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

Hello, everyone. It's great to see you. My name is Jim Suva. I'm the IT hardware analyst here at Citigroup Investment Research. We're very pleased that you could join us for our global technology conference. First, a few housekeeping items. If you are media or press, please disconnect immediately. Media and press are not allowed to attend this. We're very pleased to host this session with Keysight. During this fireside presentation, you do have an opportunity to ask questions. You can press the ask a question little button there, and they'll send you the e-mail into my inbox, and I'll try to get to it if there's time. We also note that investors that are subject to MiFID II, please ensure you have the applicable research agreement in place. Also, there are associated disclosures with this at the Citigroup site as well as upon logins, and we do ask you to please take a look at Keysight's Investor Relations website. On the Keysight Investor Relations sites, there are safe harbor, risks and forward-looking statements that you should read as investors. I'd like to officially introduce our next presenter. This is Neil Dougherty. Neil is the Chief Financial Officer at Keysight. Neil, thank you so much for joining us here today.

And maybe we can talk big picture as we kick things off about end demand. Have you seen end demand trends evolve over the past year, 18 months, especially since the initial COVID 18? Has it disrupted, helped you overall? Kind of how has COVID impacted you in your company and industry?

That's a great question, Jim, and thanks for having us here today. We appreciate the opportunity to be at the Citi conference. I think as we look back, it's been a crazy 1.5 years, right? And if you look back since kind of the time when COVID really went global back in March of 2020, we've seen -- we've been through a lot, right? So if you look at our results last fiscal year, the impact on revenues was much greater than the impact on orders because in many of our end markets, demand stayed strong even as our own ability to ship, because of factory closings and other things, was more heavily disrupted. So I think we were very pleased in the initial periods of COVID that demand for some of these breakthrough new technologies, 5G, 400 gigabit remained relatively strong semiconductor through that period, while other industries did see more of a demand side impact. We certainly saw that in the auto industry, right, where demand for automobiles fell sharply in the early stages of COVID, and that fell through an impact of our auto business. We also saw some acceleration of decline in legacy wireless technologies like 4G as our customers had to make prioritization decisions. So they continued to invest across the board. They couldn't do that, so they had to pick, and what they chose to do was to continue to sustain those investments in 5G, sometimes at the expense of ongoing investments in legacy technologies. I think since that time, though, the COVID recovery has begun. And I think there's really 2 kind of phenomenon that are benefiting Keysight at this point. First is the overall kind of strength of GDP, right? You've got kind of broad market demand is very strong, and that's the rising tides raise all boats phenomenon. But then you've got secular drivers in a number of our end markets that kind of supercede that underlying demand strength and are really additive. And so that's, again, the ongoing investment in 5G, the deployment of 400 Gigabit Ethernet, the investment that's going on in 800 gigabits in the R&D lab. Semiconductor businesses obviously remained extraordinarily strong, driven not just by supply shortages, which are happening today, but also re-onshoring of semiconductor manufacturing, China looking to build the domestic semi industry, the move towards smaller process architectures below 5 nanometers as an example. The auto industry has rebounded dramatically, and we're once again seeing steady and growing investment in autonomous driving and in electric vehicles. And so -- and then maybe the last one is, we've now seen a first indication out of the Biden administration for a budget for next year. He's included increase -- further increases in defense spending. So that was kind of a big question mark that was out there, and we view that as a positive as we look forward. And so we are in a unique situation given that Keysight has such a diverse portfolio of industries that we serve. And right now, we are in a situation where most, if not all of those things are trending in the right direction, which is a pretty -- usually, it's a little bit more of a mixed bag.

Can you maybe break it down by your businesses, maybe aerospace, defense, government, commercial communications and industrial? Talk a little bit about what the trends you're seeing there, whether it be a little bit of, obviously, aerospace. I was on an airplane. Every seat was filled, which is encouraging, but the number of flights between different destinations like San Francisco and New York is significantly less. Can you talk about each of those segments a little bit?

