Oxford Nanopore Technologies plc (ONT) Earnings Call Transcript & Summary

Thanks, everyone, for joining us. My name is Mike Ryskin. I'm on the Bank of America U.S. Life Science Tools and Diagnostics team. And we're excited to be joined for our next session by Oxford Nanopore. We're joined by CFO, Nick Keher. Nick, thanks so much for being here.

Thank you very much for inviting us. It's a pleasure to be here.

Maybe just to kick things off, sort of our standard question is, you reported first half results fairly recently. Strong results well received. Nice follow-through on the stock price after that. You want to talk us through sort of the key takeaways, the key message from the -- based on the quarter -- half payout?

Yes, absolutely. So, we clearly, at the beginning of the year, we guided that we were going to have this 45%, 55% kind of step-up in revenue. We guided to a 57% gross margin for the year. We got it to kind of like a number of things happening during the course of the year in terms of contract wins. By the time we got to the reporting September delivered GBP 84.1 million of revenue, so 12.5% underlying growth, which was pretty much towards the top end of where we guided we would be. The gross margin came in at 58.8%. So quite a good beat against what the guidance was there. And in terms of like key factors during the first half, we announced that we signed this 10,000 genome project with PRECISE, which is very excited about. We had said that the NIHR contract we had last year, 22,500 samples, that started to kind of ramp up quite nicely as well. We've seen encouraging utilization increases on our large P24, P48 devices, so up over 20%. And placement rates with beginning to kind of return to normal. So whilst it's a very tough market environment out there, we've kind of continued to kind of take share and grow organically. And actually, as we laid out the results as well, we've now kind of got this new market and split for our customers. So I think that's been well taken by the market. It's giving people a bit more confidence about the trajectory we're going in market share, the way we're gaining it. And it will lay the bread crumbs for how we kind of get to that above 30% growth underlying next year. And in the second half of this year and for the next 3 years as well. The other -- I mean, there's always a multitude of reasons. I wouldn't even pretend to know why the share price goes up or down from one moment to the next. But those things all kind of were beneficial in helping we've rebased the expectations, we've hit the expectations or beaten. And we've outlined like our end market in a bit more details so people can understand how we get the breakeven in '27, cash flow in '28. And on top of that, we did talk to the fact that we are looking at -- we're in the process of how we think about the step up to the main market and potentially FTSE 250 inclusion, which in the U.K.. For U.K. investment managers is kind of a big thing as well. And that clearly all these things combined with some more buyers than sellers has actually helped the share price.

Yes. That's a great overview. Maybe just start on sort of the second half outlook. You talked about how you expect that the first half, second half split, and you're trending well towards that. But can you go into a little bit more detail on sort of what are the drivers underpinning that confidence in the second half, a couple of different factors there.

