Avantor, Inc. (AVTR) Earnings Call Transcript & Summary

Thanks, everyone, for joining us this morning. I'm Vijay Kumar, the life science tools and med tech analyst at Evercore. A pleasure to have with us Avantor from the company we have Michael Stubblefield, the CEO. Michael, thanks for taking the time this morning.

Yes. Happy to be here. Nice location, kudos to the team for picking a warm-weather location this time of the year.

I feel like we had enough cool weather where the stock markets we're trying to compensate, but on the topic, I mean, fiscal '23, it was -- I mean, I think for tools, we haven't seen seeing this level of underperformance versus the S&P cautioned in the last 15, 20 years. I can't remember the time. But when I look at your peers, I think most of them were contemplating maybe like flattish sort of organic for fiscal '23 for the base business. Avantor, the base business, I think, is contemplating mid-single declines, Michael. Like why is the Avantor sort of the down mid-singles for the base versus like flattish for peers? Is this like a business mix difference for Avantor?

I'd certainly agree with your observation that it's been a pretty dynamic time to be operating a business and certainly the most challenging period that I steered the business through over the last decade or so. And I think your point about portfolios is important. I think each of the players in the tool space has a different mix of products, certainly a different exposure from an end market perspective. And given some of the headwinds and some of the geographies around the world, that certainly plays into this as well. When I look at our portfolio and some of the things that we've encountered, being a consumables-oriented business like we are, which traditionally adds a lot of resilience coming to the supply chain constraints and challenges associated with the pandemic, we've obviously encountered a significant amount of inventory destocking both on the lab side of our business as well as on the production side of our business that has created certain challenges for our business that were maybe overweighted relative to others in this space. But we remain incredibly optimistic about the end markets that we serve with more than 55% of our revenues in biopharma. We think we're incredibly well-positioned. The fundamentals there are unchanged. The supply chain constraints will unwind. The fundamental growth drivers are fully intact. Patient demand is high. Innovation pipelines are robust. New molecules are being approved at record levels, new modalities are emerging, and we have exposure to all of that. So despite the headwinds, I think we've weathered the storm well. And I think if you look at the phasing of how our revenues have performed over the last 12 to 18 months, we were -- we probably started to see some of these headwinds a little earlier than others. And I think if you look at the space here over the last quarter or 2, our business has stabilized. We haven't had any incremental bad news to bring to the Street here for a bit of time. And I see us coming out the other side of this perhaps a little bit earlier than others. So it is hard to compare across portfolios here. I think some of the headwinds are certainly unique. But I would agree, it's been a pretty challenging time to -- for the business.

And I think part of that was your exposure to semi, which is different from some of your peers. How much was this headwind from semis in fiscal '23? And have you started to see a turn in semis?

So I wouldn't want to overweight the impact of the -- our exposure to semiconductors, it's only a couple of percent of our overall revenues, but it has had an outsized impact on our performance this year. Very similar to our Life Sciences platform. The semiconductor market had significant supply chain dislocations a significant amount of excess inventory. And maybe unlike the life sciences space, our customers in the semi space elected to do an inventory reset on a pretty accelerated basis, meaning over a couple of quarters as opposed to maybe over a few years, which meant that for the better part of the year, we've been down 70%, 80%. We're encouraged as we saw our collaboration with our customers on a real-time basis in that space, looking ahead from a forecasting perspective, we definitely caught the bottom in the third quarter. We're starting to see some incremental improvements in the fourth quarter. We're certainly working closely with them to increase the production out of our facilities to support a ramp as we move into 2024. So we'd anticipate most of this being behind us as we end the year, which is one of the bright spots in our business, but it's been down pretty significantly this year as our customers have reset their inventory levels. But again, similar to Life Sciences, the fundamental demand drivers are there. If you look at just the emergence of AI and the proliferation of that, the digitization across virtually every supply chain, what's going on in the electric vehicle market, the long-term growth drivers for the semiconductor space are strong.

Got you. And you did mention you caught the bottom in semis in third quarter, some improvements in Q4. When I look at the guidance for Q4, I think down mid-singles for Q4, and to your point, I want to start seeing these headwinds in Q4 of last year, right? So shouldn't we be at this point in time in annualizing some of those comps and seeing some implementing trends? Maybe just talk about the macro and what went behind the Q4 guidance assumptions.

