GLOBALFOUNDRIES Inc. ($GFS)

All right. Good afternoon, and welcome to the second day of JPMorgan's 54th Annual Technology Media Communications Conference. My name is Harlan Sur. I'm the semiconductor capital equipment analyst for the firm. Very pleased to have Sam Franklin, Chief Financial Officer at GLOBALFOUNDRIES. For those of you that don't know, GLOBALFOUNDRIES is the fourth largest semiconductor foundry in the world. Leader in specialty and mature manufacturing technologies, targeted segments such as analog, power management, RF, wireless, wired networking, connectivity, compound semiconductors, advanced packaging, silicon photonics targeting end markets like data center, comms infrastructure, mobile, IoT, automotive and industrial markets. And also a growing portfolio of compute related intellectual property and capabilities to support their customers' efforts in the fast-growing areas of Edge AI and physical AI. So the team reported solid results, constructive guidance last week also held in Analyst Day, which we'll be talking about. So Sam, thank you for joining us this afternoon.

Thanks, Harlan. Please to be here.

So let's start off with the Analyst Day and when you and the team built the financial long-term model that you laid out last week, what was the single most important shift in the GF narrative you wanted to crystallize and send a message to the market. And second, from an outsider perspective, and the obvious question is whether the current business strength is structural or partly cyclical. And where do you think the team is generally structurally better today than, let's say, 12 months ago?

Sure. So Maybe just to take a little bit of a step back. It was a little over 3 years since our last Investor Day. And obviously, when you think about some of the cyclical dynamics that have been going on within pockets of our industry during that time frame, we felt that there was a prime opportunity for us to really kind of convey the way that we've been repositioning our business internally, but also more importantly, how we see that intersecting with the mega trends facing our industry and expect it to kind of drive momentum in our industry over the course of the next 5 to 10 years. So the real confluence of messaging, I'd say that resonated most with our investors during the course of that day was sort of threefold. It was these megatrends driving our industry, the intersection with GF portfolio and what it means for our financial model going forward, driving increased revenue diversification, increased margin expansion and really kind of delivering shareholder value as well. So the -- I don't think there's any disputing from anyone in the room right now that the data center is a key driver of -- those mega trends. What is probably underappreciated is the role that GF has within the data center today and the expected role we have in the data center going forward today, principally service through our silicon photonics portfolio as well, by the way, through our silicon germanium applications too. So we've seen strong momentum there. Data center going forward, we expect increased opportunities to fall through from silicon photonics, the growth of co-packaged optics but also power delivery within the data center. Again, a sweet spot opportunity as we think about the investments we've made in our portfolio as well. So that's kind of one part of it. The second part of it is, well, what does this mean in terms of all the data center CapEx build out as it relates to the applications, the devices that are connected to that. And that's where we think there's a real opportunity for us as it relates to physical AI. You think about what the core features of physical AI are, what do they need to do? They need to sense, they need to think they need to act, they need to communicate. All of those features play into the sweet spot of GF's portfolio. And we think there's a meaningful opportunity there over time. And then the third of those trends is really the surety of supply and the diversification of a globally secure semiconductor manufacturing footprint. And again, GF is very unique in so far as we have presence on 3 continents. We have increasing technology fungibility across each of those. And so when you think about some of the challenges facing global supply chains today, we think we're very well positioned to support that going forward. So really, I'd say it was the confluence of those core drivers with our opportunities as a business. Now to the second part of your question, what's different as it relates to structural versus cyclical, I'd probably bring this a little nearer to home and tie it to our results that we reported a couple of weeks ago. The best reference point I can give you all is first quarter revenue was up about 3% year-over-year. Our gross margins expanded 510 basis points year-over-year. A big component of that is the mix evolution that we're seeing within our business and also the end market diversification as well. That also carried through, by the way, in principle to our second quarter guidance. where at the midpoint, we guided to about 4% year-over-year revenue growth. But actually, if you look at the second quarter revenue from last year, we did about $1.68 billion. You take that midpoint of EUR 1.76 billion that we've guided for the second quarter. And then you take the 28.5% of margin delta that we provided as a margin guide. It actually implies that we're having a very strong pull-through from a gross profit perspective as it relates to that revenue year-over-year bridge. So look, I think that they, for me are some of the strongest indicators that we've made structural changes within our business to drive margin accretion.

On the other side of it, we'll talk about the growth rates in some of the model parameters that you laid out. But as you think about the long-term outlook for the business, the financial model that you laid out, where is the execution maybe still falling a little bit short? Where are the internal capabilities, technology portfolio, product portfolio, still falling a little bit short of expectations relative to the team's ability to confidently like hit those targets as we look forward in time.

