. And the latest case in point is relaunch of BioAccord, which we repurposed from QA/QC to clone selection in bioprocessing, and that's starting to get meaningful traction. And I think that basically is it. And you're asking me to pick amongst my kids, which when I bet on, I don't bet on any of them, I bet on all of them all together, and it's going very well. And thank goodness, it's not me operationally running. In the end, it's Jon Pratt and Ian King and I'm more watching them. It's a terrific team. So I feel very good about where we've started and where we're going.

Great. We'll certainly dig into a few of those here coming up. But maybe before we do that, I'm going to go beyond those 5, right? You start to discuss 3 new areas on top that can impact growth -- excuse me, starting in 2023, I think bioprocess characterization, bioseparation and LC-MS and diagnostics. I remember going, I think must have been ASMS like in 2013, we were talking about diagnostics at Waters back then as well. So which of these is -- no, listen, it's not you, it's like -- I mean it is a very accurate tool, right? It is extremely accurate tool. So there's so much opportunity. We once had a call with Mayo Clinic a couple of years ago, and they said they're putting the screws to the mass spec vendors because we want this tool to be more pervasive and make it simple to use. So everyone understands the power, right, there. So it wasn't a jab by any means. So maybe which of those 3 you think is the closest to being a material revenue contributor today, a? And then b, are these 3 going to come on top of the 5% to 7%? Or are they're replacing some of those drivers? Like how do we think about that kind of shift?

I mean, most -- the best way to think about it is on top, right? So we said the 5% to 7% has the possibility of becoming 6% to 8% higher or lower, I mean don't ask me. I mean, remember, when we started MilliporeSigma, the weighted average was 4.5%. We ended at -- when I at least left was about 9%. So I'm not going to say that's the trajectory. But I think if you focus on tough problems to solve and you don't veer too far away from what the heck you know about, right? I mean if you don't veer too far away from your core competencies, it works out, right? And in this case, it's really relevant. And these 3 areas, there are different levels of maturity, right? So let me take each of these in turn, right? So let's start with LC-MS and diagnostics. This is the one that you started with. We have a $150 million to $200 million business already in diagnostics with mass spec for newborn screening and for the therapeutic drug monitoring. So it exists already. And we had a dry run with the SARS-CoV-2 virus. And I can tell you, had we started that process of developing an LC-MS workflow for diagnostics for SARS-CoV-2 back in January of 2020 as opposed to September, October of 2020 when I came in, we would have definitely had it in parallel to PCR. There's no question in my mind, right? The workflow is that good, the selectivity and the sensitivity is high. And you have much fewer false positives, right? With PCR, you know -- I'm going to travel next week and I can't take a PCR test because I had even after getting vaccinated and boosted a 2-day Omicron bout. So if I do it now, there's a possibility that PCR comes up with a false positive and then I can't travel. So I'm not doing PCR, am doing an antigen test. Now with LC-MS, that's not a possibility. So it's a tool that is made for this area. And what is the challenge? The challenge is assay development, the challenge is regulatory and commercial. We have the box and we have the informatics. And that's what we're working on, right? And our target is to go into the endocrinology and go into oncology where you need multiple biomarkers to be analyzed in parallel with a multi-omic tool. It's a tough problem, Dan. Otherwise, it would already have been solved. But you know what, I think if you don't try, you don't fail. I'm not saying we fail, but we're going to try and it's going to be a long process. And anything that helps us accelerate that journey, right, be it sample prep, be it informatics, be it commercial or regulatory capabilities, we'll look at inorganic and partnership options as well. So that's one where we already have a base, right? And we know what we're doing and we're progressing very nicely, right? And we've in fact, what we've done, again, picking out from what we did at Millipore, we basically separated that business out of our big Waters division and said, okay, you're going to be globally verticalized. Jianqing Bennett has it under her now, and she is leading it. And basically, it's a global business. So you win and lose and we know what's happening as opposed to it be getting muddied by other businesses, right? So I think it's -- I feel very good about what we're trying to do. It's early days. Second is bioanalytical characterization. And again, this was somehow again a bit opportunistic, but also really thinking through where there is a difficult problem to solve. And this is our collaboration with Sartorius, right? So having worked in this industry for a long time, I mean, one of the problems I knew about is clone selection in early-stage development of biotherapeutics. So in every process lab in the world, you are trying to do clone selection or you're trying to do bioprocess optimization. And the BioAccord is tailor made for that application. So for instance, the workflow for clone selection of cell lines takes about 6 weeks with the Sartorius Ambr bioreactor system. And we were able to work with Sartorius and narrow that down to about less than 2 weeks, right? So it's a significant acceleration in timing, means you can go to market much faster. So that one is also very nicely on track. And then the last one is bioseparations, right? So again, we are a world leader in developing columns for separation of small molecules and the same logic applies for large. And here, we have invested quite a bit organically and we're looking for ways to accelerate that by partnering with reagents, companies and the like. So you can see across the board, there is progress. But if I were to pick, I would say, stable base already in LC-MS, we're just moving sideways. And organically, the team is ready and moving. In bioprocessing, we've been very fortunate. We have a lot of traction with our workflows, with phone selection and we're looking at other areas. And with bioseparations, it's early days, but it looks reasonably good.