So coming out of the second quarter, as we kind of recast the balance of the year, the approach that we took, given a short order cycle business like we run, we don't have the benefit of having a substantial order book to back our forecast or our guidance over a multi-period basis. So the approach that we took was to essentially take the exit rates coming out of the second quarter and extend those through the back half of the year on a daily run rate basis for our laboratory business and then we basically took the production volumes that we achieved in the second quarter, and we cut and paste those into the third and the fourth quarter. As evidenced by our third quarter results, that approach in a static environment seemed to serve us pretty well. And we're a couple of months now into the fourth quarter, and that also seems to be holding up well for the fourth quarter, which is just another way of saying we expect the business to be relatively stable in the back half of the year. We're anxious to see an uptick in the order rates, which would inform the inflection point and see us coming out of the trough here. I would say the sentiment with our customers seems to be improving. And I think there's a number of things to be optimistic about, but we've not yet seen a change in that order rate, which would inform kind of a full-fledged recovery. But it feels like it's around the corner.

Got you. And I'm going to switch my questions a little bit here, Michael. I think in the past, you've spoken about engineering, growing activities as being a leading indicator for customer activity or future sort of orders. And that's been gone on for a while, right? I think why are these engineering growing activity is not translating to orders?

Yes. So under the heading, I mentioned earlier around just a number of things to be optimistic about customer sentiment being one of those, the reported inventory help from our customers being another -- the engagement with our customers has been super strong. Certainly, our commercial intensity is as high as it's ever been. And one of the leading indicators that we do look for in our bioprocessing business, particularly around our single-use portfolio is the engagement with our customers on specific opportunities. And the first thing that comes from those engagements is the translation of the requirements to an actual drawing and ultimately to a prototype that we would -- that we would share with them. And so we like to see high engagement in that part of the business. That's where the activity starts and ultimately before you can get an order on a new program, you need to agree on what the specifications and the design will look like. So that's something that we've been tracking historically. It is one of the leading indicators in our business. But similar to what we see across the rest of our portfolio, the procurement cycles appear to be more elongated. And I don't know if that's just more of a cautionary posture by our customers, but things are certainly taking longer would have -- based on historical data, would have expected based on the level of activity we see to have that already translated into an uptick in the order book. That cycle for whatever reason appears to be kind of bucking the historical trend and hasn't translated into a significant increase in orders yet, but we're encouraged by the engagement and the level of activity that we have with our customers.

Got you. Maybe related to that, one of your peers recently mentioned book-to-bill for instrumentation was north of 1x. And I think the Street perceived that, took that for the most part to be a positive data point on a bottom for instrumentation. I'm curious, when you look at your instrumentation business, any change in order trends or have customers signaled anything?

Yes. So much to what I said earlier, we haven't really seen a substantial change in own order rates across our business. We have a relatively short-order cycle business. Probably more than 50%, 60% of my business is going to operate on a book-and-ship type basis. A substantial portion of our revenues come in and go out the next day. And so we don't really have a lot of forward visibility. But whether it's in our bioprocessing business or in our instrument and equipment portfolio to servicing our lab customers, we really haven't seen a significant change in orders here over the last couple of quarters.

Understood. And you did mention you recently visited Asia relative to some of the macro noise we keep hearing about, maybe any learnings insights for you, Michael, coming out of this Asia trip?