Yes. Sure. So you started with aerospace and defense. So why don't I start there as well. I mean I think the first thing to note is we don't do a ton in the commercial aerospace portion of that market. Ours is much more government and defense contracting portion. And so we make -- what we primarily sell is a lot of communications tools and communications equipment into the defense industry. And as I've said, there's continued investment we've seen really over the last, say, 5 years, nice growth in U.S. aerospace defense budgets. And the U.S. markets make up a little bit more than half of our total aerospace defense sales with the balance being in the rest of the world, primarily the close U.S. allies, but some to other countries as well. And again, primarily a lot of communications type of equipment that we're selling into those industries. I think if you look at commercial communications, obviously, the single biggest driver right now is the move towards 5G, right? And we're now in that commercialization phase of 5G, but it's still very early days in the grand scheme of things, right? Even for those countries and service providers that are actively deploying 5G today, it's primarily in the big cities. Coverage is spotty. And so there's a lot of runway ahead in terms of making it such that a large portion of the world's population can reach into their pocket, pull out a 5G phone and expect to connect to a 5G network. I think within those markets, we're seeing a number of themes that have been additive to Keysight, things like O-RAN. We had the C-band auction here recently, and those types of things are increasing complexity. They're increasing the number of band combinations that are out there that need to be tested, and all of that is additive to the opportunity for Keysight. Within the electronic industrial markets, again, we tend to think of that as kind of general electronics, automotive and semi. I'll start with semi. Obviously, right now with the global supply shortages that are out there, the foundries and the semiconductor companies are out there adding capacity to meet this demand, and that's a great backdrop for demand for Keysight's products in the space. But you also have other trends, and I touched on them earlier, right? You have the U.S. and in some cases, Europe looking to re-onshore semiconductor supply. Again, it's all about assurance of supply, right? We're looking to re-onshore. At the same time, China is looking to build the domestic semi industry for the same reasons, right? And so that is additive to demand for Keysight products. And then as always, we look forward to technology innovation and technology transitions. And in the semi space, that's the move to sub-5-nanometer process architectures which is currently occurring in the R&D space. On the automotive side, you're really looking at electric vehicles today, more so than the AV, but I think as we look forward, we're very excited about the autonomous driving opportunity. Really, again, thinking about the auto opportunity through the lens of Keysight, if you rewind 10 or 15 years, Keysight didn't have a lot to offer this industry in the world of the internal combustion vehicle, right? It was a mechanical engineering device. And Keysight makes tools for electrical engineers. But as that industry moves towards electrification of the drivetrain towards autonomous driving, towards ever higher levels of electronic content, that's everything from tire pressure sensors to blind spot detection to infotainment systems, they're moving kind of lock, stock and barrel in the direction of Keysight's strength, and we see that as a great opportunity for us going forward. And then the last one is our general electronics business, which is our broadest market segment. It really does encompass a wide range of education, consumer electronics, health care, a really wide cross-section of industries. It tends to be most tightly linked to GDP. But as we're seeing here in 2021 and with forecast in '22 for great global GDP growth on a relative basis, I think that bodes well for our general electronics business as well. That was a long answer to a simple question.

No, no, because I asked about all 3 different segments. Maybe we could take a step back and talk about the supply chain. You build a lot of your products. But as the supply chain impact to your business, either from what you can get and procure or when you go to install it, are there some bottlenecks for supply chain?

Yes. So I do think it's true that maybe Keysight is a little immune to some of the really deep impacts of supply chain shortages right now. But we're certainly not totally immune, right? I mean we are being impacted. I think as you noted, Keysight is vertically integrated, right? We run our own fab here in Santa Rosa, California. We do a lot of our own packaging technologies, thin films and thick films and ultra-precision custom machining that puts us in a situation for where most of our highly specialized parts we make in-house and control the supply chain for. And that means that most of what we're looking to procure on the open markets are kind of high run, commercial off-the-shelf types of components. And it's not to say that those things are in a great demand situation at this point in time because they're not, but it's an easier problem to solve than if you're looking to solution supply for highly customized parts. So our revenue has been impacted over the past couple of quarters, but I think that revenue impact has been relatively muted. We're working an active list of components for our instrumentation that's in short supply. I think the other thing that benefits us is we're not a high-volume supplier, right? And we don't make tens of millions of anything. We make low-volume, high-mix type of manufacturing. And so in many cases, our suppliers are again able to get us small allocation, what would be very small allocations of parts to a high-volume manufacturer, but it's sufficient to meet our needs in many cases. And so we continue to work that. I think on the customer side, it's not so much supply chain constraints. I think it's really about managing expectations. I think one of the things that happens in times like this is customers know that supply is tight, they see lead times that are lengthening. And so we get a disproportionate -- or a much higher percentage of orders that are coming in now that are marked with a customer-required delivery date of ASAP. And you really know that in many cases, a customer doesn't actually need the thing tomorrow, right? And so it's working actively with those customers to understand the time lines that they're on, managing expectations about delivery, getting an agreement, and then living up to those commitments and getting product into the hands of our customers. And I think Keysight has been pretty successful in doing that. I still can't point to a single example where we've lost a material order because of the inability to deliver. And so again, I think on balance, relative to others, we're doing a good job of managing through this because of some of these unique aspects of Keysight's business.