And in terms of the cumulative -- again, we'll find out more as we move forward and you size. But cumulatively, those work out, again, they're coming at different points, and you're saying there's investment and it's going to take time. I mean, could those be more than 1 point, less than 1 point? I mean it sounds like you've given some numbers I know on the Sartorius installed base and you've walked through some of that math. But I just forget, I apologize, like is there an aggregate potential from these if they go reasonably well?

No, I don't -- because it's so much sort of fuzzy math with it, right? The timing and the total accessible market. In all 3 cases, it's a new application we are creating, right? So there is not an existing market that's been sized by competitors, right? It's a new market that's been created. I mean, I'll give you an example of the bioprocess one, right? I mean, we said there are 800 Ambr systems, and we went bottom up and calculated $0.5 billion or so TAM. Now that could be 2x, 3x, 4x that if the traction is better and if we go to many more labs than we had assumed, right? So that is the bottom-up basis. Another way to calculate it is, say, look, every processing lab should have a simple LC-MS tool in it. If you go around with that calculation, you come to a similar number, right? So $0.5 billion or so of TAM. And then you say, well, is it in 3 years, 5 years, 7 years, 10 years? I don't know if it's a question of how fast we solve the problem. It could be a very fast traction. I mean the way to think about it is -- the leading indicator there, Dan, is how quickly some of our customers adopt the technique in their workflows and how quickly it starts to bear in the drug master files, right? And that then gives you confidence, okay, this is something that the regulators accept. Now we are in a new world of bioanalytical characterization.

Understand. Okay. That's great. So maybe circling back on maybe some of the initial 5 initiatives and then go through some other questions here. But the first one is just on e-commerce, mix of consumables. So as you go up from where you are today towards, I think, the goal of -- I think you said it was like 55 or it might have been higher than that, apologies. What's the revenue pull through as that mix goes up?

So every -- I mean, rule of thumb, right, every $5 to $7 or $10 that you spend on e-commerce, assume $1 is incremental. I mean if you do it extremely well, it could be better. But if you do it poorly, it's on the $10 range. For every $10, it's $1. If you do it very well, it's for every $4, it's $1, right? So it can be pretty significant upside.

Yes. Okay. I got it. Maybe shifting over to instruments. You had a really robust year last year after 2 years of negative growth. You've got the tough comps, so you're guiding to, I think, 5% this year. But given the trend line in 2019 and 2020, I would think that 5% still could be conservative just given the cumulative base of [ orders ] correct? So just kind of wondering on that. And then which initiative has the biggest impact for you on instruments?

So take a step back on instruments, right? Historically, instrument growth rate for Waters -- and you've been tracking Waters for a long time -- has been 3% to 4%. And I'm not talking 2015 onwards, where we had a bit of other challenges. From 2010 to '15, the average instrument growth rate is 3% to 4%, and that was somehow industry-leading, right? We grew on a stacked basis. Ignore 2020, because the growth rate was dramatic, right? I mean it was 23% versus 2020. But versus 2019, on a 2-year basis, on a 2-year CAGR, it was a 5% growth. So we were tracking well ahead of historical average, say, 20%, 25% higher than previous best years of Waters. So that is a reasonable starting point. Now look at it another way, right, so let's just take another angle. And then you can imagine, I mean, again, as you know me, we're pretty analytical about how we think about estimates. So we've basically also looked at the competitive set, right? And on [ full ] year basis -- forget 2020, if I were to just take the analytical instrument businesses of our key competitors, and you know who they are, right? You take them and you say, okay, and it's very difficult these days when people are changing nomenclature from core to organic to COVID versus non-COVID. But for several competitors, it's very straightforward. You can actually pull it out, and it's very clean as they report the full year. On a full year -- on a 2-year basis, I mean, we grew 7%, 7.5% on a stack basis. Rest of the industry was slightly shy of that. I mean, the average was around 5-ish percent. And when you look at the instrument growth rate, it was again between 3% and 4%. So the 5% number is actually pretty ambitious, right, on an ongoing basis, right? So in general, when you think of the algorithm, 5% of instrument growth, if you just think of what happened versus 2019 and 2021, on a stack basis, we said 7% stack growth, 5% on instruments, 6% on service and then 9% to 10% on consumables. And that's how the 7% number was deconvoluted. So that algorithm feels pretty relevant.