As you can imagine, we spent a lot of time thinking about this and also presenting and putting together the most credible model for you all and for the investment community as well. And so we sort of broke it down into 2 distinct phases as we set those objectives for ourselves. The first is a set of metrics exiting 2028 where we're looking to solve for a 40% gross margin business and then longer term, looking to solve for a 45% gross margin target. Now the dynamics, I would say as it relates to where some of those challenges were some of those capability opportunities like rust is really the second phase of that model. And I'd say it falls into a couple of very distinct categories. The first, as I touched on earlier, as it relates to silicon photonics, is this growth of co-packaged doctors France packaging. That is an area we've been investing for the past decade. But nevertheless, when you look at what broader industry expectations are for the ramp of CPO. It's sort of in that late '28, early '29 time horizon. And so as we think about the step change in revenue pull-through and earnings leverage going beyond 2028 CPO will be an important dynamic. The second dynamic is very much 1 that we've actively pursued to, again, diversify our service offering to customers and it relates to the acquisition we made for MIPS at the back end of last year, which I know we'll talk about later. But that drives a few things, but 1 of them as it relates to that second time horizon window is the advent of custom silicon -- we're working with customers at this point as it relates to new opportunities and some of those will have custom silicon attached to them as it relates to new designs on GF technology. Now typical life cycle from where we're at now from a design win through production ramp typically in that sort of 2- to 3-year time horizon. So as it relates to those phases of opportunities, I'd really say Phase 2 is an area of capability where we've been investing in heavily, the ramp associated with that is more in that end of 28 and beyond time frame.

Got it. And then on the financial model and the midterm targets, 10% to 12% revenue CAGR over the next few years, $4 of earnings power exiting $28, $6 long term sort of earnings power. When you and the management team and the Board were sort of stress testing this model, where did you put the most conservatism? Where is there sort of the least amount of uncertainty? And is it the gross margin trajectory, operating margin leverage, revenue CAGR or CapEx envelope, like where were the areas where there was a little bit more uncertainty, a little bit more conservatism.

I mean, look, as you can imagine, I'm sure you give to these as well there's no sort of singular area of the financial model where a degree of stress testing or conservatism is applied. We stress test all elements of our model or through the top line assumptions through the P&L. And so we spend a lot of time going through those various different scenarios from a bottoms-up perspective. How do we control cost? How do we drive productivity what's the design win momentum that we're seeing with customers. Fundamentally, we wanted to build a model that was anchored in the momentum we see today and the technologies that we're investing in today. Then, of course, you have your top-down set of assumptions as it relates to the end markets that we service, the relative SAM growth within those end markets as well. But really for us, it's how do we drive that growth control that from within. So maybe if I kind of take you through a couple of elements from a P&L point of view. Look, the way we positioned the end market growth opportunity that drives up to that 10% to 12% that you mentioned at the enterprise level is really through a few core vectors of growth that we're seeing meaningful acceleration of today. Comms infrastructure and data center. We grew that almost 30% last year. We grew 32% year-over-year in the first quarter. Our expectation for that end market is high 30s percent growth in 2026. We built a model where we expect that to go through the cycle in the 30-plus percentage points, which is pretty similar to where you're seeing the broader market. But again, we've got some pointed technologies within that to capture value. If I then take automotive, another core end market for us, we've grown our automotive end market revenue 14x in the last 5 years. It has been a really strong growth story, particularly when you look at some of the other cyclical dynamics over the last few years, Auto has been a bright spot for us. And that's really the function of the design wins that we've been building, but it's also the durability of that as an end market. So we put a low double-digit percentage growth around that. And we think it kind of drives from 2 key areas that, again, is based in momentum we have today. Number 1 is the increase in semiconductor content within the vehicle, roughly $1,000 of semi content in the vehicle today, the expectation is that, that has a 1.5x increase through the end of the decade. It's very easy to get attached to SAR dynamics within auto, but content really matters. And that's where we've been picking up opportunities. The second part is the gaining share dynamics within that as well. The third piece is really a function of what I mentioned around MIPS and the mega trend around AI, which is the opportunities we see in IoT we've baked into our model and expected growth rate or mid-teens for IoT through the cycle. And again, the reasons behind that are not just the additional capabilities we're adding through the likes of MIPS and Risk 5, but also the momentum we see with customers. We had about 200 design wins in IoT last year. That was about a 40% uplift than the year prior. So we're seeing good traction with customers there. And then it would be remiss of me not to mention smart mobile because historically, that's been our largest end market. But what I think resonated with our investors is that we've built a model, which is not dependent on smart mobile growth. And so we actually view that as a flat revenue end market our model. So that's sort of how we think about it from top line and then just briefly touching on the other components, mix really matters in our business. And when we look at the progression from where we are in the first quarter, which was up to that near-term target of 40%. Mix will be an important component of that, call it, sort of 5 to 6 points as we see increased pull-through from those end markets that I just mentioned. Then as you get into some of those other margin bridges, the likes of optimization of our footprint, tooling productivity, really kind of filling out the falls of our capacity, we're a high fixed cost business that matters. And then just taking it a little step further. And if I go into our operating margin targets, we've set a long-term target of 35% nearer term, 25%. I do actually expect our OpEx to grow over the course of that period, but call it roughly half the rate of revenue. The composition of that OpEx really matters as well. And as we add on more capabilities in risk 5 and MIPS as another example, our expectation is that the proportion of R&D is probably slightly larger than what it is today. But overall, that envelope is roughly half the rate of growth. So there's a lot to unpack there, but the punchline is we stress tested a lot in terms of how we think about this growth.