In your backlog, it's been very, very healthy. Some people ask, is it actually being inflated or abnormally higher than sustainable due to COVID in these longer lead times.

Yes. So first headline I put out there is we believe the quality of our backlog is very strong, right? I often get questions from investors about double booking of orders and those types of things. And that's just not a phenomenon that we really see in our industry. I think the one place where we have seen customers placing orders noticeably further ahead than they otherwise would is in our semi business, right? And keep in mind, in the grand scheme, I think this is roughly 10% of Keysight's business. But semi is an industry where they're used to long lead times, right, 18-, 24-month lead times are not uncommon for some of these big pieces of semi equipment. Keysight's leads times in the space are much shorter, relatively speaking, but long by Keysight standards. Our semi lead times are in the 4- to 6-month kind of a range. And so we have seen our foundry customers place orders further ahead. They're looking to secure a spot in line. More than anything, they're looking to share their own road map so that we are ready with capacity and able to deliver. And so that's the one place where we have seen that. We have seen elevation of our backlog over the past couple of years. Last year, obviously, driven by our own factory shutdowns and kind of inability to ship as COVID was just beginning. More recently, it's shipment constraints around supply chain as we've been talking about. As I look forward, I don't envision that we're going to have a 2 or 3 quarter period of time where revenues elevate to a dramatic degree and we flush through all that backlog. I think it's going to work off over a much longer period of time. And ultimately, I think Keysight grows into a higher level of backlog. Obviously, we're approaching $5 billion in revenue this year. $5 billion company needs significantly more backlog than a $4 billion company. And so I think we're going to grow into that backlog over time and work it down far more slowly.

I get asked a lot about the peak of the 5G cycle. Where are we in the 5G cycle? And what do you think about the concern of people who think that when the 5G cycle peaks, the stock and business fundamentals are going to see pressure?

Yes. So I'd say a few things. I think we're still very early days, right? I mean look at the percentage of people you know that are operating on a true 5G network today on a regular basis. It's microscopic, right? Even for the countries that are actively deploying 5G today, Korea, Japan, China, you're still largely talking about spotty coverage and deployments in big cities only, right, at this point in time. And in the U.S., it's neighborhoods, right? The coverage is still very, very thin. You're also looking at a situation where most of the deployments today are sub-6 gigahertz. I think the whole high-frequency opportunity is largely in front of us. And we are hearing Japan and China and Korea looking into deploying those higher-frequency solutions. Also, you're going to see this move from non-standalone to standalone deployments of 5G occur over time. And so we believe there's a tremendous amount of runway ahead of us in terms of 5G. And then the other thing, it's kind of getting to the second point of your question, which is around what happens when 5G peaks. Again, I would just point to that the overwhelming majority of Keysight's business, particularly in commercial communications and in 5G, is focused in the R&D lab rather than the manufacturing line. And these R&D teams at these core communications customers are going to continue to work, right? So they're going to move from initial deployments of 5G, just like we saw in 3G. 3G went to 3.5G, went to 3.9G; 4G went to LTE, LTE Advanced. We saw these progressions of the technologies over time. Those R&D teams continue to work, though. Eventually, after multiple revs, low frequency, high frequency, standalone, non-standalone, next generation of 5G, they'll eventually migrate on to 6G. But the point is they're going to continue to work and Keysight is going to be there to continue to provide them the tools and solutions that they need to stay with this market as it evolves over time.

