This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.
spk03: Greetings and welcome to the Teradyne second quarter 2023 earnings call and web chat. At this time, all participants are in a listen-only mode. A brief question and answer session will follow the formal presentation. If anyone should require operator assistance during the conference, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Andy Blanchard, Vice President of Investor Relations. Thank you, Andy. You may begin.
spk09: Thank you, Operator, and good morning, everyone, and welcome to our discussion of Teradyne's most recent financial results. I'm joined this morning by our CEO, Greg Smith, and our CFO, Sanjay Mehta. Following our opening remarks, we'll provide details of our performance for 2023's second quarter, along with our outlook for the third quarter. The press release containing our second quarter results was issued last evening. We're providing slides on the investor page of the website that may be helpful to you in following the discussions. Replays of this call will be available via the same page after the call ends. The matters that we discussed today will include forward-looking statements that involve risk factors that could cause Teradyne's results to differ materially from management's current expectations. We encourage you to review the safe harbor statement contained in the earnings release, as well as our most recent SEC filings. Additionally, those forward-looking statements are made as of today, and we take no obligation to update them as a result of developments occurring after this call. During today's call, we'll make reference to non-GAAP financial measures. We've posted additional information concerning these non-GAAP financial measures, including reconciliation to the most directly comparable GAAP financial measure, where available on the investor page of our website. Looking ahead between now and our next earnings call, Teradyne expects to participate in technology or industrial-focused investor conferences hosted by KeyBank, Jefferies, Deutsche Bank, Goldman Sachs, Citi, and Evercore ISI. Now, let's get on with the rest of the agenda. First, Greg will comment on our recent results and the market conditions as we enter the third quarter. Sanjay will then offer more details on our quarterly results, along with our guidance for the third quarter. We'll then answer your questions. This call is scheduled for one hour. Greg?
spk10: Thanks, Andy, and good morning, everyone. Today, I will summarize our Q2 and first half results, comment on the current business conditions, and our view of the second half. Sanjay will provide the financial details on Q2, our outlook for Q3, and offer some financial guideposts for the rest of the year. Second quarter sales were at the top of our guidance range as supply constraints eased. Earnings were above our guide on higher gross margins. At the halfway point of the year, overall company performance has unfolded as expected, but at the segment level, test was incrementally stronger and robotics weaker. In semiconductor test, we are four quarters into a correction cycle driven by excess inventory, which has hit the mobility part of the market hardest. Automotive demand has remained strong, and in memory test, the growth of DDR5 and HBM devices for data center applications are driving retooling. In our wireless and system test businesses, demand remains muted, unchanged from our April view. Robotics demand has softened over the past three months with worsening PMIs and the short-term impacts of the transformation of the UR distribution channel. Looking forward to the rest of 2023, we estimate the 2023 SOC test market will be $3.7 to $4.1 billion, down 13 to 21% from 2022, but up from our outlook in April. The continued weakness in mobility has been offset by sustained strength in the automotive segment. Also since April, the accelerating build-out of AI-enabled cloud computing is driving test demand for compute and networking. In memory test, we expect the market will be at the low end of a $900 million to $1 billion range we described in April. In this market, the impact of AI is evident, especially in the HBM DRAM segment, where we see incrementally stronger test demand for Magnum products through the second half of 2023. Although HBM represents a small portion of the overall memory market, it is in a rapid growth phase, moving from under 5% of the test market last year to 10 to 15% of the market in 2023. And our HBM share is higher than our overall memory share. This technology-driven strength is offset by weaker capacity buys, especially in flash for mobile applications. Despite the current downturn, we remain confident about the future of the semiconductor test market. The primary growth driver is an insatiable demand for increasing device complexity. This can be seen in cloud and edge AI applications like ADAS systems, spatial computing, and privacy-focused consumer applications. These require enormous compute power, higher wireless data rates, and more sophisticated power management. We expect the automotive TAM will grow at a faster rate than the rest of the SOC market through this midterm, driven by the growth of EVs and hybrids, broader adoption of features such as ADAS, cabin lighting, and infotainment, and the high test intensity required to achieve automotive quality levels. The trend to vertically integrated producers, or VIPs, is a fundamental disruption to the computing segment and is beneficial to Teradyne. Companies that provide cloud computing, cloud AI, and edge AI are seeking to differentiate their solutions by taking control of chip design. Efforts that began a few years ago are now beginning to proliferate in data centers and vehicles, and this trend will accelerate. Over the midterm, we expect the VIP portion of the compute segment will grow faster than the overall compute segment, and our share will be higher than in traditional compute customers. The semiconductor industry has a process technology roadmap that supports these new, more complex devices. 