Cirrus Logic, Inc. (CRUS) Earnings Call Transcript & Summary

I am Thomas O'Malley, semiconductor and semicap equipment analyst here at Barclays. Lucky enough to have Cirrus Logic here with us. We've got John Forsyth, President and CEO; and Venk Nathamuni, the CFO here at Cirrus. Thank you both for joining me.

Thanks, Tom. Good to see you.

As most of these start off, I want to talk kind of high level strategy. So if we go back to last year, when we sat here, this strategy was 2-pronged. You had a multi-generational, multidata content, [ Square ], your largest customer, the diversification effort in the Android and the notebook market. Can you just update us on how that strategy is playing out? I know, unfortunately, your lead customer didn't adopt the chip in the iPhone 15, but I would also say that the non-app revenue has struggled a bit. Where are we today and where we pointed in the future over the next couple of years?

Right. Yes, I generally talk about our strategy in terms of 3 main vectors. So the first of those is maintaining and building our leadership in smartphone audio. The second of those is expanding our high-performance mixed signal, or HPMS content in smartphones. The third vector is leveraging our leadership in audio and HPMS to attack and penetrate adjacent markets and other new markets for us. So across those 3 vectors, I think we're very excited about how we're doing and what we've got ahead of us, notwithstanding macro events and conditions beyond our control this year. I think there's a lot of exciting stuff going on. So first of all on smartphone audio and kind of maintaining and building our leadership there. We've said publicly that in the coming year, we have new generations of our biggest selling audio products in smartphones, so codec and boosted amplifier. So we're excited about that. In HPMS, expanding HPMS in smartphones, we got a lot of things going on where we continue to be really excited about the opportunities, what we've done so far and the opportunities for the future around camera and power, especially. So we have made a lot of great progress there. We just recently launched our latest generation -- third-generation camera product, for example. And then when it comes to leveraging HPMS and audio content into new markets, which I think is very important for us long term, we've started to see momentum in the laptop space that demonstrates that we can do that. And that's in both audio and HPMS content.

Very helpful. On the technology side because that's a broader strategy side, can you talk about where your core competencies are and how that's changed over time, right? So for analog, you're at leading nodes. You pushed the 22-nanometer. I assume there's some competitive landscape that become more challenging, but where do you continue to out-execute the competition? And how do you continue to be at the leading edge, particularly at your largest customer?

Right, right. Yes, that's a great question. So I think we are at our best, and this has been the case really for a long time in Cirrus' history, even though the kind of goalposts have moved and we've kind of pushed the boundaries over time. We're at our best really when our customers have some really, really challenging mixed signal problems that typically require very high performance, very high precision, combined with extraordinarily low levels of power consumption, extraordinarily low levels of latency and very often for all that to be done within a very, very constrained space. So to give you an example, I mentioned the camera part of our business. In our camera controller products, we're doing -- we're taking in data and moving elements many, many thousands of times a second in order to provide a great stabilization and autofocus experience. If you look at the size premium on like a -- kind of a prosumer camera, for example, stabilization -- the stabilized lenses were as big, stabilized body is bigger and so on. And you have a lot of space to work with. The amount of space in a smartphone camera module is effectively nothing. It rounds to 0. So we've got a tiny, tiny amount of area in which to pack all that processing and control. And so in order to be able to do that, you've got to have really leading edge analog design because you're going to be on an advanced node to get a small chip that packs in analog and a lot of digital processing capability as well.

Yes. I think that diving deeper into the camera side, you have a portfolio where on the low end, you may have one controller in a device. And at the high end, you can have multiple, as high as 4. So can you walk through why are you adding that functionality in a phone? And like if there's a road map that you could talk to as to why that may continue to happen in other modules because you're seeing in certain modules, multiple portions of your product. And so just walk through why it's needed and how you see that road map progressing, such that, that works its way down where all phones may have multiple of your products?