And we hear a lot about 5G, and I think people think about like their cell phone. Or right now, mine shows 5G, even though my cell phone doesn't have 5G in it, so it's kind of a version of enhanced 4G. What about like millimeter wave and O-RAN? Are any of these catching on?

Yes, certainly. Let's talk about O-RAN for a minute, right? I mean I think if you rewound a couple of years, there's a lot of pressure that was being put on, for example, European providers to not adopt Chinese technology as they put in the radio access portion of their networks. With everything that's going on from -- has occurred over the last couple of years from a trade perspective, and now, we are now seeing European operators start to adopt O-RAN, right? And so O-RAN has proven to be a significant opportunity for Keysight, right? We like complexity. And as you disaggregate this radio access portion of the network, you add complexity, you add hand-off points, you add customers, Keysight customers, but you add many, many different customers or companies that are looking to do a piece of what was a market that had been dominated by 3 or 4 different players. And so that's created a tremendous opportunity for us. I think the -- we've heard a lot about kind of delays, getting to the second part of your question, which is about millimeter wave delays in the deployment of millimeter wave. I think from a Keysight perspective, our customer base are continuing to invest in research and development around these technologies. And so we've seen steady investment in that area, and we do believe that ultimately, you are going to see meaningful commercial deployments of high-frequency 5G networks. That's what you really need to get the performance advantages that have been promised from 5G. And so that opportunity is ahead of us. And Keysight has thought ahead, right? A lot of -- most of the sub-6 gigahertz tools that we put into the marketplace are upgradable to higher frequency. So we provided our path to our customers to stay with Keysight, to stay with the instrumentation and the user interfaces that have been working for them, and they can easily upgrade that instrumentation to higher frequencies as their needs evolve. And so we see that as not only a way to continue to enable the success for our customers, but to continue to provide a growth revenue stream for Keysight going forward.

Neil, can we focus a little bit on the automotive sector. I drive a car, my wife drives a car. It goes beep, beep, beep before I hit something, it stops itself. It has a lot of sensors. It calibrates the air conditioner of 68 degrees where I sit and 72 percent (sic) [ degrees ] where my wife sits. And it calibrates airbags for my children in the back, do they fully go on or not? Just a lot of things in the automotive sector. And then you think about cars that are plugging in and charging in the future to make sure that they don't get overcharged. Can you talk about the auto sector and what this could mean to you potentially? Because I believe it's a pretty small part of your company. Is there an opportunity there? And where would you be in the auto sector as far as so people can grasp, while they drive a car, where exactly you would have been part of that car?

Yes, it's a great question. And you're right. The auto business today is a relatively small one within Keysight, but it's one that we're very excited about. And as we look at kind of the growth opportunity for auto, we think about it in terms of not quarters or years, but actually in terms of decades, right? Because when you think about the amount of time that it takes to turn over the auto install base with people buying a car every 10 or 12 years on average, this is a very long period of time. And we're still very much in the infancy in terms of the percentage of cars that are sold today that are truly electric, or even hybrid, for that matter, is still a very small percentage, and that's obviously the direction that we're heading. You're getting regulatory help from governments to move the industry in that direction. And ultimately, for safety reasons, you're going to move towards more complete autonomous -- levels of autonomous driving. So you already noted, Jim, through your own personal experience, how you've seen this expanding electronics content that exists in these vehicles today. All that electronic content needs to be designed and developed and ultimately tested to make sure that it functions appropriately. And Keysight makes tools that essentially enable electrical engineers to do this kind of work. I think the big opportunities for us in auto are, as you mentioned, around electrification of the drivetrain in electric vehicles and then around autonomous driving. If you think about electric vehicles, a lot of the work that's being done occurs in the battery space, how do you make batteries that are more efficient, both in terms of how long they can go on a particular charge, but also how they can last an ever-increasing number of charge dissipation cycles, and Keysight has a power management business where we make tools that can essentially help battery developers make better batteries. We also acquired a company several years ago called Scienlab which makes products that we can sell actually to car manufacturers, electric car manufacturers and to the folks that are making charging infrastructure. And the problem that they're looking to solve is how do you get a safe and effective connection between the car and the charging infrastructure when a car of unknown brand pulls into a charging station with charging equipment of unknown brand. You need to make sure that those protocols, the safe transfer of power, the transfer of payment, the transfer of software updates all happens effectively, and Keysight makes an instrument that can essentially sit between the car and the charging infrastructure in a lab environment to help make sure that, that happens effectively. And then on autonomous driving, I think that the opportunity is even bigger, right? Autonomous driving is going to be enabled through a set of very complex communications using multiple standards, everything from Bluetooth and Wi-Fi to 5G. You've heard about C-V2X communications. In other words, how does the vehicle communicate with everything, right, communicate with infrastructure and pedestrians and traffic signals and the other cars on the road. And as you know, Keysight is the #1 player in the communications test space, and there's going to be a huge market for that as you look to enable autonomous driving. The other component of that is complex RADAR and LiDAR systems. Keysight has been providing tools to develop complex RADAR systems to aerospace, defense markets for decades. And that's now being able to be leveraged into the auto markets. The work that we've done in commercial communications around 5G is being able to be leveraged into the automotive markets. And so there's a tremendous amount of synergy that comes across these markets as we develop tools for one space and then see that technology applied in other markets.