Yes. So what Vijay is referring to is before we started the discussion, this one, I mentioned to him, I just got back from a trip to Asia. For us, Asia is roughly 5% of our overall revenues, which probably doesn't tell the full picture. When I look at the 2 parts of my business, the lab side of the business or the production side of the business, we're relatively underrepresented in Asia on the lab side of the business compared to our positioning in the rest of the world. And that's really what drives this kind of 5% exposure. On the production side of our business, though, we are incredibly well positioned when I look at our positioning in bioprocessing, for example, the work that we've done historically to get ourselves positioned with the Western OEMs who have set up production hubs in Singapore, for example. We have a very significant presence there. Maybe you've seen some of the headlines this year. We've opened up a manufacturing center there. We have a GMP distribution center there to support those customers. Similarly, in Korea, where there's a substantial base of business there, we're incredibly well positioned and have historically had strong exposure to the key customers in that region. We've got a pretty compelling footprint in India, multiple manufacturing sites, hundreds of sales reps on the ground pretty well positioned there. And I think we're excited by some of the trends we see in India around things like biosimilars. And then you get to China, which is why I spent most of my time here on this recent trip, prior to the pandemic, we opened up an R&D center there specific to support the collaboration with our customers in China. The RIM Bio acquisition that we did in 2021 was specifically meant to give us a manufacturing footprint in China for China. And although China today is a relatively small base for us, we do think long term, it has a lot of growth prospects. China generally has missed the monoclonals wave. But when I look at their level of activity on cell and gene therapy, certainly, it's something to be excited about. There is a tremendous amount of activity there. There's some -- a lot of technology being developed there, a lot of collaborations that we're involved in. And I think our strategy for China is playing out well and that we're investing with local capabilities to be able to serve China on a China for China basis. And kind of quietly here, we'll continue to develop that. And at some point, that will be a more meaningful part of our business. But I like our positioning in the region overall within bioprocessing. You mentioned semiconductors earlier. We have a pretty significant footprint in Taiwan to support the strong end markets around semiconductors in that part of the world as well. So from a production standpoint, well over 1/4 of my revenue comes from the Asia region.

And just sticking on with the region of China. I think for some of your peers China, minus 30, minus 40, some pretty dramatic numbers. Curious when you met with customers, whether it's the macro funding challenges, et cetera. Are we at a close bottom to these macro pressures in China?

Yes. I wouldn't be naive enough to think that I could call the bottom in China. There is considerable headwinds there. I think if you start with the macro economy in China, it's clear the consumers are nervous and the spend levels, certainly are being constrained. I think the government has tried various stimulus efforts over the last year or 2 that have had maybe less success or impact than expected because I don't think the consumers have the confidence to return to the market. So it is a tough macro environment, and it's not clear what it's going to take to push them out of that. When I look at the life sciences end market, there's still a lot of activity around the biopharma space, particularly around cell and gene therapy and a number of things to be excited about. So we've probably done a bit better than our peers broadly speaking in China. But it's -- obviously, it's a relatively small part of our business with only a couple of percent of our revenues coming out of China.

Understood. Maybe switching gears to biopharma. That's the largest end market more than 55% of their revenues down high singles in fiscal '23, is fiscal '23, the bottom, Michael, for biopharma? How should we think about sort of that customer base? Different companies have given different reasons for why that market has been so challenged. But from Avantor's perspective, what have you heard from your customers and why that market has been so challenging?

Yes. So I mentioned a couple of times already, the reasons to be optimistic. Certainly, biopharma is one of those. The fundamental growth drivers are fully intact. There's an incredibly healthy pipeline. We're excited by all the new modalities. And it's important to understand, we're agnostic around the modalities. Our portfolio and our offering is going to be relevant in monoclonals, cell and gene therapy, GLP-1s, mRNA, ubiquitous presidents here really gives us a pretty strong position in a really exciting growth-oriented end market that has had some headwinds in it this year. We play on both the early phase research and discovery part of the business as well as with specified materials into their production activities. And both parts of those activities have been down in that mid- to high single-digit level as you suggest. Certainly, one element of that has been the inventory destocking given the heavy consumables portfolio that we have, both in the lab as well as in production, there was a significant excess inventory coming out of the COVID period that is impacting the business. On the research side of the business, we definitely see -- we started to see it probably in the second quarter, a more cautious approach to spending, reprioritization of early-stage pipelines. Certainly, the level of inflation that our customers have taken on over the last couple of years, the pullback of COVID revenues has challenged their profitability a bit. And so you see them going through a little bit of a reprioritization there. There's a lot of discussion around IRA, I'm not sure we could pinpoint a direct impact on the business from that, a lot of folks still trying to wrestle with what it means to their business. But I think it just adds to the little bit of the cloud over the business through the year. Biotech funding was a concern that crept into the business in the latter parts of 2022, we saw that pretty well bottom out by the time we got to the first quarter into the second quarter. So for us, these various headwinds and maybe the reset of kind of position within our customer base has been stable over the last couple of quarters. And that's one of the things that gives us some confidence here that we have caught the bottom. When we talk to our customers about expectations around '24, I think generally, they're all very positive about the trajectory. And one unique factor to think about for Avantor is related to this inventory destocking phenomena that I've mentioned -- because I think it is a little bit unique for us just given how much of our portfolio is consumable driven. There's been hundreds of millions of dollars of inventory that our customers have been working down over the last year or 2 that is actually masking stronger underlying demand for our products and what's showing up in my P&L. We recognize that during the COVID times, obviously, when we booked the revenue, a bit of a dry spell here as they're working it through, but I don't actually need a fundamental change in underlying demand for me to start to see a recovery once our customers work through that inventory. So that's certainly one of the dynamics on the -- particularly on the research side of the business and in our single-use side of the business. There is still, I think, some inventory normalization that our customers are doing of their own inventories which I think is a reflection of just their interest in cash generation as well as the restoration of just confidence in the value chain for all intents and purposes, the supply chains, mine, my suppliers through to our customers are essentially operating normally at this stage. You may have the odd friction point here or there. But generally speaking, lead times have been restored to prepandemic levels. And so the confidence in the supply chain, I think, is intact. And I think emboldening CFOs at our customers to look at resetting their inventory levels as well, which I think generally is impacting the bioprocessing space, at least temporarily.

