Keysight Technologies, Inc. (KEYS) Earnings Call Transcript & Summary

Thank you, everybody, for joining to talk with Keysight. Happy to have Neil Dougherty here, CFO, to go over the business. Tim Long here, IT hardware, comm equipment analyst. So Neil, thank you.

Thanks for having us.

Yes, love it. Maybe let's just start out with just like a little high level on the business. Had a pretty good quarter last quarter, gave some fairly positive guidance for the year. So maybe just at a high level before we get into the end markets and other pieces, talk a little bit about where the company is, what kind of momentous and areas are you excited about heading into 2025?

Yes. So as you noted, things incrementally better here in the fourth quarter. We've been through 5- or 6-quarter now correction in our end markets, but it does feel like at this point, at the highest level, things are starting to loosen up a little bit. Our funnel is growing. Opportunities are moving through the funnel at a faster rate. And so just generally a little bit less hand to mouth maybe than we had been, say, 6 months ago. I think as we look forward, Keysight serves a wide range of end markets, and not all of those end markets are inflecting positively. Some of them are. But right now, at least, in terms of our base case, and I know he said we get to the end markets in a minute. Our assumption is not that everything is going to kind of inflect positively and phase with one another. It's going to be more of a mixed demand environment where some of the things will be starting to inflect and where we might still see some continued pressure in other areas of the marketplace.

Okay. Great. Great. Yes. Let's start getting into it. I wouldn't be a tech analyst if I didn't start with AI. So I know you guys are getting a lot of questions, getting a lot more involved in that as an end market. So talk a little bit about kind of the solutions that you're seeing being adopted specific to AI. And I know you're not really sizing the business, but just talk about how it helps that wireline related business.

Yes. So within our Communications Solutions Group, we have a commercial communications business that's roughly $2.2 billion, $2.3 billion. A little less than half, call it, 45% of that is focused on the wireline ecosystem. And again, with AI is the driver that is really driving pretty broad strength across those end markets. And so the nice thing about Keysight and the way we approach these end markets, we tend to have a lot of touch points. And that's true in wireless, and it's also true in wireline. So if you think about the wireline space, we sell to the folks that are working on network silicon, and we're helping them primarily in the R&D lab to get these network silicon solutions into the marketplace. We sell to the folks that are making network transceivers, again, primarily in the R&D space. We sell to the folks that are making connectivity, connectors, cables and connectors. We sell to the folks that are making network equipment. And then we sell to the hyperscalers themselves and others that are deploying data center environments. And so a lot of touch points. We sell -- again, the business, roughly $1 billion. It's probably 65% R&D, maybe something along the lines of 20% manufacturing, and then there's 10% or 15% that's kind of post-deployment monitoring of network environments types of applications, post-deployment types of applications. And we sell tools that are focused on Keysight's traditional areas of strength, which is the physical layer test, but also via our acquisition of Ixia, which is an acquisition that we completed back in 2017. We have protocol layer solutions for wireline as well. And so again, a lot of different touch points, some AI specific, you can think about the NVIDIA ecosystem or some of the things that hyperscalers are doing and some of more infrastructure related, you can think of helping the industry to increase Ethernet speeds from 400 gigabit to 800 gigabit to 1.6 terabit, ultimately to 3.2 terabit and beyond, where we are providing essentially equipment to help in the R&D labs as those Ethernet speeds evolve, which are essential to the AI environment. So while not all of the solutions are AI-specific, AI is the underlying driver right now of the vast majority of strength in the wireline ecosystem.

Okay. Yes, we got a lot of questions trying to correlate what happened in 5G wireless several years ago with AI. Do you think there's that type of durable growth behind this newer technology given you -- as you said, you have a lot of touch points?

Yes. I do. And I think there are other parallels as well. So I just talked about all the touch points in the wireline ecosystem. We had a similarly broad set of touch points in the wireline ecosystem -- excuse me, in the wireless ecosystem. And then if you think about 5G, and I'll draw the parallel here to AI and wireline at the moment, when we started selling 5G solutions, it was to a relatively concentrated set of customers. It was the big handset manufacturers, the big component manufacturers, the small set of companies that make network equipment. But as the -- as you got closer and closer to deployment or even post deployment, the set of customers that were working on some portion of a 5G link solution really started to expand. So we went from dozens or many dozens to many hundreds of customers in the 5G ecosystem. And I think we're seeing something similar right now on the wireline ecosystem. Like it's a highly consolidated set of customers, and it's a constrained environment, right? There's just not enough supply of GPUs and some of these other things for the smaller players to broadly adopt these technologies, but that's improving. And so what is right now focused on a small set of hyperscalers and the equipment manufacturers are supplying them, I think, is going to broaden out. It's going to democratize and you're going to see other customers look to capitalize on the data that they have in their environments put in AI learning tools so that they can ultimately use AI to improve their product offerings and serve their end markets better.