One thing that investors focus on a lot of is, a metric you talk about is book-to-bill. They view that above 1 as being positive, below 1 negative, but also importantly, kind of the rate and level of where it's at. I'm wondering, has COVID impacted your book-to-bill or you mentioned semiconductor shortages, not necessarily from what you experienced, but maybe others? Is it impacting book-to-bill? And what's your kind of visibility or how should we think about book-to-bill? I know you had some issues with Huawei, Huawei coming in and out of those comparables, which I think is completely behind us now. But can you just spend a few minutes talking about book-to-bill?

Yes. So we do pay attention to the book-to-bill ratio. And again, ideally, you're in a situation where your orders continue to outpace your revenues as a growing business. I think we certainly have seen some impacts here as a result of COVID, right? Early on in COVID, we actually had to shutter our factories for a period of time. So that held down the denominator, the revenue component of that while demand for -- in our end markets, as I've already talked about, stayed relatively strong. So back in Q2 and Q3 of last year, we put up some pretty aggressive book-to-bill ratios, right? I think you saw that they normalized. Now I think to the extent that there is a much smaller constraint, there's still -- because of supply chain, there's still a constraint on revenue that maybe book-to-bills are a little bit elevated. I think I've talked about how we expect to work through this backlog over time. So you could see some compression of those book-to-bills as that happens. You may even see a quarter or 2 where it goes below 1. In this case, that wouldn't signal any lack of health of our industries and our demand environment, it may just signal the fact that the supply chain and other manufacturing constraints have let up and we've been able to clear some backlog, right? So that may be a phenomenon that we see going forward. You mentioned Huawei. Huawei has been a significant customer for us in the past. The most recent set of restrictions that were put on Huawei around this time last year were pretty complete. And so it's pretty dramatically limited our ability to ship to Huawei. While it's not 0, They went from being, call it, a 5-plus percent customer for us to something that's substantially below 1 going forward. There's only very small portions of our portfolio that we're currently able to sell to them. And so over the last year, we've been lapping some pretty tough comps on both order and revenue side with regard to Huawei. But as you noted, that's now -- it will be totally behind us at the end of Q4. But even in Q4, the numbers are small enough that we shouldn't have to talk about it. And so we're very pleased. And I think most notably is that even with essentially losing a, I'll call it, a 5% customer, Keysight's continued to grow over this period of time. Not only has Keysight continued to grow, our commercial communications business has continued to grow. Our China business has continued to grow. And I think that speaks to the underlying demand for Keysight technology and Keysight projects -- products, excuse me.

Yes. Can we switch over to software and services? A lot of times, people think about your company as more of a hardware company. But can you talk more a little bit about software and services and maybe how big or important they are of your company? Maybe quantify the percent or size because it's quite material. And I've noticed over time for covering your company for so long, it's not all about hardware anymore.