Understood. And some helpful commentary there. You mentioned customers are positive about fiscal '24. I think you guys have done a pretty good job about conducting a pretty large swath of your biopharma customers. What did the latest survey and any learnings out there, which is giving you perhaps some signs that customers feel good about '24?

So we spend a lot of time talking to them just about sentiment around demand and spend levels going into next year, and that feels pretty green across the board. Given our unique exposure to this destocking issue, a lot of discussions in serving data gathering around just inventory levels. On the third quarter call, I think I mentioned for the first time, we didn't have any customers signaling that they had inventory in excess of a year of target and the overwhelming majority put themselves in kind of a 0 to 3-month bucket of excess inventory, which -- that's great. At least it feels good and it maybe signals that the bottom is here but we need to see that translate ultimately into orders for that to be really meaningful. But I would say that just the activity levels at our customers is strong, we have been really, really deliberate about intensifying our engagement with our customers. And you see that showing up in a number of places. Certainly, we've talked a little bit about the engineering activity to support our single-use business. But things like the work we've done in our academic end market. We're clearly taking share there over the last number of quarters and the investments that we're making in those -- that commercial intensity be digital infrastructure that we've built to be able to make our platform more searchable to be able to do more targeted, customized marketing campaigns to our customers, certainly driving significantly higher levels of traffic across our sites compared to our peers, leading to higher click-throughs and conversions and ultimately more customer wins. And so the commercial intensity that we've applied here has been intentional. And I think that's also part of the picture that we see playing out in our business. I mean I think the setup here is positive. And the -- it's a little bit like a wound spring here. I think when the market recovers, I think we're well positioned to take advantage of that.

Understood. And bioproduction, and I think there's been a little bit of a debate with some companies saying book-to-bill turn positive in third quarter. Other companies saying book-to-bill was still sub-1x. Does book-to-bill matter for your bioproduction business and where are we on book-to-bill trends?

Yes. It's not a metric that we've typically cited because it's pretty volatile just given the breadth of our portfolio and the distribution of lead times across that portfolio. So we found that it doesn't there actually give us probably too much to work from there. We focus more on just the order book itself. And when I look at the order book trends over the last several months or several quarters, it's been relatively stable. We do have a relatively short order cycle business. And so once we see that order book start to turn, it will translate into revenue pretty real-time. So we're following that more closely. And as we sit here today, we've not seen a meaningful step-up in order books, just to be clear.

Understood. And you did bring up destocking in a few times. What gives us the confidence that what we're seeing in biopharma is all about destocking, Michael, when you look at that order book, is this something about daily run rate business, which is giving you the visibility like what we're seeing right now is all destocking and -- sorry, go ahead with that.