3 nanometer is ramping now, and 2 nanometer and gate all around are coming soon. In advanced packaging, we see broader adoption of stack die for high bandwidth memory and chiplets for processors. These new technologies will drive longer test times and retooling to test new interface standards. Rolling that all up, our growth hypothesis for the semiconductor test market remains unchanged. A similar complexity roadmap drives our wireless and system test businesses. In wireless test, new standards like Wi-Fi 6E and 7 require new or upgraded test equipment. In system test, increased device complexity is broadening the adoption of SLT, and a very solid pipeline in defense and aerospace puts our system test group on a solid foundation for growth. Shifting to the robotics portfolio, I want to take a moment to highlight that Ujwal Kumar joined Teradyne earlier this month as the new president of our robotics group, which includes UR and Mir. Ujwal joins us from Honeywell, where he ran the process solutions business. His industrial automation and software background, combined with the deep experience in building businesses through organic and inorganic growth across multiple end markets, is a great addition for Teradyne, and we are delighted to have him join our leadership team. In the second quarter, robotics demand softened significantly. The trends that we noted in April have continued and intensified. Challenging economic conditions, particularly low PMIs in Europe and the U.S., have resulted in lower demand in our highest revenue regions. We have previously noted that channel transformation work at UR was having an impact on pipeline conversion. This trend has continued in Q2. Large account development and the build out of the OEM channel are progressing well, but not quickly enough to cover for the market softness in the quarter. With lead times under five weeks, change in end market demand, especially with our current high exposure to small and medium sized businesses, is felt quickly. However, there is no shortage in interest in our human scale automation products. We saw record lead generation from two major automation trade shows in the quarter, but there's a clear reluctance from customers to place orders in the short term. As a result, we are now projecting full year revenue for our robotics group to be flat to down 10% from last year. Despite the difficult macro environment, and the short-term impact of our distribution changes, we believe that we will emerge from 2023 in a stronger position in robotics. On the new product front, we began shipments in the quarter of our higher payload, longer-reach UR20. Customers serving welding and metal fabrication and palletizing across a number of industrial verticals have driven demand for the UR20 and a triple-digit unit backlog. We will see our first UR20 revenue in Q3 and ramp shipments through the second half. We also introduced MIR Insights, a cloud-based tool to enable MIR customers to monitor and optimize large fleets across multiple workflows and sites more effectively. We are transforming the UR distribution channel in 2023. Recall we are complementing our existing distribution channel with direct touch coverage at large customers and adding OEM partners that have high long-term growth potential. In the short term, it appears this shift is slowing sales from distributors that were dependent on a high level of UR sales support. Longer term, however, the change puts our focus on customers with the highest revenue potential. Although these changes will take several quarters to yield, we're confident that we're on the right path. We've added 28 new OEM partners so far this year and are working with our distribution partners to directly engage with over 200 large customers. At MIR, our focus on large accounts is yielding good results. This year, our installed base at large customers has grown three times the rate of our overall installed base. We view robotics as a long-term opportunity. The market drivers are clear. aging populations, rising wages, labor shortage, and the reshoring of production to reduce cost and cycle time. We have innovative, market-leading products serving a market that has the potential to grow to tens of billions of dollars per year. Our customers have already demonstrated the value of our products and ecosystem in their operations and they're partnering with us to extend our robots' performance to expand the range of tasks they are planning to automate. Teradyne brings the foundational expertise in engineering, operations, and customer support needed to enable UR and MIR to become premier providers of human-scale automation. We are putting the structure in place to support a billion dollars in profitable sales by the end of the midterm. This is a long-term project, but we are seeing the early signs of this work yielding. Wrapping it all up, our test businesses are performing better than planned through the first six months, and we expect that performance to continue through the second half. Automotive and memory are our strongest markets in tests this year, and a combination of inventory reductions and new products should enable the mobility market to recover next year. Our robotics business was below plan in the first half, And while we expect a stronger second half for the group, we expect to be below our growth and financial plan for the full year. Our plans to transform our distribution and expand our product line are on track. We are carefully managing our spending in this business while we execute these changes. With that, I'll turn things over to Sanjay for the financial details.