Yes. I think this is one of the most underappreciated aspects of Keysight. I don't think people realize the magnitude of our -- specifically our software business, but services as well, right? Today, 30-plus percent of our revenues are coming from software and services. About 20% of that is from software with the balance from services, if you're looking for quantification. And across both of those product lines, right, there's about $1 billion -- or north of $1 billion now of annualized recurring revenue. And so this is a much different than Keysight's kind of history, right? Much different than the company looked at the time of our spin, for example, which was just 7 years ago. And so I think it gets to the stability of our revenue streams. I think it gets to our ability to provide differentiated solutions and provide value to our customers beyond hardware, right, ensuring uptime of their instrumentation, providing them with the software that's necessary to analyze their industries to, in many cases, emulate and simulate performance of their instruments in real-life networks. Take 5G, for example. And so we have done a good job of adding software content over time and dramatically increasing the proportion of those businesses as a percent of the total. And both our software and services businesses continue to grow at rates that are above the company average. And so we continue to add. Ever-increasing portion of our mix is coming from these businesses, which is a trend that we expect to continue at least as far out as we can see in the marketplace.

Neil, my memory might be off on this, but I think I recall both you and Ron mentioning, or maybe it was just Ron, about you see a potential for still some margin expansion. Is my memory right on that? And were you referring to gross or operating? Is that due to more sales off of a kind of fixed base and leverage? Or -- well, you just talked about the services and software being more accretive relative to hardware.

Yes, it's a great question, Jim. So on our most recent earnings call, so one of the things that we've gotten a lot of questions about is at our Analyst Day, which was March of 2020, so about 18 months ago, we outlined a new targeted operating model that we hope to get to. And we said that we'd hope to get to 26% or 27% operating margin and achieve that objective by fiscal -- our fiscal '23. Well, we've actually achieved that objective 2 years early, right? We will be at 27% operating margin, so the high end of that range here this fiscal year. So we're getting a lot of questions from folks about, is this the end? And I think -- so we did make the comment that both Ron and I do see the opportunity for Keysight to continue to drive margins northward. I think the comment was specific to operating margins. But to your point, Jim, I think a key component of us getting to increased levels of operating margin is going to come from us continuing to drive gross margin further northward from its current levels in the mid-60% range. I think our business continues to grow nicely. We continue to see strong demand in our end markets. And as you know, we've talked about a model where when our business is growing, we can deliver 40% operating margin incrementals on that revenue growth. And so there is opportunity as we continue to operate to that model for us to continue to drive operating margins north from where they are today. And you hit on some of the drivers, right? We're seeing increasing software content. We're seeing increasing services content. Increased differentiation in the marketplace, right, enables us to monetize that because we're adding value to our customers. At the same time, we look to leverage our G&A infrastructure as the overall business grows. So -- and benefit from the economies of scale that come from being a larger business. So I think there's no one single driver. I think we've got numerous things that are going to enable us to continue to deliver to that 40% incremental and continue to drive operating margins north even from where they are today at 27%.

I actually got e-mailed in a question that I'm going to combine and let you know, with a couple of other investors, they all kind of asked the same thing. And they said, Neil talked about a vertical integration model which is a little bit different and helps during supply component shortages. Have you been able to gain share in the 5G test equipment by being able to deliver more if your competitors are seeing shortages of some chips, or the design cycle is just simply so long that you would not see some share shifts?

Yes. So certainly, our vertical integration, the fact that we run our own fab, do our own packaging of subcomponents has certainly been a differentiator that has enabled us to be able to continue to [ show ]. On the surface, I think it makes sense that, that's been a contributing factor to us gaining share in 5G and in other markets. But I think the bigger factor is the differentiation that we have in our technologies in those spaces, right? We got a first-mover advantage by investing early in 5G and got kind of embedded with the key customers and the key market makers across the industry very early. And we've continued to innovate, we continue to move the bar forward in terms of the value that we're providing to customers. And in many cases, those big customers are now kind of committed to the Keysight platform and the Keysight portfolio. So as the industry is continuing to move forward, we're in a natural position to capitalize on that. And I think the fact that we have a vertically integrated supply chain and the fact that we can -- are in a better position to deliver is just additive. It's -- in and of itself it's not driving this market share gain, it's the differentiated portfolio of technology that is ultimately enabling our success in the marketplace. I think we do view that vertically integrated supply chain and the fact that we control that, particularly for a business like ours that's high mix, relatively low volume, is a competitive differentiator. I mean I think we view our fab in Santa Rosa as one of the crown jewels that enables us to achieve some of the technological advantages in the marketplace that we've been able to achieve. It's co-located with our R&D team. So we can run R&D lots through the fab on an expedited basis that you wouldn't be able to do if you were partnering with a foundry type of environment. And again, I think truly a crown jewel of the company in terms of our ability to differentiate along technological spectrum.