Okay. One of the areas also with AI that you guys have talked about is a little bit more on the software and emulation. Can you talk a little bit about how meaningful that can be? And obviously, that's probably higher quality revenues for Keysight as well?

Yes. I think there's 2 things there. So first, I'll talk about it as it specifically relates to AI, but then I think there's a broader software story within Keysight that cuts across end markets. So specifically, as it relates to AI, one of the new things -- a new product that we brought to market, so it's released in the last few months. So still relatively small from a dollars perspective, but I think is indicative of what we can do. We're now selling tools to customers that are deploying data centers. You can think of customers like the hyperscalers that enable them to, in a lab environment, model these AI workflows across a simulated network environment. And what this enables them to do is they're planning their next data center deployment, it allows them to optimize those data flows, again, in a lab environment so that ultimately, they can deploy a smaller but more efficient, i.e., less expensive network down the road. And so again, bringing in Keysight's emulation capabilities, largely software solutions to help solve an industry-specific problem for AI. I think the broader trend within Keysight is how do we use software to help the industries we serve solve complex problems. So we did this in 5G, providing a 5G protocol stack that enabled our wireless customers to emulate how their devices or wireless devices would perform on a deployed 5G network in a lab environment or allow network equipment manufacturers to see how their network equipment would perform when flooded with voice traffic or data traffic or video traffic. Again, in a lab environment to allow them to optimize. And so we continue to look for new ways to use software to help our customers advance their time to market and their solutions.

Right. Okay. Great. Great. Maybe just the last one on -- well, on just the more traditional optical networking pieces of the wireline business. Other than AI use cases, are we still seeing growth and rebound in that category as well because there tends to be pretty steady upgrade rates, speed and kind of hamster wheel on optical. So is kind of the core wireline business? How is that looking?

Yes. I mean, again, I think I'd go back to AI as the driver across the ecosystem. So I actually think you're seeing an acceleration or at least an increased sense of urgency by the markets of how they want to move from 400 to 800 to 1.6 terabits. So I think if anything, there's only increased pressure to increase the speed of evolution. And then in order to enable that, obviously, you need faster network. So, you need connectors that are keeping pace. You need network devices, switches and routers and things that all keep pace. And so -- so yes, the underlying fabric of these wireline ecosystems are all being driven. But again, the catalyst is this AI trend.

Okay. Great. Maybe let's touch on wireless a little bit. We mentioned 5G there. Last quarter, it sounded like wireless was a little better. So maybe talk to us about kind of the near-term outlook and what has been improving on the wireless side?

Yes. So again, we have a wireless business that's significantly north of $1 billion, somewhere in the neighborhood of $1.2 billion. And so while the market has been under pressure now for a period of time, there's still an awful lot of activity that is happening in the wireless ecosystem. And I think as we look forward to FY '25 and the environment that we've seen over the past couple of quarters. I think what we've seen is this business stabilize, albeit at a lower level than we were at, say, at the peak in '22. But we've at least seen a nice stabilization and steady investment across a number of themes, right? So we continue to look to the ongoing releases of the 3GPP standards. The one that's currently being worked, not yet released is Rev19. And Rev19 is going to have a lot of the product definition for what they're going to call 5G advanced. So if you think back to -- went from 3G to 3.5G, eventually 3.9G, then LTE, LTE Advanced, right? The 5G that we're all using in the networks today is kind of the first instance of 5G, but there will be a 5G advance. It's going to deal with things like satellite communications, lower -- improved power usage, increased security features. Also included in that, will be starting to think about AI at the edge kind of almost as a transitionary technology towards 6G. Things like uplink are going to become much more important part of the wired network as you think about this. You've got this AI-driven transition that's happening in wireline. But ultimately, the primary access point for those technologies are going to be wireless devices. It's going to be tablets. It's going to be cell phones. So the wireless networks need to keep pace.