So we're at the beginning of the year, and this is exactly the same message we'll give now. We kind of triangulated how we would kind of see the year play out. We saw at the end of last year that device placements weren't as strong as we'd like them to be in Q3, Q4. And anybody who follows our story knows that the majority of our device placements are done by an OpEx lease. And that means when we place a P24, P48 actually, we have 9 to 12 months visibility in terms of Flow Cell shipments. And that means we have essentially an order book, but essentially we're going to -- we know when we're going to deliver it. Because last year was weak, we knew that the first half of this year was going to see a lag indicator of that, particularly in the U.S. Actually, the device placement in the first quarter we knew were a bit weaker, but we could see utilization was increasing. And that's the best organic metric we could have. So the year, we knew if we kind of placed out the same device numbers as we saw last year with a bit of haircut at the same utilization rate just by critical mass. We knew we were going to see this kind of second half bump. And that's kind of what's seen come through, but utilization has kind of increased a little bit ahead of that. We said that if things remained weak and our utilization kind of took that downturn, actually, we've got these big contracts. These bespoke contracts like we knew PRECISE was in the wings, NIHR we knew it hadn't ramped up yet, but they could help make up the shortfall. And the other metric that we're kind of guiding people to is our sales force like efficiency. So -- we've spent -- since IPO, we've tripled the size of the commercial infrastructure and doubled the number of heads that have a target on the inside sales and sales. And we've done that clearly a very difficult time during the end market. But what we know is that in 2022, our average orders per head was up here. And it's trended down because the dilutive effect of adding all these heads. What we can see, we saw in the first quarter, it started to turn because these people are starting to get tenure of 6 to 9 months. And like anybody in a new job, it kind of takes time for them to bed in, get to know the technology and then they start to kind of execute and particularly in the U.S., again. Q1 '24 to Q2 '24 is a 30% increase in average orders per head. That trajectory is kind of maybe continuing, not that kind of same clip, but it's continuing into the second half and the way I think about this is orders per head essentially is going to lead to revenues per head. Conversion rate is very strong. And we're kind of seeing that play through. And the encouraging thing for the medium-term view is if we deliver the same revenue per head in '22, in this year, we don't just deliver this year's guidance number. We delivered next year's guidance number. So it's about the sales force kind of bedding in, becoming more established, and that is we're starting to see the fruits of that. And without a doubt, we're going to have to -- there's always a case of kind of making changes here and there and filling in some gaps. But actually, that's kind of beginning to come hold. At the same time, the encouraging utilization we've seen on the larger devices. So that means our customers are essentially getting more comfortable with the product. And that means they're kind of pulling through more consumables, which are higher margin for us as well. So some encouragement essentially on that piece, but that's why the second half we've got that. And so we talked about -- we delivered ex revenue, GBP 84.1 million for the first half. We have a schedule -- because that we got 9 to 12 months visibility. So we've got scheduled book of business. And we've got a run rate essentially like the underlying pull-through from existing devices that are out there. And then we have an opportunity funnel, which by sizable potentially is all of our converting as we think about that, that opportunity funnel is going to be where it's going to position us inside that range of 20% to 30% underlying growth for this year. And what we've said is opportunity funnel gets us to that midpoint essentially as well.

Okay. And there are also -- in the earnings call, you also discussed some of the dynamics within the various product lines and also between some customers. Anything that stood out there in terms of what's doing a little bit better versus a little bit softer?

Yes. So clearly, I think everybody has seen regionally, I should start regionally. We actually delivered double-digit underlying growth ex currency in China. China has clearly been a headwind for everybody out there, not the same sort of headwind for us because we've just seen greater utilization by customers pulling through. But it's been a headwind against where it could have been as I think as for everybody it's a biosecure, difficulty in placing larger devices into the market. . Regionally in the Americas, we've kind of had that lag from lower number of devices placed last year. First half of this year, certain months, they were now beating Europe. So there's a bit of healthy competition dynamic there, which we're hoping kind of plays through at least that Americas growth as well. Then if you think about customer groups, this is why we split it out. So 70% of our business now comes from the research market, we call it. Those are customers that are funded by government bodies, public bodies, grant funding, things like that. It includes our distributor network. These people are funded for a novel science, research base. It includes the whole genome sequencing market. That's clearly had a different dynamic in terms of funding and whatnot to our other segments where applied industrial biopharma, which is splitting out because we think it will be a big growth driver on its own and clinical. So within the research space, we continue to take share that PRECISE contract showed. And actually, we think we've got quite a lot of building opportunity here, particularly because of our telomere-to-telomere workflow, which is seeing some encouraging signs of adoption, particularly in APAC. That means that we might get a bigger pie there in the whole genome sequencing space. Then within the applied market, plasmids, something simple as plasmid sequencing and our technology being really applicable to it is actually delivering quite substantial growth. So this is a very well-established market, a 50-year-old technology, but we're actually quicker, cheaper, more accurate. So essentially, those kind of benefits to our product grades are going there has meant that we've seen service providers start to adopt it at scale. And so, Plasmidsaurus, great name. They've got a multiyear, multimillion dollar contract for them in the U.S. But they've got a place in Canary Wharf now. And they're actually starting to branch out across Europe. They'll go into APAC. They're a great customer, but an exemplification of work and a new body wants to kind of take share from the existing parties out there, and they've created a multimillion dollar business not overnight, but very quickly. And they've done it by [indiscernible] share and plasma sequencing, this is great for us. Here's all the kit, you guys go about it. And they're starting to make the other larger, more established service providers now go, we've got to start doing this as well. It then has made the biopharma customers. We actually do a lot of this, maybe we should be doing this as well. And so they're starting to evaluate the technology. We're moving from that research evaluation phase adoption in more commercial volumes, as we say. And what they're also doing is well, what else can it do? And the two big other things that people look at is sterility testing and mRNA vaccine production where these kind of 2 of the workflows are gaining a lot of traction. And it's very -- this is early days, only 9% of our revenue, GBP 7 million in the first half. It's not a big number, right? It's lower and smaller numbers time. It's not huge, but these contracts that we're looking at a multimillion dollar in size per customer per facility. And so we're kind of looking at this and going, well, if we place out 10, 20 GridION to Grid Q, it's a lockdown version per test basis, and the adoption is kind of coming through. We have a London Calling Event, and in fact no one is talking about -- Boston, Thursday, Merck, BioNTech, big companies talking about as well, kind of seeing this kind of critical mass where they've all -- we've had hundreds of customers who are evaluating research and now taking it going actually, now we've got a lockdown version. We can validate this. We can put stick on it that says we don't need to validate it again from the 3 or 6 months, and we can use it in our workflow. And I mean one of the best things about this is we're not going head-to-head against the conventional NGS players. We're going up against either a 50-year-old legacy technology or we're going up against in the case of our mRNA, no sequencing is done there because they take pictures. Essentially, we got the modifications because taking live RNA. And that means we're going up against 8 orthogonal tests where people having to spend millions of dollars to install the equipment to get approximation, whereas we can say, well, we can do that for 24, 48 hours. So we work with Lonza on that as well. And then we're quite excited.