spk01: Sanjay? Thank you, Greg. Good morning, everyone. Today I'll cover the financial summary of Q2, provide our Q3 outlook, and full year planning assumptions. I will also update you on our supply chain and resiliency progress. Now to Q2. Second quarter sales were $684 million, which was at the high end of our guide, with non-GAAP EPS of 79 cents, which was above our high guide of 74 cents. Non-GAAP gross margins were 58.8%. above our guidance due to deferred resiliency costs until later in the year and improved product mix. Non-GAAP operating expenses were $251 million flat with the first quarter. Non-GAAP operating profit was 22%. We had one 10% customer in the quarter. The tax rate excluding discrete items for the quarter was 16.5% on a GAAP basis and 17.2% on a non-GAAP basis. Semi-test revenue for the quarter was $475 million, with SOC revenue contributing $355 million and memory $120 million. As Greg noted, we continue to see SOC strength concentrated in the auto end market. Memory sales were weighted towards technology-driven buys as the industry ramps new, higher-speed devices like protocol flash for smartphones, DDR5, and HBM DRAM for server applications. System test group Q2 revenue was $94 million with $38 million in storage test, which has been down all year on continued low SLT and HDD production demand. In wireless test, revenue was $44 million in Q2 with ongoing low demand from both PC and smartphone end markets. Now to robotics. Revenue in Q2 was $72 million with UR contributing $58 million and MIR $14 million which was below plan as Greg noted. FX did not have a material impact on our top or bottom line results. Shifting back to the company level financials, our free cash flow was $104 million in the quarter. We repurchased $135 million of shares in the quarter, paid $17 million in dividends, and settled $2 million of debt. We have $33 million of convertible debt remaining. We ended the quarter with $813 million in cash and marketable securities. Now to our outlook for Q3. Q3 sales are expected to be between $650 and $710 million with non-GAAP BPS in a range of $0.61 to $0.81 on 163 million diluted shares. The third quarter guidance excludes the amortization of acquired intangibles and restructuring and other charges. This outlook is above our April view at the company level, but under the covers, test is incrementally stronger and robotics softer. Third quarter gross margins are estimated at 56% to 57%. OpEx is expected to run at 35% to 38% of third quarter sales, down slightly from Q2. Non-GAAP operating profit rate at the midpoint of our third quarter guidance is 20%. a little more color on our revenue and gross margin profile for the second half. Our revenue guidance excludes approximately $35 million of test demand tied to supply constraints in Q3, which we expect to resolve in Q4. As a result, we expect Q4 revenue to be similar to the Q3 level. If we clear some of these supply constraints earlier, some Q4 revenue and resulting profits will shift into Q3. That will put second half revenue at roughly 51% of the full year. Shifting to gross margins. Our long-term model has gross margins at 59 to 60%. In 2021 and 2022, we were in our model range with margins at 59.6% and 59.2% respectively. In April, we noted that we expected full-year 2023 gross margins to be at 57% to 58% due in part to transitory costs of adding new manufacturing sites for our test products, geographically multi-sourcing components and sub-assemblies to reduce supply line risks. We're also developing multi-country capability of some of our design services work. In past calls, we noted expected margins to be lowest in Q1 and improve through the year. However, the timing of these resiliency costs shifted to later in the year, which was a driver of outperformance on gross margin in Q1 and Q2. Since that spending will now happen in the second half, our margins in Q3 and Q4 will drop to reflect these costs. Net-net, we continue to expect 57 to 58% gross margins for the full year. However, some of these transitory resiliency costs have been deferred until the second half of the year. Regarding OpEx for the full year, no change from our prior guide as we expect full year 2023 OpEx to be roughly flat compared with 2022. Now to profitability for robotics for the full year. As Greg mentioned, we expect lower robotics sales for the year. Even at these reduced revenue levels, robotics gross margins are above the corporate average, and while we expect UR to be profitable, the robotics group overall is now expected to have an operating loss for the year. Our gap and non-gap tax rates are forecasted to be 16.5 and 17% respectively in 2023. Summing up, our stronger test businesses in Q2 more than balanced weaker robotic shipments, leading to sales at the high end of guidance with gross margin improvements yielding profits above our high guidance. We're executing our plan to improve the resiliency of our supply chain and engineering services. We're modeling full-year gross margins in the 57% to 58% range, in line with our April outlook, and are on track to hold our full-year OpEx flat with last year. We have a strong balance sheet to support organic and inorganic growth, while we continue returning excess cash to shareholders through share repurchases and dividends. With that, I'll turn the call back over to Andy. Andy?
spk09: Thanks, Andre. Operator, we'd now like to take some questions, and as a reminder, please limit yourself to one question and a follow-up.
spk03: Thank you. We'll now be conducting a question and answer session. If you'd like to ask a question, please press star 1 on your telephone keypad. The confirmation tone will indicate your line is in the question queue. You may press star 2 if you'd like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star key. One moment, please, while we call for questions.
spk11: Thank you. Our first question is from Timothy Arcuri with UBS.
spk03: Please proceed with your question.
spk11: Thanks a lot. I had a question on the revision of the TAM, SSC TAM. So you took it up about $300 million. Can you talk about where that's coming from? Is that all in autos? So autos goes from six to nine this year. And then also in that, I think you previously had said that your share would be in the kind of 38 to 39% range this year? Is that still the expectation even on this higher TAM?