Okay. Maybe touch on some of the like sub verticals here, O-RAN, enterprise 5G. Are those stabilizing as well? Or where do you see growth coming from?

Yes. So I mean, I think if you look at where investment is happening, O-RAN is certainly one. We talked about democratization earlier, right? The radio access portion of the wireless network used to be a 2 or a 3 customer space. But now with O-RAN, you have many, many customers that are looking to do a portion of the network device. And that brings in from a Keysight perspective, all sorts of new interoperability testing that we didn't have to do when it was kind of all designed under one roof. Satellite communications, I think, is another area where we're seeing significant increased levels of investment as it relates to 5G. You're going to be able to get better coverage with a direct satellite to user link than you can. It's going to be easier to provide coverage to rural areas or hard-to-reach areas via satellite link than necessarily going through the cost of deploying base stations into all of those areas. So that's another example of an area where we are seeing investment today within the 5G ecosystems.

Okay. Yes, there's probably been some pessimism around 6G timing given that 5G from a business model standpoint for operators doesn't seem to have moved the needle that much. So what is your view on -- we have these big cycles that you mentioned, 3G, 4G, 5G. What's the latest thinking on 6G? I know there's some initial activity, but how do you feel about the kind of upcoming cycle?

Yes. I mean, again, I look at what's happening in wireline. I look at AI moving to the edge and the advancements that are going to need to happen in wireless technology to enable this. And I just think the pressure to move to 6G is going to continue. And we are looking to trials in the late 2020s deployment, commercial deployments, first deployment either late this decade or early next decade. And we're already seeing our 6G research business ramp significantly. Initially, that was almost exclusively with governments and research institutions in a university environment. But increasingly, we're seeing big market-making customers start to fund early 6G research. So it is happening. Obviously, small dollars relatively speaking today, but those things are kicking off and people are looking forward to deployments of this technology.

Okay. And I think when 5G happened, Keysight was in a much better competitive position than maybe the 4G cycle. Do you think the success you've had in 5G puts Keysight in a pole position for 6G deployments?

Yes. So first of all, we never take success for granted, right? And so -- but we're very much -- we know the things that we did that drove our relative success in 5G. We're working on repeating those. We started to pivot our own R&D investment budgets towards 6G years ago. And so that we will be ready with the appropriate 6G products when the market is ready for those things. And I think, to your point, the successes that we had during 5G, the deep customer relationship that we built during that period of time, the big installed bases that we have only serve to strengthen our position relative to where we were at the similar point in the 4G to 5G transition.

Okay. Great. Great. Maybe if we touch on A&D and kind of really focused on the defense side. There's been a lot of moving parts there. There's obviously a change in administration coming. So maybe talk a little bit about the outlook there. It is an area where I think satellite and space has been a good driver for Keysight? What's kind of the outlook for the defense line?

Yes. Again, I think this is an area where we're quite optimistic looking forward. Our business was down here in fiscal '24. But just would remind everybody, in '23, we were still shipping off a lot of backlog from the supply chain disruptions of '21 and '22. And so relative to the other end markets from a demand perspective, the aerospace defense business was by far the most stable of our end markets here, even though revenues did decline during the fiscal year. I think as we look forward, again, there does seem to be political alignment across the 2 parties in the U.S. of the need to invest in defense technology. We've seen that now in 7 consecutive U.S. defense budgets across parties of both presidential administrations of both political parties. So that's positive. We're seeing NATO allies commit to spend on defense at a higher percentage of GDP. So that's a long-term positive trend for this end market. And the unfortunate part is the conflicts that are underway around the world are only creating an increased sense of urgency of making sure you stay on the cutting edge of defense. And so I feel like we have a differentiated position. I feel like there is a broad alignment and the need to continue to evolve defense technology and that should position Keysight pretty well. And these markets, once they get going and get momentum tends to be sustained over time.

Okay. Great. Maybe if we hit the EISG businesses a little bit, start with semi. It does sound like, I think you guys have talked about this being one of the areas that's starting to inflect more positively. So what's driving that? How good of a recovery can there be? And I'm assuming things like U.S. production would be really -- are there really any potential very large green shoots for Keysight?