A lot there. I want to follow up on one of the points you made was -- you're talking about competition both in the first half and some of the big program wins. Can we talk a little bit about the competitive landscape and what's going on in the market because, yes, I mean, as you called out, it's a very tough market environment, but you seem to be handling it better than others. So both from the long-read perspective of PacBio and the bigger players in short-read sequencing. How is the competitive landscape evolving? And what's enabling you to gain some share?

Yes. Look, it's evolving is the right word. And I'd much rather -- I think we all would be sitting here going, actually, the end market is so buoyant that everybody is doing very well because that's what all of us want to say. It's not nice. We're seeing the companies having tough times because we've all been through them. So it's -- but it is evolving. In the long-read space, the PRECISE contract, 10,000 samples, we think that's just the start where in APAC we've got quite a few opportunities that are coming down the pipe in that T2T piece. So now you can use, I think, five Flow Cells, PromethION Flow Cells into T2T on our device means that you don't need quite a few other devices. So there's a price play economic pull and as CFO, I quite like economic pulls by customers because they generally get traction. So that's positive. And that's evolving. Clearly, we're seeing the competition is intense, but we are managing to take churn, maybe weathering better because like for all devices, in particular, it's a different price point. I mean, times are tough, interest rates are high, there's less investment, there's less CapEx going through. We can place a device out with customers at a much lower price point and they can just get going on it. I think that's helped us weather a bit the storm. But interestingly, we are seeing more CapEx purchases now, particularly from biopharma and applied customers. So it's in the mix. In the short-read space, we don't consider long-read, short-read. We're just a sequencing player there. Actually, we're a sensing platform. But in the sequencing space, there's a lot happening in the short read. And clearly, there's a lot of price kind of pressure going down there, just trying to -- well, avoiding that because we can offer people a different product. So we can offer people more data, more comprehensive picture and what the genome looks like. And we can add in that methylation, which I think people started to realize it's actually quite important, particularly from things like oncology. So -- and right now, cardiovascular as well. So there are different plays for methylation that are going to cut stock to be more apparent. So it's definitely an intense market at the moment in terms of competition. But we're not playing into that. Where we are coming up against long-read players, it's usually hold genome sequencing, holding on price, still winning business, and it's not a -- just maintaining our premium product status. We don't really go head-to-head against the other players unless we're talking about applied, clinical, maybe not really in biopharma. And in those niches or maybe in research. I think the fact that our accuracy is now at the right level. Post chemistry changes whatnot. We're in a good place. Actually, we started to kind of see more adoption by customers who get more comfortable with the product.