So a couple of things. First, when I look at just biopharma at a macro level, a super exciting time to be part of that ecosystem. When I look at the science that's being developed, the innovation that's in play here, the emergence of new modalities, GLP-1 is maybe a topical example, maybe less relevant for the space at large, but as a specialty chemicals, process ingredients, manufacturer, this is actually very relevant for us. And I think one of the differentiated opportunities to play the GLP-1 space through a company like Avantor who is going to have content both on the synthetic route as well as the more biologic fermentation route. And so the pipelines are robust. We follow things like new molecule approvals very closely. That's an important growth driver for our business. Approvals in 2023 are trending above historical levels, the mAbs pipeline, which has driven the bulk of our revenues historically, continues to be very robust. Cell and gene therapies, I think you mentioned on the third quarter call, we'll have more approvals on that this year than the last 5 years combined. So you see some momentum picking up there, and it's been a strong growth driver for us. We've been growing our gene therapy platform well into the double digits throughout the year. And so the fundamentals in this space are fully intact, and I really like our positioning. We've doubled down on investments in the footprint. Our innovation engine is running extremely well. We're excited at our Investor Day next week to highlight for you some of the things that we're working on and the impact that innovation is having on our business. And so we're continuing to extend our positioning across the space. And in bioprocessing, this is an end market that we have historically outgrown by, call it, 300 to 400 basis points. And I think you see that even in the numbers this year. You talked about biopharma being off high single digits. Our bioprocessing part of that is off mid- to high single digits. And I guess we don't have maybe the same level of exposure to China as maybe some of my peers. But I think when you even adjust for that, our performance this year is going to be best in class and continuing a nice string of -- continuing to outperform the market. When I look at the current dynamics, you mentioned the destocking, we can work with our customers on trying to get better visibility into that and the surveying is one way to gauge where they're at in that cycle and the majority of customers are saying they're within kind of 0 to 3 months of their target inventory and so that's certainly one thing that we look at. The patient demand, though, on the other end of this has been extremely strong. And so where a customer might be in their inventory journey is probably unique on a customer-to-customer basis, but no doubt it must be coming out because the doses are being produced, customers are certainly taking advantage of the therapies that are out there. So the end patient demand is extremely strong. The pipelines are robust and our positioning is secure.

You did [indiscernible] GLP-1s. Is that -- any way to quantify what would it mean for Avantor?

It's a little bit early innings. I think everyone is trying to figure out what this is going to mean, what label extensions are going to come with the therapies where all -- they'll be able to take these, what the dosing levels are going to be. And those obviously all have impacts on how much material is ultimately consumed. It is important to recognize though that whether it's a synthetic route or a fermentation route, our materials are relevant. We -- it's obviously a fairly concentrated customer base at the moment that we have good access to. So it's an exciting area. And just another data point of we've referenced recently this idea of working the golden age of science, and this is just one of those examples of why it's so exciting to be positioned in this space with a portfolio that's as broad and relevant as ours is.

Got you. And from a revenue perspective, it is looking at prescription volumes. Is that the right way to look at it or perhaps production volumes? What's the revenue model?

Yes. So I mean, production, we're going to be more linked to the production volumes that come out of our customers on this is probably not a bad way to think about it.

And then switching -- moving away from biopharma. You did bring up share gains within government, education. I think there's been some sort of concerns, maybe pricing was a tool that enables share gains. Like is there any basis for those statements, Michael?

So we love the positioning that we have. We have an extremely broad portfolio. I have a supply chain that gives me the ability to meet the agile requirements of our customers. And particularly in the academic environment, the digital capabilities that we have built over the last several years has really strongly positioned us there. And then you overlay that with just a very focused commercial intensity that we've brought to bear that has resulted in the share gains that you've talked about. When I think about price in our business, this is certainly one of the drivers for our margin algorithm and the academic market is probably one of the more price-sensitive end markets. They tend to be a little less brand loyal than, for example, a customer in biopharma, their budgets tend to be a bit more fixed. And so they tend to be a little bit more aggressive and willing to shift if they can satisfy the requirements somewhere else. So I think the propensity to make a change to find value elsewhere, combined with the intensity that I'm applying has kind of come together at the right time. But we watch price over COGS extremely closely in our business as it is one of the important drivers of margin in our business. And I would say the price over COGS differential that we would normally expect in our business, pre-COVID, during COVID and even now, has been very, very steady. And so it's something that we think that we're quite good at. We bring to bear some pretty sophisticated AI-driven algorithms to help us monitor these things and help guide the price actions that we take in the market. We think we're disciplined in that regard. And I think we've done a nice job managing kind of volume and price and margins in that regard.