Yes. Okay. That makes sense. No, that's totally fair. So maybe just back to the replacement cycles. Just what was the -- is it possible to tease out like how much the replacement cycle impacted you in '21? And what's kind of baked in this year for that? I mean, is it kind of -- just trying to unpack the replacement cycle impact for you?

No. I mean, it's another $10 million on top, right? So $30 million incremental in 2020, because [indiscernible] we've sort of baked in 2022. And we had about 13,000 we found across the board, across LC and different types of mass spec and UPLC that we found should have been replaced over the last 5 years or so. And we basically said, okay, put them into a database and we started going through it. So as we've gone through it, we've contacted each and every customer, we feel there is still a nice runway over the next 2 to 3 years that we will contribute as much or more than it has in the previous years. So still a lot to do. And you can ask, why does it take long? Customers are not -- they don't have just capital sitting around when we go and say, we've got to replace this and replace that when their capital cycle comes up -- and they're not replacing 1 or 2 instruments. These are big replacement, right? So people usually buy 50 LCs if they're going to replace everything in a lab. And that's a pretty significant purchase. And so some of them have said, not this year, but next year, for sure, we will do it. So that's how we do the projections, yes.

Got it. Okay. So maybe looking at a little higher level on pharma. There was a few smaller competitors who noted some near some pressure from emerging pharmas. I'm sure you've got this question 15 times. But net-net, we're still trying to get level set with some of the larger players in this space. So emerging biopharma, balance sheets are great. Now the capital markets are effectively shut. So are you seeing any impact from that group of customers in terms of demand trends? And maybe could you give us a sense of how big maybe pre-commercial pharma is for your business?

Less than 5% of the total pharma business, right, that we have. So 60% of our total business is pharma, I mean, less than that, less than 5% of that is pre-commercial pharma. And you know Waters' business model. I mean, we are more heavily weighted towards QA/QC. And they're so in development and at least [indiscernible] right? And that sort of jives with what I just told you, right? So our -- I mean, our business is very resilient because it scales with the volume of testing of existing molecules that have been already filed, right? So it's very predictable. Most of the revenue comes from medicines that are already on the market and repeatable testing. As long as we replace the instruments, as long as when the customer calls and wants a service call or some consumable, we supply it. I mean, that part of the business is super robust. The early stage is more relevant for chemistry, there too, in pre-commercial pharma, it's less than 5% of the total business.

Got it. Okay. Maybe just one more broadly on LC-MS. So it's just interesting that we -- you don't hear who's losing share. I mean all the vendors say they're gaining share, right? So -- or they're holding share. So I'm just trying to -- you and I have talked in the past about trying to compare some relative share numbers. And I'm just wondering, is there a decent part of the market that is from smaller players that we're not really focused on that are just not as competitive and that all the big players at varying degrees during their strong cycles can really take some share? Or just how do we think about the cumulative market growth and then also the share shifts?