Absolutely, absolutely. And we've seen the attach rate grow over time and the total blended value of our content -- camera controller content grow over time as well. Because at this point, we have multiple generations of the camera controller shipping. And as we brought newer generations, we've packed in more processing, more channels, the ability to do more in the camera. I would say when we look ahead of us, we also have a rich and aggressive road map for further features. I see no sign whatsoever that this area is done. And so it continues to be a marquee differentiating feature that you see talked about a lot in keynotes, and we've been working very, very closely with our customer to make sure that we're the partner for their innovations. And the kind of things that we're doing today are, obviously, enabling stabilization, which allows you to use, for example telephoto lenses, which otherwise are very, very susceptive to camera shake, night mode photography and so on. The abilities that we've brought with very, very advanced processing and high-speed processing to do stabilization very, very rapidly, have been kind of central to enabling those features. So our plan for the future is continue to partner with our customer in innovation and help them deliver even more advanced features. I'm very, very confident that they're going to do that. What we typically see is that those features come in, in the high end and then cascade down, and we've been benefiting from that and just seeing our camera content continue to step up year after year over time.

What do you think the bigger opportunity is? One, your customer adding more cameras or more functionality from an audiovisual perspective? Or 2, do you think that in existing cameras in phones continue to see a penetration story that is more robust?

I think it can be both over time.

Yes. Okay. Very helpful. So kind of related to that question, circling back to the largest customer, I know you can't speak beyond next year. But can you talk about the level of engagement that you have on various projects right now? Clearly, what occurred this year, this is the first time at least in my memory that you guys have ever had a product that didn't go into production, where it was at least generally spoken about that it may ramp. So does that change the way you're working with that customer? Is there a move forward to next year right now, sort of mentality? Or do you try to use that chip in another function? Just anything about where that relationship is and how this cycle changes that?

Right, right. Yes, I don't think it changes anything. I think you'd have to look at it in the context of the length of the relationship that we've had. We worked very, very closely with our customer for 15 years now, 15 years plus. And in that entire time, I mean you're absolutely right. This is the only instance like that. And the hit rate has been phenomenal. And as much as this was a tough break for us and traumatic for us, I think it was for them as well. And so that will very occasionally happen even with great people, when you're trying to push the boundaries of innovation. Our attitude has been that, that doesn't change our appetite to continue to partner with that customer in bringing innovation to market. So I think the relationship is in a great place. Obviously, we work very closely with the customer on the practicalities of dispositioning material we had in flight and so on, and that's all been resolved very smoothly. I think the best indicator though is the number of opportunities that we have in discussion or at some level of being in flight with that customer. And I think that's -- I think there are more of those across a broader range of areas than we've ever had in the past. And that really is a testament both to our consistency and execution and the fact that they see us as a great partner and supplier.

And just to add to what John said. I mean, just to put things in context, as you know, we started out being primarily an audio supplier to this customer several years ago, almost a decade ago. And as you fast forward to today, we have grown the multiplicity of our capabilities, not just in the audio front with codecs and amplifiers but also in the area of high-performance mixed signal. And as we see the pipeline, we continue to work very closely with that customer in terms of expanding the portfolio. And obviously, at the appropriate time, we'll share more details about what's going down the pipe. But suffice it to say that it's been a very strong relationship. And I think, as John alluded to, even with this particular part that is no longer shipping as was originally expected to, we have been working collaboratively both with our customer and with our foundry partner in terms of dispositioning those wafers. So I think it's another indication of the level of customer intimacy that we have and clearly relationship has continued to grow over time. And as demonstrated by more products, more technologies and the range of the opportunities expanding over time.

Right. And to put, I guess, some numbers around, how that relationship has broadened and deepened over time. If you go back to the beginning, obviously, we were shipping one piece of silicon into their devices. Today, in the high-end devices, there are 9 pieces of Cirrus silicon in there and we're still very, very active across the whole range of areas for new programs.

So you've opened the door for the breadth of product question, and I want to dive a little bit more into that. So in the all-important shareholder letter, which is loved across your investors by the way, you mentioned customer interest in our capabilities around battery and power. And when investors hear power they think PMICs. Can you talk about what you're doing today with your PCC chip? And what can you say regarding Cirrus' capabilities in batteries and power?