spk01: Sure. Roughly at the midpoint, last time we got to the SOC TAM at the midpoint, about 3.6. Now, obviously, as Greg noted, 3.7 to 4.1. So roughly a $300 million increase. And that's really tied to the compute market as a key driver. And there's, you know, we're rounding to hundreds of millions, but it's mainly in compute. Then I would say from a share perspective, our view on the full year is that we're roughly flat and may tick up a bit.
spk11: Okay, got it. Then can you give a sense of just storage for the full year and sort of how you see that evolving next year? I think the expectation is still that it's going to be very soft in Q4 and And then, you know, do you think that's going to rebound as you kind of get into the first part of next year?
spk10: So this is Greg. Yeah, we think that storage is sort of running along at a minimal ship rate. We're going to expect to continue at that rate. And there might be a recovery in 2024, but I don't think it would be a first half thing. I think that there's still – a fair amount of softness in the end market for HDD, a fair amount of oversupply. So it's really only when we're going to see significant capacity inflection. When Hammer comes online and demand increases, that will drive higher test times and ultimately drive additional capacity purchases. But that's probably further out in time than the first half of next year.
spk03: Thank you. Our next question is from Mehdi Hosseini with SIG. Please proceed with your question.
spk05: Yes, thanks for taking my question. Just going back to SOC, Tam, I want to get an update on how you see the transistor density growth evolving into 2024. I remind you a couple of years ago, The SOC test was benefiting handsomely, and then we hit the pause, and I'm just trying to get an understanding how that growth rate is evolving to next year, and I have a follow-up.
spk10: Good morning, Mehdi. This is Greg. So the semiconductor process roadmap seems to be pretty much in line with what we saw before. Perhaps a little bit of ramping three nanometer a little bit more slowly than we would have expected maybe back in 2020 you know, but in terms of compared to recent events it's pretty much on track. The The key thing is that those those complexity transitions are occurring in a market where end unit volumes, especially in mobile are significantly down so. Even though the complexity increases are happening and we are seeing sort of the expected changes in test time, there is a fair amount of idle capacity because of the unit declines that's being filled versus it driving new tester purchases. So I hope that answers your question, but basically we think that the pace of new process adoption is kind of going as we expected.
spk05: Does that imply that the growth rate would accelerate in 2024, or would that remain the same?
spk10: So I think our view of 2024 is that the number of high-volume 3-nanometer devices is definitely going to increase. So there will be more chips on 3-nanometer. And the sort of third-party reports that we see around end market is that it's up but not up dramatically. So I've seen numbers between 3% and 5% in terms of smartphone unit volume increases. So we definitely think that there's going to be 3 nanometer coming from more fabs, there are going to be more parts on 3 nanometer, and volumes are going to be up. So we think that 2024 is going to be incrementally stronger than 2023.
spk03: Thank you. Our next question is from Vivek Arya with Bank of America.
spk13: Please proceed with your question. Thanks for taking my question. Greg, I think in the last call you said test day utilization in Q1 was, you know, at kind of a generational low. Where do you think it is now? Because when I look at, you know, where consensus expectations are for Teradyne next year, they are for almost 25% case growth. And I'm trying to understand, you know, you know, when do you start to see the orders to justify those kind of growth expectations if utilization still remains at these kind of generational lows?
spk10: Good morning, Vivek. Thanks for the question. So the good news is that utilization hasn't gotten any worse. So the, you know, and frankly, it's up a bit, you know, so in the two to five percent increase in terms of utilization of Teradyne equipment. So our view is really of our own fleet. The thing that I will tell you is that there's a significant disparity in utilization between IDMs and subcons. So utilization inside of IDMs is, you know, in the upper 80s, and that's well into the range where our customers are actively adding capacity, and that's really the strength that we see in automotive. In subcons, the utilization is nearly 20 points lower, but it is heading in the right direction. So I think it's a really good question in terms of how much that idle capacity will impact the size of business in 2024. Um, so we, you know, we, we are, uh, you know, that's sort of factored into our, our plans. We think 2024 would be stronger, but we really don't have a great idea about how much stronger it will be.
spk13: And, um, then maybe one follow up on the IA business, um, right. It's very innovative. The value proposition makes sense, but it has sort of consistently underperformed its potential. And I don't think it's ever been profitable. And I'm curious, what has been the disconnect? Is it really a structural growth business? Because it's actually declined two out of the last four years. So what's been that disconnect? What will change in the future? And just kind of related to that, do you think the competitive and pricing environment will be more or less favorable in that business for the next few years?