That makes a lot of sense. And if we talk about the -- you talked about the model looking beyond 2028 and the team at the Analyst Day, you definitely brought up co-packaged optics. And you're right, I mean, NPL, CPO, those are dynamics that when you think about scale out, maybe we see some in '27, but more in '28 scale up maybe '28, '29 time frame. So I totally agree with that. And we'll get back to your silicon photonics portfolio. But you also mentioned 2 other drivers, advanced packaging and custom silicon, and this created a lot of buzz after the Analyst Day. And so the question is like what is the team doing here? Where are you going to be focused on advanced packaging and custom silicon, right? So advanced packaging, there's been a lot of questions, are you going to be pursuing things like [ 2.5D cost-like ] capabilities. And on custom silicon, the GF team used to have an ASIC team, right, doing custom silicon solutions but sold that portion of their business to another semiconductor company, but are you planning to rebuild the custom basic team as well to go after the custom silicon market? Is that what you mean by custom silicon?

Why don't I -- I'll start with the advanced packaging and finitely get into the custom silicon. The thing that we're excited about from an advanced packaging point of view is that it's areas of capability that we've already been investing in for years as it relates to wafer to wafer, diet to wafer bonding. I know we'll come on to [ Sipo ] later. When you think about some of that wafer to wafer bonding capability, we have qualified solutions in silicon photonics for PIC to EIC bonding. That's going to be an important dynamic. But the capability doesn't stop there, is another great example of where we're in development with bonded solutions for our NSW technologies, high-performance RF at 45% less of the die size when you think about the real estate within the phone and the RF capability that you need, that kind of size matters. And so we're seeing good opportunities there. And then we're in early-stage development for things like Siggi on FDX as well. This is a broad area of advanced packaging and die to wafer wafer-to-wafer bond that we see as opportunities, both within and beyond silicon photonics. And then if I switch over to your question around custom silicon. Maybe just taking a bit of a step back. We acquired MIPS for multiple reasons. One, we have a strong thesis in the belief and the applicability of risk 5 but also the relevance of that business in terms of the revenue composition. It is a business today, which is principally IP licensing software-based revenue. Over time, that drives towards more custom silicon. We put a target out as part of our Investor Day a couple of weeks ago of looking to get to $1 billion asset in 2030 for that business. And I will say that the reaction from customers since announcing that acquisition has been overwhelmingly positive. It changes the nature of the service offering that we can provide to our customers and it brings us into the discussions much earlier than we would have otherwise had as just a pure play foundry recipient of an RFP for a D win for example. So we're finding we're getting that early traction, and we're having the types of design conversations, which are very relevant to the GF portfolio, which then takes me on to the last part of your question around legacy ASIC business we did sell to Marvell. I view that very different for a couple of reasons. That ASIC business that we sold to Marvell at the time was principally focused on designs for 5-nanometer embolus.

Leading edge

Right, that's right. It was outside the wheelhouse of the technology portfolio that we were focused on as a company. So strategically, it wasn't the right fit for us going forward. What we found with MIPS and what really gives us some optimism around the opportunity there is that touching back on those features I mentioned earlier as it relates to sensing at Unica, the relevance to GF technology portfolio is very well suited there. So as you think about the opportunity for custom silicon, those design capabilities are not just to support a customer on another node, it's to support technologies that GF services today and going forward. So that's really the nuance between the 2 dynamics between that legacy ASIC and where we see ourselves today with --.