You mentioned the CapEx difference and the price perspective. But it can't just be the -- any remnants from customers to spend money because like you said some of your instruments are sold that way as well. And that environment should normalize over time. So is it, to your point, on accuracy. Is it -- do you feel like there's a threshold that was passed where sort of enabled some of these projects that couldn't be done a couple of years ago?

So when we -- yes, absolutely. So I think that's definitely been the case. And this will take time to come through in terms of we are there. We've got some legacy kind of noise to kind of flesh out over the next few years, but we are there. And that is just going to be a natural thing that will happen. People will have in build kind of, oh, we didn't like the device for these reasons, but actually tries now and see what they think. And that's kind of gaining traction. I think the fact that we've got more workflows. We're increasing the number of things you can do on it as well. It's also healthy. So it's a bit of everything. But I mean, fundamentally, what do customers while the choose a device in this space because they want the technology, there has to be able to hit certain. We're hitting those criteria now like accuracy. But fundamentally, the technology is what opens the door with the customer, then you're going to have discussions on price. But fundamentally, they're choosing the product because they want the technology.

Okay. You touched on NIHR. You touched on PRECISE already, but I want to dig into that a little bit different -- a little bit further. For programs like that, I mean, you've been involved in some POPC programs in the past, and those have come and gone around the world. Can you talk us through the customer decision process. How long does that take? How long have you been sort of negotiating with these? How are the protocols refined over time? Just how does that evolve? Because that's -- it's a major customer, a major program.

It is -- well, let's, I mean PRECISE, it's a landmark kind of -- we want that to be open light, and it's a great customer. We're really excited by it. It's not like the EGP contract that we have, for instance, that essentially materially impacted the P&L in a good way than a bad way. We're not -- what we're going to see here is small types of deals, so not the 100,000, 200,000 maybe one day. But really, people are going to want to focus. We've got something we really want to look at. We want to understand it properly in detail now. That's where they're going to need a different picture. And that's where we step in. . So these contracts do take a long time to negotiate. PRECISE was multiple months. I think it's fair to say people were hoping it was going to be announced in the second half of last year. We announced in July. So it took a bit more time than everybody would like. There's a big evaluation phase. They do go head-to-head against other devices and they try to figure out what's the best thing they're going to get out of it? And how is it going to work? And clearly, there's a price element as well. So all of these things come into in it and it just takes months.

When that gets announced, do you notice a pickup in your conversations or change in conversations with our customers from sort of a recognition perspective, a validation perspective?

Absolutely with that project, particularly, yes, there's -- I mean, Gordon was out in Singapore a lot long ago, and there's a lot of interest. But we -- it's interesting. The thing that maybe got more of the people excited, London Calling and T2T kind of work. That has essentially kind of started something and then PRECISE kind of building on that as well, where we're also doing T2Ts. So there is growing momentum. And I think for whatever reason, shift to focus on customers geographically. We've definitely seen in APAC and EMEA, a lot of adoption and an opportunity funnel that's growing particularly in that whole genome sequencing space. Not seeing the same thing in the U.S. in terms of like things coming through at the moment. U.S. has been more applied, clinical, biopharma. It's certainly research as well, but in a different way, whereas those larger kind of programs, Europe and APAC seeing quite a bit of interest. .

Yes. I think the U.S. is having a lot of other issues right now. So I think we're a little distracted for that, unfortunately. You mentioned some of the new products. I want to touch on Q-Line and clinical markets. Just sort of talk about the opportunity there? And what's been the reception so far?