Yes. So semi is one of the areas that we're optimistic about looking forward. I think largely, we're doing the same thing that a lot of you are doing. We're looking at the big public statements of the large foundries around the world or the large semi capital equipment manufacturers around the world, all of them pointing to calendar '25 is likely to be significantly better than calendar '24. Now we have seen some small level of uptick in our business in the most recently completed quarter. It's not clear that this is the start of that. It could be or if it's just a blip. But I think as we look forward, certainly, there is -- certainly -I'd start with the long term. If you look at the long term, the outlook for semi is great, right? There's an insatiable need for more advanced chips at volume and at scale. The assurance of supply in different regions around the world is becoming a national security issue, which is driving ways of investment, often aided by government assistance. And all of that serves to create a longevity in this end market. And I think we have 2 particular niches in which we play in which we have highly differentiated solutions. And so I think we feel confident in our ability to capitalize on a market recovery when it comes. I think there's some question about when exactly that does turn on. Again, they're pointing to calendar '25. Does that mean first quarter? Does that mean fourth quarter? I don't think we really know the answer to that question. But again, I think specific timing aside, the overall outlook for the industry is very positive.

Okay. One of the areas that hasn't been great is auto. So can we -- what should we be looking at for a potential recovery in the auto business?

Yes. I think auto is probably the area of our business that continued to be the most under pressure, right? And again, if you think about Keysight's auto business, really 3 segments to it. There's roughly 1/3 of it that's historically been manufacturing with auto production down, that business has been under pressure. I think we have a high degree of confidence that, that business will recover as auto volumes recover. And so we're just kind of riding the cycle, if you will, in that place. The parts that's been the most robust for us is because kind of R&D development around electronic content in the vehicle, and that can be everything from infotainment systems to airbag triggers to probably the best opportunity for us long term, which is continuing to advance the level of autonomous driving capabilities within these vehicles. It's an area that we're very excited about because it's going to leverage the communications technologies that Keysight is very strong at, whether that's WiFi or LAN or 5G. All of those things are -- radars and LiDARs, all those things are going to be necessary to continue to advance autonomous driving. That leads you to the area that's currently the weakest, and that's the EV space. Obviously, we've seen a slowing in the adoption rate of EV vehicles. We've continued to see this mismatch between charging infrastructure and vehicle deployment. And I think there's -- as well as kind of an influx of relatively low-cost Chinese batteries. And I think that has a lot of the OEMs rethinking some of their investments in battery development, and we'll have to see how that plays out over time. I think the long-term trajectory for EV adoption is good. but it's unclear how exactly those batteries are going to be brought to market.

Okay. And then maybe just the last one here, general electronics. I'm guessing there's some definitely macro impacts here. So we think the macro is improving. Is that a good way to look at this business as we go through 2025?

Yes, it's such a broad cross-section of industries that fit into that general electronics buckets. Essentially, any industry that would have reason to be hiring electrical engineering talent that's not commercial communications, aerospace, semi or auto, right, which is a -- it's a broad cross-section. We did see some strength in the most recent quarter, driven around areas like med tech and advanced research in the university environment. And so those are 2 areas where we've seen some recent strength. I think as I look forward to '25 and beyond, there's a couple of things that we'll be watching. The business does have a higher exposure to our distribution channel than many of our other businesses. And we have seen elevated inventories and relatively slow distributor flow-through, through this downturn. That has been improving for multiple quarters in a row, but it has not yet normalized. And so I think once that distributor inventory situation normalizes, we'll get back to a more regular buying cycle, which should benefit this business. I think the other thing you need to watch is the business also has a little bit of a higher percentage of the business coming from manufacturing. And we know a lot of capacity went into place during the supply chain disruptions of '21 and '22. And so I think -- and some of that's no doubt excess capacity that needs to be absorbed. So we're watching global PMI very carefully. I think you probably need to see PMI sustained above 50 for multiple quarters to absorb that excess capacity. But once you do that, you'd expect those manufacturing buyers to reengage and you can start to see some strength in this end market.

Okay. Maybe just on China, you still have some business there. Obviously, Huawei came out a few years ago. What's your outlook for that region for Keysight? And do you think there's any -- we don't really know much about tariffs at this point, but do you think there's anything of note that is Keysight specific, good or bad with the potential for tariffs?