Oh, that's good insights. And then maybe switching over to sort of the cyclical dynamics of the business and the resilience of the business to this cyclical downturn and now we're in the midst of the upturn. After 8 consecutive quarters of year-over-year declines, the team starts to see revenue recovery trends at the beginning of last year with year-over-year revenue trends inflecting positively. Q1 remaining positive throughout the year. As you mentioned, you entered this year driving 3% year-over-year growth, guiding up 4% or 5% in 2Q. Street has you guys exiting the year at about 7% to 8% growth. Could you just share your thoughts on the current state of the industry and the impacts on GLOBALFOUNDRIES' recent earnings update.

Yes, absolutely. And look, you hit it right on the head there, which is I think we're in an important inflection point, not just an important inflection point for our business, but also for the industry more broadly, it was encouraging to see a return to year-over-year growth about 3% in the first quarter. And as you say, we followed that up with roughly 4% of growth at the midpoint for the second quarter as well. Now all of that is actually in the face of a couple of ongoing dynamics, which I'll come on to from an end market point of view. But we're really, really encouraged by the diversification that we're seeing from an end market perspective. I will say that if you looked at our business 5 years ago, you would have seen revenue contribution from Smart mobile devices well over 50% of total revenue. In the first quarter of 2026, revenue for smart mobile devices was 34% is the lowest that it's ever been as a percentage of our total revenue. And that's with enterprise revenue growth. So notwithstanding that reduction, actually, the growth has come, and it ties to what I said earlier around the mix and the margin dynamics, growth has come from some of those fast opportunities that we're seeing in other end markets. Communications infrastructure and data center up 32% year-over-year. I see us as upwards of high 30s for the full year. Automotive, we delivered about 24% of growth year-over-year in the first quarter. Again, automotive proved to be a very resilient business for us. We're continuing to grow customer partnerships in that end market as well. And although pockets of IoT were a little bit soft for us in the first quarter, we are seeing good momentum pull through from a take-out perspective, which is a good leading indicator for where we see future revenue. So we still expect IoT to be about mid-single-digit growth for the year as we go through. So that end market diversification has really helped drive some of that inflection. The other point as well is we're seeing good contribution from our technology services revenue as well. This is revenue we use to categorize as non-wafer revenue. We broaden that definition to technology services revenue to really reflect what we're doing with our business and adding that service capability. So that was about 13% of total revenue in the first quarter. Historically, that revenue has been sort of 8% to 10%. It's trended upwards for a couple of reasons. One of those is that we saw good momentum from a mask and radical perspective in the first quarter, particularly on some of our aerospace and defense applications. But also, we're very gradually seeing that broader service offering pull-through from the likes of the MIPS acquisition. Small in the first quarter, but expected to grow as we go through the year. So I would say that, yes, on paper, it's an important inflection, but it's taken years of effort to kind of drive that diversification and get us back on to the track that we see for future rate.

And I think to your point, I mean, a lot of the things that you're talking about today and going forward have been sort of newer dynamics to the team's profile, right, MIPS and the focus on compute IP, the inflection that you're seeing in some of your data center photonics related sort of businesses. But if we go back a couple of years, I mean, your peak to trough decline in the last down cycle was only 25%. I mean it was significantly better than in your customers or your competitors, right? And so even 18, 24 months ago, I mean, the team was doing something, right? I mean your peak-to-trough decline was very minimal. How was the team better able to weather the down cycle. And it clearly seems that on a go-forward basis, the potential for you to do even better, given the mix effects and some of the other diversification of the portfolio is going to come into play. But how was the team able to very well managed through the last down cycle?

Yes. Look, our team has done a phenomenal job for years, and we've got 1,000 very strong workforce around the world that is constantly looking to drive progress for the company. So yes, they've done an excellent job. I'd kind of really bucket into a few different categories as we think about how that relative progress we saw for the last few years. I touched already on end market diversity in, so I won't labor that too hard. But what I will say is that, that was a set of decisions and wins and opportunities that crystallized well before you started seeing the cycle turn the other way in some of the consumer-centric end markets.

And auto, I mean auto is really strong during the...