And since it brought pricing, what does fiscal '23 contemplating from a price contribution? And I think some of your peers have made comments about pricing still remaining positive in '24. Is that a reasonable expectation for Avantor?

Yes. So on 2023, we're probably getting 300 to 400 basis points contribution from price across the portfolio, which is probably close to 2x what we would get in a normal year, which, again, is just a reflection of the hyperinflationary environment that we're in. It obviously hasn't led to outsized margin expansion opportunities for me, but it certainly has enabled, I would say, a normal contribution to margin based on that dynamic. Pricing has been relatively constant throughout the year, which has been, I think, a welcome dynamic for our customers. The pace of price increases, say, in 2021, 2022 was a little bit untenable and led to a lot of, say, transactional inefficiencies with our customers. And so I think our customers have been satisfied that we've been able to keep things relatively stable throughout the year. Now as I think ahead to 2024, obviously, we kind of sidestep the question on the Q3 call around guidance for 2024. We're not planning to give guidance until we follow our normal cadence, which would be at the -- on the fourth quarter call. But one of the variables when I think about revenue guidance that we'll certainly have to take into account is where do we see price landing. Is it going to be closer to what we've experienced over the last couple of years, which is that kind of 3% to 4% level? Or does it start to move back towards a more normalized pricing environment of, say, 1% to 2%. And I think the driver for that will ultimately be the call that we take on inflation. We're right in that process, actually as we speak here of working with our suppliers to get their price nominations and to understand their aspirations on price. We're pushing hard to try to contain inflation best we can. And we'll see where that lands here over the next couple of months, and then we'll take pricing to our customers to make sure that we protect our margins accordingly. But it does feel that just from where we sit today that there's still going to be a reasonable level of inflation in the business again in 2024 that we're going to have to protect through price. Yes.

Should -- when we look at broad inflation indices, right, the PPI or CPI, is that like a good barometer for judging what inflation within your end market should be?

It's not an index that we necessarily look at all that closely for us. It's -- what are our raw material suppliers and our third-party partners, what are they looking to do? And then what's the labor inflation? When I look at the heavy SG&A model that I run with hundreds of facilities around the world running distribution centers, position inventory close to our customers or the 30-plus manufacturing centers we run or the 13 R&D centers that we run, 5,000 customer-facing associates. Labor inflation is important to us. And heavy exposure to Asia, even though I may only have 5% of my revenue there, we probably have 1/3 or more of our associates are in the Asia region, significant presence in India with a lot of back office functions as well as our footprint, heavy exposure into Europe. So labor inflation, which has been running at elevated levels over the last couple of years is an important variable. And then what's the input costs to our business is something we're looking at pretty closely. And then the model historically has been, whether it was high or low inflation, don't matter. We do have a culture of continuous improvement in productivity, all underpinned by the Avantor business system, strong, strong DNA within the company at offsetting inflation with productivity. And we've had to kind of overplay that card in 2023 and you should expect that we'll do that again in 2024. Brent will share some details with you about our thinking around productivity next week when we get together for our Investor Day. But that's another tool in the toolbox here that I think plays well to our culture and our DNA and our capabilities. And I think we've got a very, very strong track record of driving value by also keeping a close eye on costs. And particularly in this environment, I think that's especially important.

Understood. And then on your advanced tech and materials, which is about 1/4 of revenues, the semi is low single digits of total company, it's probably bottomed out. But what's the remaining exposure or cyclical exposure within advanced tech materials? If there is a recession, how should we think about sensitivities?

Yes. So the applied part of our business is about 25% of our revenue, as you suggest. Broad strokes, the way I think about it, about half of that is going to be in more a cyclical or defensive-oriented type end markets and then the other half, maybe 10%, 12% of our overall revenues would be in more cyclical applications like oil and gas, petchem mining, these kinds of end markets, which tend to follow GDP relatively closely. The offering that I take into those end markets is really the broad lab portfolio of chemicals, reagents, equipment, instruments, consumables that go into their QA/QC function associated with the testing of their finished goods. And so it's going to mimic the output there in those end markets pretty closely, which lends itself to kind of a GDP-type view. And so I think where we're at now, you see kind of flattish to low single-digit performance for most of those end markets. The reason that the applied markets for us overall are off as much as they are as the 70%, 80% headwinds that we've been experiencing in semiconductors, which fortunately, we found the bottom, we're turning our units up again and the outlook from our customers there is pretty promising.