Yes. So we launched Q-Line, first of the May. We actually -- we worked really hard to get this out before the end of first half due to customer pull. And the reason for that is a Q-Line, what is it? So got a GridION device -- so with MinION. So my opportunity to pull it out. But that's -- clearly, the MinION we've got the Flow Cell within it. And essentially, the grid is essentially five of those combined into one device that's like this big. Essentially, it's very small but quite portable, but essentially bench top. And the Q-Line is essentially the exact same product, except for the software essentially been locked down. So that now it's out in the market, there will be no revision for set period of time, that's important. So if I go all the way back to when I was in the labs and whatnot. What you'd have seen is [indiscernible] days, you'd have had engineers coming around all the time to revalidate machines to make sure that essentially they were performing exactly as they should. We all know this. So the kind of the measurements are always the same. They want the same thing with the sequences because they don't want to have, oh, hang in a minute, different readings coming out. We have to revalidate machine, costs money. They've been calling out for a lockdown version. We've now got that. That has led to sales of Q-Line and there's a significant opportunity for that product as well, where it's going to be a growth driver, and we're very excited about it. So GridQ, the Flow Cells for RNA, DNA, all of it essentially is getting a lot of traction in that market. We've done the grid as the first Q-Line version, but we will go into PromethION, we'll also go on as well. And yes, that will lead to another boost down the line.

Yes, that was going to be my next question. Is there an opportunity to do the same thing with other platforms? And how much work and how hard is it to create a lockdown version? Is it on the front end, on the back end, software or so...

Software is a big piece, but there'll be the device fundamentally -- device is the same box, but the -- we need to be absolutely sure that the Flow Cell is ready and from a chemistry perspective, manufacturing perspective, we need to be sure that there's no major change that's going to come through. It's going to be stable to the point where actually, we don't need to push the tech any further. . And we know that this is going to be the sort of thing that those markets are going to adopt and grids there with the MinION Flow Cell, PromethION coming right behind it. So it's also ready. We're going to continue to push the capabilities of both PromethION and MinION in the research market. But we think that is ready for a Q-Line and lockdown now as well. And it is not insignificant. The amount of work the teams did to make that got over line was huge. And it's -- at the beginning of the year, we said the investments that we're putting to R&D, less about maybe the early-stage research. There's a really good chunk of money there anyway for them. Development teams, quality, it's the apps teams, essentially them to kind of get the money they need to kind of get these things out on time. And as we look forward, those teams are probably going to be more investment again. But it's step-wise fashion.

Is there -- I almost -- I mean, I know the obvious focus for lockdown platform would be the clinical markets. Is there -- could there be any interest in research more in pharma just for a sort of more straightforward lockdown version?

Pharma, not sure. Research customers potentially yes as well. It really...

In some cases.

In some cases, I mean, without a doubt in some cases.

Yes. Just because there's such a variety in terms of what the customers are looking for.

Yes. And I mean, it's worth perception you to talk about accuracy, that perception is going to disappear, we're there. The perception could also be about variability in Flow Cell and things like that, and that's because they're not -- we don't put a limit on it. We could put limits on it. And if we did, every Flow Cell would perform 100% like expectations all the time. With actually research, we take away the constraint so that people could try and push it to the limit, and that's why they get the variability output. But I like to say, lockdown versions, we'll cap it and then everyone will know exactly where they are with all the time.

Yes. I mean I know historically, that was a lot of the debate or I don't want to say controversy, but debate on the MinION, on PromethION, on the Flow Cells is the variability and read length and Flow Cells. So again, the research setting, it's more acceptable and it's more understood, but there will be applications where...

Certainly. But it's now at that point where we can lock it and cap it. And then those questions kind of go away and people as well. But it's been purposely done that way.

Any questions from the audience? Thrown in there? Okay. I want to hit on a couple of other topics, China. I mean you mentioned China earlier. I've seen some weakness there, obviously, in terms of the broader market from elsewhere in tools and elsewhere in sequencing. You've reported pretty solid underlying growth. Anything in particular stood out that allows you to deliver that?