Neil, it's amazing how fast time goes. We've got about 5 minutes left. I do want to touch on capital allocation, you being CFO. You're a high-growth company. You have a stock buyback and there's probably a chance for even more M&A. That's been very successful. I think you've done well over a dozen companies of acquisition in a handful of years since you've spun out from Agilent. Can you talk about your capital allocation policies and programs of how we should think about you deploying cash?

Absolutely. So I mean I'd start with kind of the basics, the blocking and tackling, right? The #1 priority that we have as a company is to make sure that we're continuing to make the organic investments that are necessary to drive the profitable growth of Keysight going forward. What that's meant historically for us is obviously increasing levels of R&D investment. At the time of our spin, we were investing like 12 points of revenue in R&D. That's up to 16%, 16.5% today, which is a level that we're comfortable with, but we want to make sure that we continue to seed money towards those positive MPV R&D projects that are driving technology forward. And I think notable in our space that we have such in-depth relationships with our customers that there's a lot of demand for our customers for development work. So unlike other industries where, to some extent, you're out there doing market research, you're defining products and hoping that you defined products correctly such that they hit the market desires. In many cases, our R&D programs are customer-driven. And so it derisks those R&D investments for us. The second big place that we're investing through the P&L for organic growth is in our sales force, right? We've been on a multiyear effort to double the number of direct sellers that we have out there calling on customers. We are almost to the end of that investment. We have a little bit more to go here as we go into fiscal '22. Beyond that, Jim, as you noted, we do have a strong appetite for M&A. I think the most notable thing there is that we stay disciplined with regard to our strategic and return hurdles. I think, as you noted, I think we've done 14 acquisitions since our spin in early -- in late 2014, been very successful in adding growth and profitability through those transactions. And I think right now, the idea -- what I've said in the past is idea generation is not the challenge, right? There's a lot of interest -- there are a lot of interesting targets that are out there. But valuations are high, and so the key is to stay disciplined and look for the right opportunity at the right price that enables us to generate strong returns for our investors going forward. = And then beyond that, we're returning capital. We're conscious of the cash balance. We're trying to limit further growth from where we are today. We've -- over the past 3, 4 or 5 quarters, we've taken roughly 50% of free cash flow and returned it to shareholders. Roughly 50% has gone into M&A. And so I think when we're not finding targets, we're proving that we're willing to and able to quickly return capital to shareholders via buybacks, and I think you'll see that behavior continue as well.

And Neil, in the last concluding minute, what's the 1 or 2 things you really want investors to know about why they should be owning Keysight as an investment in their portfolio?

Yes. I think that the one thing that I'd like to leave investors with is the kind of the confidence that we have in the demand side of the equation for Keysight. I think there are a number of secular drivers across our end markets. We've talked about them today, 5G, 400 gigabit, 800 gigabit, autonomous driving, electric vehicles, semiconductor shortages, onshoring. All of those things are kind of underlying secular drivers that go beyond the macro cycle that are going to drive demand in these industries. And I think it -- not only is Keysight, a diverse business across multiple industries, but those industries all have strong secular demand that provides resilience for Keysight as we look forward. And so we're very encouraged about the prospects for Keysight over kind of that intermediate to long-term horizon.

Well, I want to thank Keysight so much for sending their Chief Financial Officer. And Neil, thanks so much for joining us from your headquarters in Santa Rosa. And I really hope next year we can do this in person because I'd love to have you on stage next to us and doing some in-person meetings.

Thank you. Thank you, Jim. Thanks for having us. It's been a pleasure.

Bye-bye.