Right. Yes, absolutely. So first of all, for anybody who's not familiar, PCC refers to power conversion and control. It's not a hugely informative name in its own right, but I thought at least I'd elaborate on PCC. We are -- first what I'll say, we're not especially going after kind of conventional PMICs and the power conversion and control chip certainly isn't that. And there's a variety of reasons for that. It's a market that's in most cases, pretty well served to start with. But also going back to the earlier points we were talking about and the fact that we're typically investing in advanced geometries. They don't make a lot of sense for a conventional PMIC. They make much more sense when you need a lot of logic in there as well, and there's going to be a fair amount of digital alongside the analog circuits. And so when you look at the power conversion and control chip, yes, it's managing power at some level, but there's a lot of logic and processing in there as well. And what that logic is doing is really 3 things. One is it's looking across the system and the power -- it's sitting between that battery and rest of the system. So it's looking at what's happening with the power demand in the system and avoiding -- preventing system brownout when something places a very high level of power demand on the battery which, for example, can happen when you start using AI accelerators and so on. They're a kind of classic case where there will be a big spike in power demand. So typically the lights will dim in the rest of the system in that situation, we prevent that. We also make sure that performance across the system remains consistently really good even as the battery state of charge starts getting much lower. And this is one of the kind of things where you almost don't know the benefit of it until you see it side by side with a device that doesn't have it as the battery state of charge declines. And you can see every -- the performance of just about everything declining. And then the third thing, which is a longer-term benefit of what we're doing is, is managing the battery's health or rather preventing damage to the battery, preventing premature aging of the battery or limiting that in order to extend the life of the battery in the system and the user experience of the device. So all of that requires both a lot of kind of conventional power, circuits in the analog domain, but also a lot of digital logic and control. That means that there are many people who could do that in a single chip, but it's exactly in our wheelhouse. When we look more broadly for other opportunities in the power space, that's the kind of thing that we're looking for, places where there's a lot of -- there's a compelling case for there being a much more digital approach to power.

Yes. So it sounds like you're displacing the classic analog guys. Can you point to certain devices that make sense for a solution like that? I mean obviously, you have PCs but are you're talking about AI at the edge, wearables. Where does it make the most sense?

Well, I think certainly, I think right now, both in smartphones and PCs, we see opportunity -- lots of opportunities for these devices and then this kind of IP and then we anticipate there will be more markets beyond that, hopeful.

So you talked about a refreshed version of your codec and amp at the beginning of this fireside. So there used to be a 2-year cadence, but that clearly extended this cycle. So you've got 4 different chips you sell to Apple, each with their own cadences. I assume, design resources are at a premium these days. Can you talk both at your own R&D center and your customers, how do you balance those different processes? And how do you look to make sure that you're making the right decisions on the right product so that you're spending money wisely?

Well, it's very collaborative to start with. I think one of the many, many things outsiders aren't necessarily familiar with when it comes to working with our largest customer, is the size of the investment they make in a chip program as well, which is considerable. So you -- although we're doing the specification and the design there, they are passionately concerned about how that's going. They want to be able to sleep at night. It's a big investment. You have a lot of resources on their side dedicated to seeing that program through to success. And so all of those discussions about what the right clock rate is for bringing out new versions of existing sockets, for example or other new products and sockets, they inherently have to be very collaborative. I think it used to be the case that the idea of doing a new amplifier and the new codec in 1 year was kind of bad juju or just considered like not good planning because of the sheer amount of work involved. I would say that given the confidence in our execution record and the fact that these are kind of well-known areas now, like 10 years ago, you wouldn't have done that. But I think now we can do that and be confident in our plan.

And I think that also just speaks to the level of customer intimacy as well as the execution excellence in terms of being able to bring on these parts, working collaboratively with the customer. And in one case, we're moving from a 55-nanometer node to 22-nanometer node, but obviously also has the heavy lift of the process technology aspects of it in addition to the architecture. And then on the amplifier side, there are 3 [ instance ] stations of it, clearly a different architecture, much lower power consumption and so forth. So the ability to work in close conjunction with the customer in introducing a multitude of products, both in existing domains as well as in new domains shows the level of customer [ assurance ] and execution excellence that we have.