spk10: So it's a good question. The first thing that I'll point out is that the group has consistently underperformed our profit targets. It hasn't consistently lost money. So this is the, you know, I think this is really the first year where we're projecting a loss for that group. But you're right, you know, we are in robotics for growth and we're not delivering the growth that we expected to from this business. And what we're really discovering is that there's a, If I'm going to sort of sum it up, we misunderstood how large the potential end market is and how much the early years of those business was driven by very sophisticated early adopters, people that were, you know, like robot enthusiasts. And what we found is that as we satisfied those early enthusiasts and started moving into a larger market where people were more focused on just buying a solution, that they didn't have the skills that they needed to put these robots into operation. So that's the logic behind our transformation in the channel at UR, that By shifting more emphasis towards solution providers like OEMs, we're going to be able to provide solutions that are at the level of the capability of our customers. And by directly connecting to larger customers, those are customers that have the heft to be able to maintain and implement the robots within their organizations. So I think that the thing that we're discovering is that the challenge in robotics isn't as much around technology as it is a go-to-market and a product-market-fit challenge. And we have a lot of evidence that our products do deliver value. We just are working out the best way to get them into the hands of our customers in a way that they're going to be able to easily adopt them.
spk09: Competitive environments.
spk10: Oh, yeah. So thanks for the cue, Andy. So in terms of the competitive environment, the UR is a clear leader in its space. It has more than three times the market share of our nearest competitor. And, you know, our share is in the mid 30s. We are we see two kinds of competitors, industrial robot makers that are pivoting into co-bots and and sort of pure play cobot companies, mostly from China. We have not lost any significant share to the industrial robot players that have entered the cobot market in the past couple of years. But we have lost a few points of share to the Chinese pure play entrants. And almost all of that share loss is specifically in China. So I said that our share is like mid 30s worldwide. If you exclude China, our share worldwide is over 50% or around 50%. Um, and it's much, much lower in China. And to be really Frank, we are, we are struggling with coming up with the best solution to gain share in China, because a lot of that market is being served at prices that aren't profitable for us. And we're not really interested in growth at the expense of overall profitability. In AMRs, it's the wild, wild west. We have about the total AMR market is about $2 billion per year. And our share is probably in the range of about 3%. And we're in a crowd of a whole bunch of other companies that are in that 1% to 6% range. The thing there is that it's very fragmented. There's no clear leader. we are really working on trying to establish strategic relationships with global customers because they're the ones that have the ability to adopt large fleets. So I think, you know, we see competition, but we believe that it's fragmented and we'll be able to outperform.
spk03: Thank you. Our next question is from CJ Muse with Evercore ISI. Please proceed with your question.
spk08: Yeah, good morning. Thank you for taking the question. I guess first question was hoping you could speak in more detail around the 35 million shortage. Is that specific to auto? And I guess what would enable those revenues to come in earlier in Q3?
spk01: Sure. Hi, it's Sanjay. Thanks for the question. You know, I'd say it's spread across the entire semi-test portfolio, across many of the product lines. And the supply shortages are, as I've noted in the past, really tied to analog and linear logic. A couple of other items, but those are the ones that as, you know, we're working with our supply chain partners as supply comes online to enable that. And so, you know, right now it's excluded from the range and we're trying to be transparent just as we have over the last couple of years. And, you know, I just want to be clear that if that does come out of Q4 and is provided into Q3, it shouldn't have an overall impact on the second half of the year.
spk08: Makes sense. I guess for mobility for 24, you gave some great detail earlier, but I was hoping perhaps you could kind of rank order three nanometer complexity, unit growth, potential for edge AI demand, and any other kind of drivers we should be thinking about to support growth for that business next year?
spk10: Specifically in mobile?
spk08: Yes.
spk10: Yeah, so in mobile, I would say that we bake edge AI into the complexity factor. So the edge AI is the market pull that is causing the chipset, the mobile processor developers to go to three nanometer and to increase their transistor counts a lot. So I lumped that in. I think that's probably the key driver for complexity in mobile is increasing the capability for AI in the handset. Having said that, the pace at which complexity increases is kind of linear. It hasn't inflected up. It's not inflecting down. It's been a pretty consistent factor. The thing that we see inflecting in 2024 is the unit volume. So when we look at this, we tend to look at peak quarters, peak shipment quarters. And the peak shipment quarter in 2020 was 20% higher than the peak shipment quarter in 2022 for smartphones. And that's a pretty big hole to try and dig out of. So next year, unit volume is gonna inflect and complexity has been increasing over the past, you know, it will have been over the past four years. So we expect to see unit volume being the driver.
spk03: Thank you. Thank you. Our next question is from Sameek Chatterjee with JPMorgan. Please proceed with your question.
spk00: Hi. Thanks for taking my questions. I guess if I can start on a somewhat related topic on AI. You mentioned the benefit or the demand you're seeing on the memory side, but maybe if you can share your thoughts about 2024, if you're looking at next year. where the biggest impact on the business will be from an AI demand perspective, memory, and any other pieces that you see that demand coming through and have a follow-up. Thank you.