So in -- the China, it's a very different market to the rest of world that we kind of conventionally see the active service labs that have actually adopted the technology very quickly. And the IVD market is something that we are looking at over there as well. But the service labs that have adopted it, it's that pull-through. So the pull-through by those customers has been very strong. And that's helped offset the fact that we could really place any other bigger devices into that market because they did [indiscernible] piece. We've now got a combined unit and get it in. But even with that, you're still seeing delays. Everything needs to be signed off to go to China. That means every single shipment, there is a risk that it can see a delay or a rejection because it's like an application, you have to put in, and we're seeing that. So it's not the go-go years. I think that we've kind of seen for the last 10. And we're not -- we don't believe it's going to be returning anytime soon either. But it's still been a healthy market for us to your point, still seeing a lot of interest in the product. We're still trying to enable the right people to kind of get the product as well to kind of do what they need to do.

Okay. And one of the hot topics in the space has been China stimulus. It's been debated over the last 6 months -- 6 to 9 months. Is that something that you think you benefit from sort of...

A rising tide lifts all boats. Yes, we would. So I mean even though we've done well, and like you'd have to tell me relatively speaking, how we're doing against peers in the space in China. But I think we've done all right then like if growth stimulus comes back, we'd be a beneficiary like everybody else.

But no indications or anything yet still in terms of orders accelerating from it or?

Let's wait I see. It's not -- we're not factoring it in.

Yes. Okay. All right. I want to touch on margins a little bit. You talked about the first half, second half margin ramp, but then beyond that, your long-term gross margin target, sort of what's underpinning that progression?

Yes, absolutely. So we delivered -- we said 57% for the year. And we've not changed that guidance number. We did 58.8% for the first half. We did say in March that we would see a stronger first half than second half. I think it's fair to say we did a bit better than we thought we were going to do. We had a 120 bps headwind from currency. So ex currency is 60%. It's not that our medium-term target is 62% -- above 62%. So I think we kind of demonstrated to the market that this is possible. A few years away. What's been driving that margin increase, two things. Devices. We've done very well at getting returned devices and recycling them. And what does that mean? So the sequencer, essentially, we can -- depending on the type of device, we can recover quite a lot of the bits that go inside and put it into inventories, use them again. For the GPU to -- the GPU itself, we can also recycle that. And so when we place our P24 we get the GPU back, you can put it to a new device and send it out. So that is a -- that's a great part of the business model. The other piece that essentially helped the margin is even the MinION with the decreased volume that we've seen through because of the COVID headwind in particular, the Flow Cell margins stayed the same. And that's because we can recycle the most expensive part of the Flows Cell. So we can clean the chip essentially. And we are doing, and then we're putting it into a new Flow Cell selling that. We can go around the loop quite a few times on this. So that's helped maintain the margin on the MinION in spite of lower volume, which is a good thing, but for PromethION. So the chip is essentially the same product at a different design, same fundamentals. We've seen yield improvements, and that's driving us to drove a double-digit increase in gross margin. So improving yields essentially is driving a double-digit increase in PromethION Flow Cell gross margin. When we actually start to recycle, we'll see another double-digit increase as well. And we're hoping to be able to recycle to a year or two. That will be one of the things that drive us to that above 62% gross margin overall for the group. The other driver is -- so let's be balanced, talk about the negatives. More whole genome sequencing kind of contracts, these larger kind of programs, they're a drag on the margin but they're very important strategically, and we're going to continue to kind of grow into that market. We know they're got to help. They're going to bring down the margin. We know that we've got a mix impact because more PromethION sales relative to MinION but MinION is a higher margin, there's a mix impact there as well. All these things are just kind of part of doing business, which you're going to see as we grow. However, growth into biopharma, applied and clinical, we're going to sell on per test basis, particularly for the Q-Line and the margins are healthier, maybe lower volume, maybe, but a higher margin. So that will help. PromethION fundamental improving, that will help. And the other piece is just scale. So scale helps yields, but it also helps leverage the infrastructure we've got there. So when we sell a device, we have to allocate which is cost of goods, the human element of setting up the device, some customers to get the logistics or customer solutions, field application, technical support. That infrastructure is set, we'll have to add in pockets, but not majorly. So double the revenue, much better drop through. So we've got a few kind of things that are helping us get there. We think we've been cautious with the 62% greater than gross margin by 2027. Hopefully, the fact that we delivered 60% underlying in the first half shows that we've got a bit of kind of keeping to that.