So you're doing more at once, which is a good thing. But I think that there's been a lot of challenge in this past year with getting more done at the same time, just given wafer supply and a variety of different headwinds in the marketplace. So you signed a large deal with GlobalFoundries to secure supply and you've had to juggle those commitments with your new components that didn't ramp. So can you talk about when you'll be able to reduce inventory levels? And does this leave you with less flexibility from your supply base? Can you just talk through the moving parts there?

I think I'll say a couple of words and let Venk speak more to inventory. But just to give some broad color that the -- first off, the agreement we have -- the long-term agreement that we disclosed with GlobalFoundries is not product specific, it's process specific. We have many products running on that process. So there wasn't really a scenario where we thought we would be -- we would have a particular challenge consuming those wafers. In fact, I think if you go back to what I've said in previous quarters, going back to last year, I think we were still anticipating that our wafer demand would actually exceed those -- that agreement. So that we feel is in a good place. I think the part of the challenge for us, though, has been responding to both a need from customers almost across the board to see more supply assurance, supply chain resiliency and geographical diversity, while at the same time, making sure that we have the right set of partners, enough competition amongst our foundry partners and the right process technologies to serve our future road map. So it's certainly part of our goal is to continue to develop our foundry partnerships and that part of the supply chain, so that we meet all of those objectives.

Yes. And then just talking about inventory, if we just step back a little. As you know in the prior couple of years, we were all in a hand-to-mouth existence as it relates to supply. And so clearly, we were at, at some point, running at less than 50 days of inventory. So even prior to this particular chip no longer coming to the market, we had talked about increasing our inventory position, such that it was much more balanced in terms of the supply and our ability to meet customer demand. And so obviously, we started building some inventory, and then when this situation happened, to John's point, we have a relationship with GlobalFoundries such that the wafer capacity is fungible across the whole slew of products that all supported in the same process node. So as a result, we were able to use that supply, [ that it was virtually they're ] able to get it to that particular chip to other components, which we expect to chip for multiple years to come. And then specifically as it relates to inventory dollars and the days of inventory, we said on the most recent call that the -- that particular quarter represents the current peak in inventory for this fiscal year and we do expect the inventory level to come down, especially as we're building inventory in anticipation of the new product that was launched a couple of months back. And usually, with the seasonal uplift of demand, you'll see an inventory drawdown that's natural for this part of the year, and we expect that to continue for the rest of the fiscal year.

John, you brought up a point that I've heard now in several firesides which is geographical issues of -- with supply. So can you talk about what those customers -- well, what the conversations are like with your customers in terms of what they expect now from you in terms of where you are sourcing product? And can you also talk about where you think that direction is headed because I think it's becoming more and more an important theme amongst the companies that we're talking to.

Yes, absolutely. That doesn't surprise me. I think it really came back on the agenda as the wafer supply situation eased up. If you go back a year ago or 15 months ago, it was maybe at the back of people's minds, but it wasn't a topic of conversation because they just wanted wafers, right? They didn't care where they were from. Just get me the products.

[indiscernible].

Right. But now that we've got past that, it has come to the forefront of conversations in quite a marked way. And I'm not talking about any one customer here, I'm talking about across numerous of our customers wanting both a degree of --so I mentioned supply assurance, but also this diversity in supply, so things running in multiple facilities, for example, and getting away from too much geographical concentration. So that has very clearly come onto the radar as a major theme. From our point of view, it very much fits with the direction we were going in. And I think it's absolutely consistent with our strategy of cultivating foundry partnerships that give us leading edge technology and that geographical diversity. But I've been, if anything, surprised by the degree to which customers started making that a kind of -- a central part of the conversation just about as soon as we got past the crunch.

Very interesting. All right. So we dove down the supply chain and geographic diversity road. Can we pull back to more of the product side in notebooks? Can you talk about your road map in notebooks? You've mentioned that you're shipping to 5 primary ODMs, and you're working with Intel on their Lunar Lake reference design. Can you speak to the size of this opportunity? And what drives customers' decisions to use Cirrus versus their existing suppliers?