spk10: So, you know, the cloud AI driving HBM and DDR5 is something that we expect to continue. It's still a... minority of the DRAM shipments. And so there's significant room for it to grow. And there's also less oversupply in those parts than there is in the general DRAM market. The application of AI in mobile is going to drive the adoption of next generation flash protocols. And that will also drive performance tester sales into that segment. And then when you look in the SOC space, there's significant business, of course, for traditional compute suppliers. But it is the primary driver for vertically integrated producers to be developing their chips. And so We're looking for growth in those VIPs in 2024, especially for cloud AI, hyperscaler type stuff, but also in terms of AI capabilities and vehicles. We expect that that's going to be a driver of business from those nontraditional compute players.
spk00: Okay. And a follow-up, just going back to Vivek's question on robotics profitability, I mean, we can see that based on what you've done in that segment, like, looks like breakeven point is more sort of around 360 to 400 million annually in terms of revenue. Any thoughts in terms of what level of scale you need to get to your internal sort of targets in relation to profitability in that segment? And as you think about the sort of next few quarters, what does the turnaround in revenue look like? How much longer before the distribution channel hiccups are out of the way and then you're more dependent on the macro? Any views on that as well? Thank you.
spk10: Sure. So let me take that backwards. So in terms of how long does it take, we expect that we're going to see significant improvement in Q4 of this year related both to the release of the UR20 and also that will have given us kind of 12 months of direct account coverage in some large accounts, which is on par with what our normal sales cycle is. So we're hoping to see some improvement in Q4 results over where we are now. In terms of the scale for profitability, I think it's probably best to talk about that in terms of where we are now and where we're heading to. So our target going into 2023 was that this group was gonna yield five to 15% profit. And we've capitulated against that because we're not achieving the growth. That is our baseline in terms of our planning for next year is to return to that range. And by the time we are at the end of this midterm in the 2026 timeframe, we expect that profitability range to eke up a couple of points from there. But the answer to when the profits that we make from robotics is in the same range as our sort of mature semi-test business, that really depends on the growth rate that we're seeing as we exit this midterm. So if we're on track and that business is growing 20 to 30 percent per year, then we're going to continue to drive it in that 10 to 20 percent profit range because we see a tremendous, a larger future value by doing that. If we're seeing that we've misunderstood the scale of this $500 billion potential end market, then we will be changing gears and trying to reduce our OPEX in those groups to try and get into the same range as the rest of the company. The rule of thumb that we're planning to is we want that group to operate at rule of 40. And I would be delighted to have it delivering 30% growth and 10% profit exiting the midterm because that means that we'd be on track for having that unit at multiple billions of dollars over the long term.
spk03: Great. Thank you. Thanks for taking the question. Thank you. Our next question is from Toshi Ahari with Goldman Sachs. Please proceed with your question.
spk07: Hi, good morning. Thank you for taking the question. Greg, a question on the auto and industrial semiconductor test businesses. I guess the level set I was hoping you could share, you know, roughly what percentage of SOC test revenue in calendar 23 will likely come from auto and industrial? And I guess more importantly, you know, how are you thinking about sustainability into 24? I think this has been one of the most extended upturns, you know, for your customers and for you guys as well. I think collectively we appreciate the long-term secular drivers and you spoke to some of them in your script, but how are you thinking about the cycle into 24?
spk10: Yeah, so why don't I, I'll take the second half of your question and I'll pass it off to Sanjay for the sort of exact splits in terms of this year. In terms of sustainability, the key thing that we're looking at is the increasing attach rate of semiconductors in cars. So the rate of EVs as a percentage of the whole fleet is about a 20% CAGR. The portion of EVs is growing as a portion of the total car market And there's twice as much semiconductors in an EV as there is in a standard internal combustion car. When you roll all that together, what we're seeing is that the dollar value of semiconductors in each car shipped is increasing at about a 12% rate. And that's a lot faster than the total semiconductor revenue CAGR. So there's this underlying driver in terms of an increasing attach rate That's a fair amount larger than the variation in automotive demand. So the automotive demand is going to continue to cycle, and there certainly could be periods of overbuying and periods of absorption coming in the future. But the fact that the utilization is as high as it is right now, and the end market fundamentals are this strong, We don't see that kind of softness. Coming into this year, we weren't as confident that it was going to sustain. Now, I do separate industrial out from automotive, and industrial is, it doesn't show the same kind of fundamental strength. There are some parts of the industrial market that are booming around clean energy and in terms of manufacturing infrastructure for EVs and batteries and stuff like that. That is certainly consuming semiconductors, and it's helped driving significant growth of our business testing high-power discrete devices like GaN and silicon carbide. So we see strength in industrial. It's not as pervasive as the strength we see in automotive. Does that answer that part of your question?