Got you. And my understanding is within semis, like customers give you a 6-month lead time, what have -- how conversations with the customers evolved on order book?

So it is a very concentrated space. We have more than a handful of customers, but there's probably only a handful that really drive the significant volumes. And so we're able to have kind of weekly supply chain connectivity with those customers where they're openly sharing their demand expectations. And particularly in a time that's dynamic like we're in, they're particularly concerned about how quickly I can start to turn my units back up to support their outlook. And so we're very well coordinated with them. They are working hard with us to make sure we can secure all the raw materials that they need to support the ramp that they have in mind here. We started to turn the units up in Q3. We'll ramp those fully by the time we exit Q4 here, and we should be back to relatively normal demand environment moving into 2024.

Understood. I think moving on to more near term, your most recent earnings call, I think the cadence for Q4 on the gross margin commentary, I think, was something which was -- caught by surprise, is the simplistic way I'm looking at your dollar revenues for Q4 are in line with Q3, it -- so your production volumes should be at similar levels. So why would Q4 gross margin step down? And I think some people would point out perhaps it's pricing. Maybe just talk to us about the gross margin dynamics.

Yes. So I think we highlighted maybe just 2 or 3 factors driving the step down in margins moving from Q3 to Q4, probably half of which are, I would say, onetime in nature. With Brent coming in, certainly looking to end the year from a working capital perspective in a pretty strong position. So some of this is just under absorption associated with us, taking some pretty deliberate working capital actions on turning back our units to manage our inventories of our products, a little bit of inventory cleanup that Brent is working on. And then not really a price dynamic so much, but the strong growth we are seeing in our academic sector, traditionally speaking, that is an end market, as I referenced earlier, it's a bit more competitive, where the margins relative to other end markets are a little bit lighter and you certainly see the influence of that. So those are probably the 2 or 3 key drivers that we see impacting Q4 relative to Q3 and probably half of those, I would say, are onetime in nature.

Understood. Talking about the bright spots. I think free cash that's been a bright spot. You guys are -- I think you guys are the highest in life science tools, 95% free cash conversion so far. Is that sustainable when you think about fiscal '24?

It is. I mean this is a cash flow generation machine for sure. This business throws off a tremendous amount of cash. It delevers quickly in a normal environment. We're also growing EBITDA. And the conversion that you mentioned, 90%, 95% should be the norm for this business. We were a bit over 100%, 110-plus percent in Q3. But I think over the cycle is probably more 90%, 95% is the right way to think about it. It is a very capital-light model. We can support all of the maintenance, growth aspirations, productivity and investments in our business by investing, say, 1% to 2% of revenues. And we've probably been investing at the higher end of that, 1% to 2% over the last couple of years to strengthen our supply chain. But that algorithm has been pretty fixed over a pretty long period of time, and it does lend itself to significant free cash flow generation, particularly when you can manage your working capital. We were in a phase there where inventory levels were quite low, service levels to our customers were quite low as we work through the pandemic. And as you look to restore inventory levels to better support the service commitments to our customers. We had a period there where we were overinvesting in inventories. We've been able to arrest that trend for the most part of this year and now will start to operate more normally. So you should see, I would say, from this point forward, pretty consistent free cash flow generation for the business.

And maybe my last question here. Should we expect any margin expansion for fiscal '24?

So the margin algorithm for Avantor is fairly straightforward. It's price over COGS, which that part of the algorithm has been fully intact throughout the pandemic. No reason why that won't contribute at normal levels next year. Mix is important. And mix for us is the ratio of proprietary to third party. And that proprietary is primarily bioprocessing, our biomaterials platform and semiconductors. Biomaterials has been running well, semis turning the corner. As soon as bioprocessing comes back, we should get a normal contribution from a mix. And then the third component is operational leverage. I need volume growth to cover the fixed cost of the platform. And when that happens, you should see an outsized rebound in our margins. And then we'll take some self-help measures here around productivity that should also be a contributor in '24.

Fantastic. With that, I think we're out of time. Michael, thanks so much for spending the time with us.

All right. Thank you, Vijay. I hope to see you all next Friday at our Investor Day. Thank you.