Precisely. Yes. I mean, look, if you stripped out auto, from our business over the last 3 years, you'd have a very different peak to trough ratio. And so it really kind of hits home on the importance of that diversification piece. We also had a pretty pronounced migration away within some of our data center revenue as well. So that data center momentum that we've seen over the course of the last couple of quarters is really off the back of some declines in the prior year. So auto end market diversification more broadly, strong dynamic. The second component of it was really this focus that we have from a business and a manufacturer -- how do we drive improvements from a productivity point of view, constant focus on cash cost per mask place, focusing on improving cycle times as well. There are certain things when you get into a cycle which are outside of your control, there are certain things which are inside your control. And how we perform better as a semiconductor foundry was clearly in our control. And so we look to really kind of drive some improvements there. I will say that there was probably a third component here, which we've talked about a lot in the past, less so now, which was some of those longer-term customer agreements, which provided a degree of visibility. I say that with a degree of caution around that statement because the reality is that we wanted to preserve the long-term nature of those customer relationships. And so we were very thoughtful over the last couple of years to make sure that where possible, we negotiated. We supported our customers and helped each other through that period. But I'd say they were some of the core structural dynamics that we really focused on to try and help weather the storm and look, it's positioned us well, we think, in terms of the growth opportunity ahead.

Before I move on, any questions from the audience. If you do have a question, raise your hand and we'll get a microphone over to you. Let's talk about the design win and customer pipeline. So you exited calendar 25 with an increment to 500 new design wins. 95% sole source to GLOBALFOUNDRIES. Additionally, you expanded your customer pipeline to include Renesas. I know the CEO of Renesas [indiscernible] extremely well. And you now with Renesas, you now supply manufacturing services to the top 3 microcontroller suppliers in the world, Infineon, NXP and now Renesas, right? So first, help us understand the end market breakout of your design win pipeline. And what was the major motivation for Shibata song and Renaissance to partner with GLOBALFOUNDRIES?

Sure. Look as you rightly touched on, 500 design wins in the year was a strong high water mark. So far for us as a company. And so we feel good about what that means in terms of the future opportunities for the business. And I gave you a little bit of an indication earlier around some of that composition. IoT was a big component of it design wins across IoT. The way to think about some of the balance of those design wins, it was actually quite evenly spirit across automotive, comms infra data center and smart mobile for that matter as well. So relatively good diversification. It actually adds further credibility to what we believe is possible in terms of end market diversification over the years ahead. And look, the first quarter started quite well with the continuation of the design win momentum. So really kind of a basis for growth, which is steeped in that customer momentum. Renesas partnership, really important to us and a multiyear, multibillion-dollar partnership. Shibata-san, I think, made some really pertinent comments around why he viewed GF as an important part to Renesas. And it's really kind of a combination of factors. One of those is technology capability. We have got BCD, GAM, our MCU business as well, more broadly to be able to support everything from data center power to automotive to IoT. So there's a broader opportunity here as we see it across several end markets that we service and really a kind of good opportunity for Renesas as well. It plays into the sweet spot of their capabilities, too. The second part of it, which Shibata-san included in some of his comments was the resilience of global semiconductor supply chain and the relevance of GF to meeting that objective. We have the ability through the application of our BCD, GAM, other automotive applications to be able to service that demand both here in the U.S. as well as just in Germany as well as Singapore as well. So we're looking to create that fungibility and that broader supply chain security for our customers, and I think Renesas is a great example of that.

And the next thing about doing business with the top 3 microcontroller suppliers in the world is they always love diversity of compute architecture, right? It's ARM, proprietary and a lot of them now, as you probably know, are starting to do some work on risk 5. So having the top 3 microcontroller guys is your top customers and having a great risk 5 business, I mean I feel like it's very synergistic and very conducive to -- as these top 3 microcontroller players start to diversify their portfolio to include Risk5, I mean you guys have the IP portfolio ready to go.

Yes, that's absolutely right. And I think that's where we've been most encouraged by some of the momentum that we've seen since closing on the MIPS acquisition and even since announcing the acquisition of Synopsys [indiscernible] business as well.

Let's go to Photonics, which is clearly the centerpiece of GLOBALFOUNDRIES growth narrative and where a lot of investor energy is concentrated right now. The team is a leader in silicon photonics for 400 gig, 800 gig, 1.60 pluggable transceivers, right? That is the sweet spot for optical deployments today. And in a strong position and in the strong position for a lot of the components that not only -- you don't only build a silicon photonic substrate. A lot of the components you guys also make those components, too, like the silicon germanium drivers, the transimpeded amplifiers with their silicon germanium?

That's right.