Right. What drives customers to choose Cirrus? Well, I think they are choosing Cirrus if they want the best audio. If they want to make the devices thinner, if they want to better track pad experience and if they want more efficient power conversion with less waste heat warming up the planet. I think those are the main things. So if I break those apart, nobody is going to argue that laptop audio, in particular in the Windows world, has been anything to celebrate over the past 20 years. I think we saw several of those top 5 OEMs come out of the COVID kind of work-from-home period with a very different view on that. And so -- and that is sustained, they still hold that view. It's now vitally important that the audiovisual experience of communicating with somebody through their screen is significantly better than it has been in the past. So we see that commitment. I honestly, I think once you go down that road, it's very hard to go back from it. So we're able to provide people with -- provide our customers with the tools, boosted amplifiers and codecs which really do make a kind of night and day difference to the audio experience. I mentioned thinner devices. This is actually closely connected with the transition to boosted amplifiers because if you make your devices thinner, you're going to have less of a cavity for the speaker, you're going to be able to make less air move. It won't be as loud. So you're going to have to put more power through the transducer, which means you need boosted amplifiers. Mechanical trackpads also typically have a bigger stack-up in terms of height than haptic trackpads. So again, thinner devices are going to use -- are much more likely to use haptics drivers instead of mechanical switches for trackpads. And there I mentioned fanless designs and less heat. If you generate less heat, you can deliver a device without a fan, which means it isn't making a stupid noise all the time and it can be thinner and more compact. And our power solutions are significantly more efficient than legacy power converters. That means you basically are doing the same amount of charging or DC/DC power conversion with less waste heat. So it gets you closer -- it helps you stay within the thermal envelope of designing a fanless system. So we're very excited about the kind of range of products that we have and the number of reasons that we believe people are choosing us and looking for solutions like these. So those represent a pretty good tailwinds. We feel it means the SAM is expanding even without any significant assumptions on units. So we've obviously kind of backed off this year, maybe laptops 180-ish million units, something like that. I think most people seem to think that probably 200-ish is a reasonable assumption. In that scenario, the SAM for us continues to expand very significantly over the coming years, so that could easily be $1 billion plus.

And just to get some specific numbers around that. So we've talked in the last few earnings calls about the fact that we've got about 70 design wins plus in this market across the top 5 OEMs and across a multitude of different product categories as well, some of which John talked about. And in particular, in terms of the revenues, we said in fiscal '25, which for us starts in April of 2024, we expect to see low tens of millions of dollars and with a significantly faster ramp in fiscal '26 and beyond. So quite excited about the opportunity. Obviously, we've brought to bear all the capabilities that we have in smartphone space in terms codecs and amps, and all of those characteristics are now relevant in -- especially the high-end of the laptop market, and that's where we're really refocusing in terms of our ability to reuse that IP and expand the SAM in a meaningful way.

The strategy in the logic makes a ton of sense, right? You're moving towards a device that is more and more used in the work home and workaround world. And I don't think -- I think that there is some of that, that moves away from how extreme it was during the pandemic, but I think that clearly things are heading in that direction. So I guess, the question is really, it seems like the strategy is very sound, but talk to me about the time line for that portion of the business becoming more significant. And what are some significant milestones that we should look forward to as investors, as to what you're deeming a success in those landmarks that we can be like, wow, like this is really progressing as the company kind of expected.

Yes. So it will take some time to get to needle-moving revenue. But I would say that the first milestones, the first external signs that you're going to see are when next calendar year, we see customers -- top tier customers bringing product to market that have, in some cases, 7 pieces of our silicon in there, like 4 boosted amplifiers, 2 haptics drivers and a codec. And in addition to that, we're sampling our products today. So I can see instances where we're going to have even more content in that. Our revenue -- we will provide updates to our investors on the quarters, on how we're doing and how we're tracking. And as that revenue increases, we obviously look at providing more visibility into that. But just to recap on Venk's point, we think as we go into fiscal '25, I've said low tens of millions and then a more significant step-up into fiscal '26. And sometime in fiscal '26 or '27, I think if Android stayed where it was today, we expect the PC market to outstrip that as a part of our business.

Very helpful. I think we've run out of time here. Thank you so much for joining us and have a great rest of the week.

Yes. You, too. We appreciate it.

You, too.