spk07: That's super helpful. And then, Sanjay, any color on how – yeah, thank you.
spk01: Mike Pratt- Sure. From a, let me take it from a market and then bring it down to revenue for context. So, we view in our, you know, in our view of the market size, we view that auto and industrial as about 25 percent of the SOC TAM this year. And relative to our revenue, we believe that of the SOC revenue that we have, it's about 35 percent of our revenue, or just under roughly 20 percent of our revenue overall. That's how we see the year unfolding.
spk07: That's very helpful. Thank you. And then as my follow-up, I'm curious, Greg, how you're thinking about the memory test TAM into 24. You know, I think the market's been very consistent over the past several years, somewhere in the $900 million to a billion dollar range this year is, I guess, no exception. But given what you're seeing in HBM, given what we hope to be a recovery in 24 across DRM and NAND, is there a possibility the market kind of breaks out and it's closer to 1.5 billion? Is that realistic? And I guess importantly, how should we be thinking about your market share? You spoke about your presence in HBM being higher than your average in DRAM, but if you can provide a little bit more context there, that would be helpful. Thank you.
spk10: The memory market is famously volatile, and it is remarkable that the TAMs have been as stable as they've been over the past couple of years. It's too early in the year right now for us to really peg a dollar value or even a range for next year. I am optimistic that there are more up forces than down forces against the memory TAM next year. So you have sort of a consistent pace of technology adoption that we think is going to continue. We think that HBM is going to continue to grow as a portion of the memory business. We think that transitions to new protocols are going to happen. And from what I've seen in the analysis of the memory makers, it appears that they're working through their inventory and that they may be in a position to start adding more capacity in terms of the mainstream of their business. So I think it'll be up, but I'm not going to agree with the number that you put out there.
spk03: Thank you. Our next question is from Brian Chin with Steeple. Please proceed with your question.
spk12: Hi there. Good morning. Thanks for letting us ask a question. Maybe actually just to re-clarify something, and this is sort of the math around industrial automation. I think to achieve the low end of your revised flat-to-down 10% forecast, I might have to input a very strong sequential growth in 4Q, even beyond kind of normal seasonal. Is that correct? And if so, what are you factoring in? And does that include like RevRec on that backlog of UR20s that you have or something else?
spk10: Yeah, so I think the way I would model it is normal seasonal Q4 plus the additional shipments of UR20. So we're expecting – our Q4 is always our strongest quarter of the year, and we have – we've been taking orders for the UR20 for nearly a year at this point. and we have a significant backlog that will support, you know, sort of a new layer, or basically it represents additional SAM that we're serving that we were unable to serve before.
spk12: Got it. And is that just sort of the release point is 4Q, why that's sort of backlogged up to the end of the year?
spk10: No, so we – We began shipments of the robot at the end of the second quarter. We will recognize that revenue plus continue shipments in the third quarter. But the volume per week is increasing as we go through Q4. So the bulk of the revenue is going to be in Q4. Okay.
spk12: Got it. Thanks, Greg. I'm just curious, you know, on the other side of sort of this inventory correction, I'm just curious, how are you thinking about sort of native demand? A lot of talk about complexity here of tests and understandably so. When you think about native demand for smartphones, on the other side of this unit correction, do you kind of, even though it's been a big driver of the test CAGR historically, you know, in relation to auto, in relation to high performance compute, maybe parts of the memory TAM as well, You think SOC mobile potentially is a bit of a drag on the test kicker over the midterm?
spk10: So, you know, it's interesting. We had a conversation about this yesterday. And we expect, you know, a pretty robust recovery in mobile over time. But at this, like, while this has been going on, the amount of semiconductors in cars and the complexity of semiconductors in cars is is really accelerating. And the investments around cloud computing, and especially AI-accelerated cloud computing, are proceeding at a pretty good clip. So when you look back in time, a large portion of the whole SOC TAM was mobile. And there were years for us where it was over 50% of our SOC sales went into that space. What we see looking forward is, you know, we think the whole market's going to grow at kind of, you know, 8% to 13% CAGR. That mobile will come back, but the rest of the market will have grown so that it's less concentrated in that area. So it's not like mobile's dead, but we think that there's a greater level of diversity across segments looking forward.
spk12: Okay. That makes sense. Thanks, Greg.
spk03: Thank you. Our next question is from Sydney Ho with Deutsche Bank. Please proceed with your question.