I mean not to ask you to raise the number, but just sort of hypothetically, what do you see as sort of longer-term steady-state or ceiling margins. I mean industry, 70% from Illumina was more or less their number. I mean, obviously, the businesses are very different from an instrument and a portfolio perspective, but that's sort of the one comp we have out there. Is there a road map to something like that in the longer term?

It depends on mix. So it really depends on mix and I don't have that crystal ball. So yes, that could actually be possible if 99.9% of our business was that applied biopharma. Maybe 99%. If a substantial weighting was towards these customer profiles, then yes, that's entirely possible because the gross margin on the consumables will be good enough to get into that. The mix of device relatives or consumable anyway is going to be a swing factor on it. There's too many variables. But if all the variables worked out in the way we wanted to, then yes, 70% is entirely possible. But I wouldn't want to commit to that because I can't control mix, and we're not going to say no to business.

Yes, just hypothetical. Other parts of the P&L, R&D, SG&A, just sort of OpEx in general. Any -- we talked about a lot of different investments you're making both on the sales force and on the R&D front. How should we think about that in the coming years?

Yes. So R&D, for anybody did catch the results as well. At a total level, we see double-digit increase in our cost base versus the same prior year period, it's first half of 2023. However, sequentially against the second half, it was actually only up 2%. And that's because we have heard the market. We've listened to our investors. We have given, we've made the substantial investment over the last 2, 3 years post IPO, we're taking the feedback, we've taken the pill and we've really slowed that head count increase and be very targeted on where the investment needs to go whilst we're going through this phase of execution. And the execution will come through on the revenue line increasing with -- at a faster clip now than the cost base. And that's delivered 12.5% underlying growth on only 2% increase in SG&A -- sorry, total cost base sequentially. So we're quite happy about that. So the next kind of step forward is, well, in R&D, the split of investment is going to be much more towards development and research. We're not going to stop doing research. We need to innovate because -- or you know you're dead. There's a great paper on nature about proteomics that everybody's kind of fixed upon as well and our technology being used there. And that's huge potential going forward. But for development, we need to continue to invest to make our products more suitable for those kind of emerging markets that become very visible to us. So that will be a big part of it. As you know, it's a good financial analyst, but that also means that a lot of that is kind of capitalized. So that will quite hit the P&L in the same way. In SG&A, it's about sales and marketing enterprise. And as I said, the revenue per head -- the average orders per head isn't even back to where it was in 2022. So we listened and we'll hold that cost base until we kind of start to see the metrics go into the area where now we need to invest again. So very simple on the sales one, very, very top level. We have an opportunity funnel like make number, orders per head, you have revenue per head and the opportunity funnel is kind of ramping because I expect it to -- because we've doubled the number of people out there, they're going to get orders. So they're going to say that they've got all these opportunities coming. When that starts to plateau, that's when I know that actually probably need to start adding in more heads. And we're not seeing that yet.

Do you think it forward when you do see that, will it be more gradual in terms of future sales force expansions? Or do you kind of see the same sort of step function?

So we've got a big presence in the U.S., APAC and EMEA. Really, they're like we're talking let's go direct in South Korea. Let's go direct in -- yes, exactly. Yes, all additive. Not all worth doing, but they're not the kind of same thing as right. We need to go put 100 people in the U.S. to cover every territory.

Yes. Okay. We're almost out of time. Any last questions from the audience? If not, we'll close with our standard question, which is you only have one minute to answer this. What is the most underappreciated or misunderstood about Oxford Nanopore?

Good question. What's most misunderstood? I think at the moment, the misunderstood or underappreciated piece is that people think we're another sequencing device, and we're not. We're a sensing platform. And we keep on saying it, but actually, we're going to start seeing it in the numbers. We're growing underlying when others aren't. We're going into white space where our conventional competitors can't play because the tech won't work there. And I think that's going to start making people realize that actually, there's more to this than DNA. So...

Okay. With that, thank you very much. Thanks for joining us. Nick, thanks for being here.

Thank you. Appreciate it.