Yes. So China is certainly a watch for area, about a high teens percentage of our revenues are certainly within China. I think as it relates to tariffs, we're not necessarily thinking about tariffs in China as specific to Keysight, but really what are the macro impacts of those tariffs. We don't manufacture in China. Our supply chain is largely Southeast Asia based. And so those direct impacts are maybe less of a concern as they relate to China. But we've seen big numbers thrown around as China tariffs up to 60%, right? If that were to happen, that would have broad macro implications, which could impact our business. And even then, there's probably some positives and some negatives, right? On the negative side, what is the impact that, that has on broader growth. On the positive side, it could accelerate some companies to move their manufacturing operations out of China, creating opportunities for us in places like Mexico and Vietnam and other areas that are the beneficiaries as production moves out of the Chinese markets.

Okay. Great. I did want to talk on software and services. I feel like it snuck up on me almost 40% of revenues last quarter. So it's obviously been outgrowing the overall for Keysight. So talk a little bit about customers' appetite for a little bit more software type of sales. Obviously, it's a stickier business. It's got more recurring to it. So how do you see that business going? Are there other parts of the portfolio where you can inflect higher? So is it more concentrated? Or how do we think about that percentage going forward?

Yes. So as Tim said, roughly 40% of our business coming from software and services. That 24%, 25% software, about 15% services is the rough breakdown. And as you noted, those businesses are -- have a significant portion of those revenue streams that are recurring in nature. Those recurring revenues have done what we've expected them to do on the downside of the market, and they've held up really well, obviously, increasing the mix that we get from these businesses, allowing us to continue to operate in line with our operating model, deliver strong profitability, 26% operating margins, $900 million of free cash flow on the downside of the cycle. I think we look to continue to expand, particularly in the software areas. I think just given the fact that the high recurring nature and the faster organic growth that's coming from software, you will see a natural increase in the mix of software, but it's relatively slow. It's not like these things are growing 4x or 5x with the hardware is growing. And so it's incremental in terms of the mix shift. But our M&A pipeline and M&A funnel, and you can look at the transactions that we've done recently, does have a software bias to it and it allows us to accelerate that mix shift, right? One of the areas of focus for us is continuing to build out a broader suite of design engineering software tools. So Keysight for many decades has been in the EDA business with the #1 franchise in RF and microwave EDA tools. And we're looking at to build out a more robust set of engineering modeling capabilities. We bought a company called ESI, which brings computer-aided engineering capabilities to the auto space. We've announced our intent to buy the Optical Solutions group out of Synopsys, which will bring optical capabilities, which are core to this AI wireline opportunity that we've been talking about. So we continue to look to build out this set of tools. And the market phenomenon that we're observing is that our R&D customer, Keysight's classic R&D customer is looking to do more design work in software before they build expensive prototypes. And it's now practical to do that because these models, these things allow you to do digital twins. It becomes so much more accurate over the course of the last, say, decade that it's worth that investment. And so we're looking to build out that toolkit.

Okay. Great. Maybe it's a good segue into margins. Obviously, the gross margins are getting close to target, and you're talking about a 40% incremental operating margin coming back in fiscal '26 based on that 5% growth number. What are the kind of the moving parts there? And how much further room could there be in gross margin going forward? Is it really going to rely on that software piece growing more? Or is there still more manufacturing to R&D shift? What are the moving parts there?

Yes. So we'll start with gross margins. I think overall, at the Keysight level, we're pleased at how the gross margins in the business has held up on the downside of the cycle, roughly 100 basis point decline versus where we had been as our revenues have pulled back. And that's even more true on the CSG side where we've actually seen gross margins are higher now than they were in 2022, given the higher software content that exists within that business. I think one of the opportunities will be to once again improve the margins on EISG, has lower software exposure and also has a broader mix of gross margins within the hardware portfolio, some of our highest margin business as well as some of our lowest. And so a little bit more of a mix impact within that business that we'll be looking to benefit from as markets recover. I think beyond that, continuing to expand the software portfolio, continuing to -- one of the things we're proudest about, and this gets more to operating margins, is the fact that during this down cycle, we were able to sustain our investments in R&D, right? R&D dollars basically flat, but went from 15.5% of revenue to 17.5% of revenue. But as a result, we are really well positioned with a new pipeline of new products that are set to come out to hit the 6G window to enable AI. And I think as those new products come out, they tend to be more differentiated and give us an opportunity to drive gross margins even higher.

Okay. Great. I think we got 27 seconds, so we'll stop it there, but really appreciate it.

I appreciate it. Thank you for having us here today.

Thank you so much. Appreciate it. Thanks.