spk06: Great, thanks. I just want to follow up with Brian's question. If you look at Q4 guidance of being flat quarter over quarter, you talk about the robotics business up quite a bit in Q4, but what gets you to flat revenue in terms of what other businesses are down taking any particular segment you would call out that may be below normal seasonality?
spk01: Well, hi, it's Sandra here. Thanks for the question. You know, I think it's just the timing, uh, within our semi test portfolio that you're seeing, um, kind of the ebbs and flows. It's not uncommon for Q4 and some of our test businesses to have a, a lower Q4. Nothing more than that.
spk06: Okay. That's fair. Um, if I switch over to talk about the compute SOC, uh, TAM, uh, sounds like that is definitely stronger. You raised it, the overall TAM. but maybe within a compute it's now 1.3 billion instead of a billion that you talked about a quarter ago. Within that number, how should we think about the vertically integrated producers today and how that could look like next year? And maybe what kind of market share are you expecting within the VIP group that you mentioned?
spk10: So one of the things that we have highlighted about VIPs is because there are a relatively small number of players and the parts that they're producing are quite complex that it's going to be spotty, you know, that there will be, you know, capacity buys that happen for particular devices and they will, you know, it's all one customer will pop and then a different customer will pop. So we think that year on year from 2022 to 2023, The VIP TAM is going to be down a bit. Our share, you know, like last year, we had more than $100 million of revenue in VIPs. This year, it's probably between 50 and 100. So I don't know exactly what it will be in 2024, but by the time we get out to 2026, We think that that TAM is probably going to be somewhere north of $500 million. So it's going to be a significant chunk of the overall SOC TAM by the time we exit this midterm.
spk03: Thank you. Our next question is from Chris Sankar with TD Callen. Please proceed with your question.
spk04: Yeah, hi. Thanks for taking my question. I have two of them. Greg, I'm just curious on the DDR5 and high bandwidth memory test. Is it skewed more towards wafer test or final test? And what is your market share in either of those? And then I'll follow up.
spk10: So the fun thing about HBM is that it doesn't go into a package. So it's essentially all wafer test. There are two phases of that wafer test. So there's a a test of the memory itself, and then there is an interface performance test, which is done at the wafer level. And so once all of that test takes place and the HBM has been built up, then it gets shipped off to be incorporated into a, you know, a chiplet-based design for the final product. The TAM that we're talking about for HBM test is probably roughly split about half and half between the core testing and the performance testing.
spk04: Got it, got it. Very helpful. And then just to follow up on the VIP, you know, you said it's going to be 50 to 100 million this year compared to almost over 100 million last year. Is that basically skewed towards one customer or one auto customer for you, or is it more diversified?
spk10: No, it's much, much, much more diverse. So last year, like what you could probably think about is that the broad diversity of customers from last year to this year is actually increasing. But last year, on top of that broad base, a single customer did pop for a lot of capacity. So I think that's – so I guess the answer to your question is yes and no. Got it. Thank you.
spk03: Thank you. Our next question is from Atif Malik with Citi. Please participate with your question.
spk02: Thank you for taking my question. Greg, I have a question on your robotics business. You spoke about two challenges. Teradine not able to build OEM channels fast enough and then macro environment. I'm just trying to understand how much of this is on you guys and how much of it is on macro? Is it more like 60% on you guys for not building the channel for these higher weight robots and 40% on macro. If you can help us out on that split.
spk10: So it's a tough split to make because if you compare us to peers in industrial automation, we have a significantly different lead time profile than most industrial automation players. So we're working with kind of five-week lead times. And for industrial robots and for standard automation, those lead times are out kind of two quarters. So we tend to compare what our business is doing to what their businesses are doing, looking at order rates. And so if you look at order rates for, say, UR, We're down roughly 10% year on year, so first half of 22 to the first half of 23. If you look at some of our industrial robotics peers, they're down 20% or greater over that same comparison period. So there's definitely a broad economic headwind that we're facing. But because we're coming into this space as a disruptor, we would expect to have significantly higher growth than the main players. And right now, we're not happy with the growth that we're seeing. We also think that we should have greater immunity to these cycles because we're solving a problem that these manufacturers have, whether it's good business or bad business. And so I think what you're seeing is our current distribution is quite vulnerable to macroeconomic cycles. And the distribution setup that we're moving towards is going to be more suited to deliver consistent results. So I would say that it's, honestly, I think it's mostly macroeconomic, that the macroeconomic headwinds are particularly bad given the types of customers that we serve and the way that we serve them. The build out of OEM and large account coverage is something that's going to give us better immunity in future headwinds.
spk02: Thank you. That's all I had.
spk09: Okay. Operator, we are out of time. Everyone, thank you for joining us today. Those in the queue, I'll follow up shortly after this call. We look forward to talking with you in the days and weeks ahead. Bye-bye.
Disclaimer