And I think you guys even do the germanium photon detectors as well. So there's a ton of content that goes into these pluggable transceivers. The growth trajectory you laid out is pretty solid, right? 2025, $200 million from silicon photonics, 26 doubles this year, doubles again to roughly $400 million, now you're not committing to a $1 billion run rate exiting and ultimately $2 billion in 2030. You're clearly outgrowing data center CapEx spending trends. Help us unpack the components of the strong growth profile and maybe you can lay out for investors, the positioning of GLOBALFOUNDRIES product offering relative to some of the other guys out there like a TSMC Coop platform on the co-packaged optics side.

Yes. Again, maybe if I sort of start with the pluggable side and then we talk a little bit about CPO. Look, we've been more broadly investing in Photonics for the best part of a decade. We've deployed the kind of $1 billion to photonics capability. This is not a new technology capability to GLOBALFOUNDRIES. And so really, what was kind of holding back this opportunity was I think if you rewound the clock a couple of years, the market expectation for Photonics was still a case of if rather than when. And that's really been a pronounced evolution. As you think about this migration and broader adoption of optics even as little as the last year where you've seen that kind of ramp up in terms of broader ecosystem partners talking to it. So pluggables near term has been a strong growth opportunity in the public transceivers. We're one of the only photonic foundries with both 200-millimeter and 300-millimeter capability in this regard. And so strong momentum there, as you say, doubled last year, we expect to double this year. It comes with a healthy margin profile as we think about our objectives from an enterprise perspective. And really, it's another area where we're focused on investing in corridor capacity to continue servicing that demand. We see strong demand in pluggables through this year, through next year as well. So we're going to be making sure that we're putting that capacity on investing those CapEx dollars thoughtfully to meet that demand. You skip then onto sort of CPO. And again, we think of that as sort of the late '28 time frame. We announced recently our scale solution. We've actually had this solution for a long time. It was a function of the recent OCI MSA framework that was put in place our scale solution is our silicon photonics co-packaged advanced light engine, which really drives what we believe to be a very comprehensive solution as it relates to co-packaged optics. This market, if you look at it over the next decade and you look at the broader adoption of CPO, there's plenty of room for more than 1 participant in there. And so, without speaking specifically on TSMC's Coop, which is a very strong solution in its own right. We think that particularly through the adoption of the OCI MSA principles, you look at some of those founding partners behind that framework and obviously, we're engaged with several of those names within it. We think we've got a very compelling solution to support CPO longer term.

On capital return, you announced your first ever quarterly dividend at the Analyst Day and alongside it a target to return up to 50% of your adjusted free cash flow via a combination of dividend buybacks. Kind of walk us through the sizing logic for the $0.12 dividend. Was it anchored to a target dividend yield to a payout ratio to a fraction of the free cash flow target or some combination of at above. And over the long-term model, do you expect the dividend to grow at a fixed cadence with buybacks flexing opportunistically around it? Or is there other way to kind of think about dividend versus sort of repurchase?

Yes. Look, this is a really important milestone for GF, particularly when you think about our maturity profile as a public company. We've had strong free cash flow generation for the last few years. As I said earlier, I think we put together a model which we have conviction in around the opportunity growth from a profitability point of view and the opportunity growth from a free cash point of view as well. So really, this was about applying a thoughtful and systematic approach to how we think about capital allocation. And that really kind of led us up to 3 principles. The first is that we're going to reinvest in the business in accretive margin corridors. The second is where we see opportunities to accelerate our growth model. If we find inorganic opportunities, we'll look to take advantage of those as well. And the third is then looking to distribute a portion of capital back to shareholders as well. And as you rightly touched on, that's sort of where we announced at the Investor Day, the latest update. We think that a 50% distribution of free cash flow after investments is a reasonable basis to set ourselves on Actually, if you look back at the first quarter already, not that we had the formal policy in place, we bought back around $400 million of shares during that period. The dividend itself, if you sort of pro rate out a $0.12 per share dividend that we've announced, called out roughly $0.48. That's about GBP 270 million there or thereabouts of distribution from a dividend perspective. You apply the target we set this year of roughly 10% free cash flow. It's sort of right in that sweet spot with a bit of headroom above it. As you rightly said, we want to be able to fund and grow this over time. we have what we believe is a very resilient balance sheet. We've got $4 billion of cash, $1 billion of incremental liquidity. We have less than 1x gross debt. We feel like we're in a resilient place to be able to support capital allocation on a multiyear basis.

Appreciate the time spent, Sam. Looking forward to monitoring the execution of the team as the year unfolds.

Thank you, Harlan. Really appreciate it.