Yes. It's a little bit frustrating, but I mean if it's relevant, people will go after it, right? So I give you sort of a 3 faceted answer. So when I took over, I said, "Guys, I need to know how we're growing, how the market is growing, i.e., Waters is trailing the market." And people say, "Ah, but we don't have the right portfolio, other folks have this portfolio and you came from a bioprocessing player and then you had this and that." And I said, "Okay, clean it up. Let's develop an algorithm, it's where it's apples-to-apples, okay?" And we did this in Millipore as well. We said we'll compare our bioprocessing business to Sartorius, right, the fastest growing. And then I used to have debates with corporate on who's faster every quarter. And perhaps you could talk to [ Tudor Caspar ] and ask him who is faster every quarter because we used to look at each other's numbers. But you can do that if you really look at people's public reports. So all the key competitors, so that's just a starting point. And I said, "Okay, development algorithm." So we developed a simple algorithm, right? And this is the second point I'm going to make. We said, what is the analytical instrument base of these competitors? The larger ones, Thermo, Danaher, et cetera, and even Agilent, that is publicly reported, and they report these numbers, organic growth numbers, right? And that is the most relevant comparison to us, take out M&A and take out COVID, and they do that themselves. So you have very clean comparisons historically. There is no leading indicator. So historical comparisons are very easy to do. And if you do those comparisons for full year 2021, right? So do full year 2021 for all of these folks and us, what you find is, on a 2-year basis, the market amongst the analytical instrument businesses -- and these are larger ones -- you find basically 5-ish percent stacked growth, and we were clocking close to 7%. And there's another one who's closer to 7%. I won't tell you which one, but you can do -- it's all publicly reported, Dan, right? I mean it's 6 or 7 players. And that's how we look at the league table, right? So that's the second method. And the third one is self-reported, right? Our reps come in through our CRM, tell us the win-loss ratio. And that is a proxy for the market share. And it has been significantly higher than it was in 2019, right? And so I don't want to cite that externally because that's an internal number. But every direction we look at, we feel we're holding our own or doing better. There is no publicly reported market share number, and we're working on this. I mean our intent is to develop that, an objective view on that, and I'll share that once we finish that algorithm. But I feel we're definitely holding our own and doing better. And I don't know which one is losing, but if you just went through that league table yourself, you'll find some are growing at 3% to 4%, some of the larger players as well. And then a couple of us are growing at 7-ish-percent, and you'll find who might be losing share and who might be gaining share.

Okay. Maybe moving over to software. Can you just walk us through -- you've got strategic initiatives related to the LC Empower Software. Just kind of walk through your software strategy today. I know you've had some key initiatives. What kind of impact it's had thus far? And how do we think about what it can mean going forward?

Yes. So think of our informatics approach in again, 3 dimensions, right? First, we have the Empower base, the LC-CDS, right, the chromatography data system. Roughly 50% of instruments that are plugged in to Empower are competitive instruments, right, and 50% are Waters instruments. So the first and foremost task is to figure out, and we invest a fair bit in keeping up Empower, right? And we basically spent a lot of time upgrading it, improving the user interface, improving the drivers, et cetera, right? I won't get into the technical details, but we spend a lot of time. It was not being monetized as it should have been, right? And so we initiated some approaches, which meant that the competitors who wanted to plug in to Empower had to be -- had to pay a fee, right, more than what they might have done in the past. So that's an obvious first initiative. What is the economic benefit of the investment we are making and can we make sure that it is commensurate with the investment we're making, right? And it's a pretty important thing to do. Second, we said, look, can we develop -- can we help our customers develop insights from the data that they have in Empower, right? And that's a work in progress, and we're working with several vendors who are extracting data and we've been working with them to figure out how again to monetize that for Waters, but also to answer the questions that many of our customers have as they analyze the data. And the third one, which is most exciting is to develop a cloud-ready platform, right? And so this is waters_connect. And we did this first for the mass spec side, because we didn't want to touch Empower. We did it for the mass spec side. We launched waters_connect. And most recently, we launched a quantitative mass application for our tandem quads from our mass specs, which basically processes data 50% faster than the industry-leading software, right? And that's super exciting for our food safety customers. We do test hundreds and thousands samples a day. So they say 50% of the time, have a simpler user interface. We'll take that idea now into environmental and then into pharmaceutical uses before the end of the year. So a very exciting road map for informatics, but it starts with monetizing what we have. And we weren't doing that sufficiently. Again, consistent with the turnaround mentality that we had, let's just monetize what we have and then building value add and then finally the platform for the future.

So how do we see that manifest? So it's obviously stickier. You get a little higher revenues. It sounds like it's stickier. The informatics, the cloud-ready, maybe that drives higher margins. So when you -- besides strategically cementing your position, how does it manifest in the income statement, I guess?

Let me give you a simple example, right? So we were not charging for -- charging enough, in my view, for plugging into a power. If I charge $1 yesterday, I'm charging $5 today. I'm not saying that's what we did. [ For $1, we charge $2 ], something like this. And that's an obvious increase that flows right to the bottom line, right? And so that's first -- when I say economic benefit, that's what I mean, yes? And the second one on the value add, it's a service. The customer wants to process data, how can you build a software that helps them extract data and process the data. And the third one, which is again most exciting, it's a most longer build. I would not get too excited about it yet because the cloud-ready piece, I mean, the customers are super conservative. I mean, LIPITOR's data is not getting out of the premises of Pfizer. It's not going to go into any cloud no matter what it is. I think we have to be patient there. But the first 2 are pretty significant for us. So I just want to give you a flavor on the exact quantitation, but I think the logic is whether the price increase goes right to the bottom line.

And they're in the numbers, they're in your guidance today. Are they upside levers? Like, where do they fit?

A mix, because some of this we didn't know how well it would hit. So it's a mix.

Okay. So we're almost out of time. I'm going to do a quick one on margins. So just one on margins. So you talked about 5% top line gets you north of 50 basis points, 50, 55 on operating margin, but you're investing. So you're showing less than that today. So I guess, how much -- how long do these reinvestments occur for? Like when do you show that leverage?

No, look, I mean, just the math first, right? I mean, basically, just looking at the gross margin itself, we're offsetting inflation with pricing, right? So that's why the gross margin is flat. Then 50 basis points on productivity, 50 basis points on leverage, subtract FX, and that leaves you with 50-or-so basis points for investments, so you can show 20 basis points of margin expansion. That's the math, right? Very straightforward. On the investment time horizon, we see over the next 2 to 3 years, significant investment, which is about -- in this case, if you do 50 basis points, it's for $15 million, $17 million. It's not a ton, but it's good enough to sort of get these initiatives started. And we expect then these initiatives to pay back in that time frame, right? So if I were to compare apples-to-apples, yes, you should see margin expansion in 2025, '26, that time frame. And up to '23 -- 2022, '23, '24, maybe '24, you start to see early stages of turnaround. It again depends on, Dan, how quickly some of these things get traction. Bioprocessing is out of the gates very fast. LC-MS with diagnostics is out of the gates nicely. And bioseparations has just started, and it's in the incubation stage. So I don't want to get too prescriptive, but that gives you the algorithm. Next 2 to 3 years, yes, we are investing. 20, 30 basis points is a reasonable assumption. And if things get traction faster, you'll see the margin expansion faster. And that's not to say that we will not have other growth ideas. I would much rather grow the top line double digit than have it be 5% and give you the margin expansion that you're aiming for. The overall profitability and the overall EPS will scale better if my top line goes faster. That's the ambition.

Got it. Well, we're a minute past. We're suppose to be out of time. I'm going to sneak in one quick last one. So I think you've come on, you've kind of reset the growth rate. You've done a lot of blocking and tackling. You're kind of moving forward with some mentioned growth initiatives. So I think investors kind of see that and they get in your good communicator. Is there anything you think that folks aren't paying enough attention to right now? I mean it's an execution story, obviously, but where would you say people might not be appreciating enough, if that exists?

I think, I mean, in general, I would say nothing is -- for people who followed us, nothing is really misunderstood. I think it's just a question of understanding it better, right? The first thing you need to know is Waters has a resilient business model, right? We get into QA/QC and that machine works, provided we don't take the eye off the ball. We take sophisticated instruments, we simplify them. I talked about Empower already and they can get value out of our informatics software. We have consumables and we have service, right? I mean it's a terrific business model. That's the first and the most important thing. Second, our exposure to pharma is one of the highest in the industry, right? And this whole discussion gets lost in, well, how much do you have biologics and how much do you have this? It's one of the highest in the industry, if not the highest, right? What does that mean? That means we know what the heck is going on in pharma. We know what problems are there to solve in pharma. And it's not a mystery that we figured out that bioseparations is an area that pharma has great needs in and bioanalytical characterization is another area where they have great needs. And so we are very, very deeply embedded with pharma, which is the fastest-growing market. And I think that somehow gets lost in the messaging, right? I mean how much is small molecule, how much is large molecule? Small molecule is not a bad word. 80% of the pharma market is still large molecule. And guess what? Most of the industry scales with that small molecule growth, not for the large molecule growth. So I think those are the 2 points I would say one should spend more time on. Look, it's a resilient business model. It's not going anywhere, great gross margin, highest margins in the industry, great free cash flow and its very strong position if we have the intent, the courage and the strategy to move into these faster spaces, and that's what we intend to do. We're very well positioned.

Excellent. Well, Udit, thanks again. Love talking with you. Have a great rest of the day here. Hope everyone had a good presentation here and hopefully enjoyed the conference and finish strong. Thank you.

Thank you, Dan. Always good.

Bye-bye.