Arteris, Inc. (AIP) Earnings Call Transcript & Summary

Good afternoon, everyone, and welcome to the Arteris Fourth Quarter and Full Year 2024 Earnings Call. Please note, this call is being recorded and simultaneously webcast. All material contained in the webcast is sole property and copyright of Arteris, Inc. with all rights reserved. For opening remarks and introductions, I will now turn the call over to Erica Mannion of Sapphire Investor Relations. Please go ahead.

Thank you, and good afternoon. With me today from Arteris are Charlie Janac, Chief Executive Officer; and Nick Hawkins, Chief Financial Officer. Charlie will begin with a brief review of the business results for the fourth quarter and year ended December 31, 2024. Nick will review the financial results for the fourth quarter followed by the company's outlook for the first quarter and full year of 2025. We will then open the call for questions. Before we begin, I'd like to remind you that management will make statements during this call that are forward-looking statements within the meaning of federal securities laws. These statements involve material risks and uncertainties that could cause actual results or events to materially differ from those anticipated, and you should not place undue reliance on forward-looking statements. Additional information regarding these risks, uncertainties and factors that could cause results to differ appear in the press release Arteris issued today and in the documents and reports filed by Arteris from time to time with the Securities and Exchange Commission. Please note, during this call, we will cite certain non-GAAP measures, including non-GAAP net loss, non-GAAP net loss per share and free cash flow, which are not measures prepared in accordance with U.S. GAAP. The non-GAAP measures are presented as we believe that they provide investors with the means of evaluating and understanding how the company's management evaluates the company's operating performance. These non-GAAP measures should not be considered in isolation from, as substitutes for, or superior to financial measures prepared in accordance with U.S. GAAP. A reconciliation of these non-GAAP measures to the nearest GAAP measure can be found in the press release for the quarter ended December 31, 2024. In addition, for a definition of certain of the key performance indicators used in this presentation such as annual contract value, confirmed design starts, active customers and remaining performance obligations, please see the press release for the quarter ended December 31, 2024. Listeners who do not have a copy of the press release for the quarter ended December 31, 2024, may obtain a copy by visiting the Investor Relations section of the company's website. In addition, management will be referring to the fourth quarter 2024 earnings presentation which can be found in the Investor Relations section of the company's website under the Events and Presentations tab. Now I will turn the call over to Charlie.

Thank you, Erica, and thanks to everyone for joining us on our call today. In the fourth quarter of 2024, we achieved a record annual contract value, plus royalties, of $65.1 million as demand for commercial semiconductor system IP products continues to grow. Our success during the quarter was fueled by increased adoption of AI-driven enterprise computing and automotive SoCs. We also continue to generate growing momentum in other key verticals including microcontrollers or MCUs. Business in the fourth quarter was driven by a mix of the addition of new customers, including several market leaders, as well as increased penetration in our current customer base, demonstrating the success of our land and expand strategic approach. For example, the largest win in the quarter came from a global top 5 technology company that expanded its use of the Arteris product portfolio, complementing previous NoC IP orders with the addition of Magillem and CSRCompiler SoC integration automation software for their high-end AI SoCs for enterprise computing applications. Also, a major automotive OEM and a top 5 automotive semiconductor company expanded their use of Arteris products for several additional SoCs given the combination of superior performance, power and area efficiency as well as functional safety for their mission-critical applications. Last quarter, we shared that we are strategically expanding into the microcontroller, or MCU space, where designs have grown in complexity in recent years to benefit from low latency, flexible power and area efficient commercial NoC IPs. We are pleased to report the strategic expansion has already started to bear fruit with Infineon, the leading microcontroller manufacturer becoming a new customer standardizing on Arteris NoCs from automotive MCUs, which serves many of the world's top automotive tier 1 vendors and OEMs. We believe this strategic MCU win will help to accelerate our growing royalty stream. Another key customer win was GigaDevice, where Arteris was selected by the microcontroller business unit as a result of our optimization in interconnect area and power consumption while ensuring functional safety. We're also seeing increased adoption of Arteris technology for chiplets, particularly for high-performance enterprise computing applications, sophisticated autonomous driving and smart edge devices across market-leading companies, midsized players and innovative startups. These customers are increasingly pursuing a multi-die strategy to expand compute power with Arteris as the core interconnect IP for each chiplet due to our technologies, superior power, performance and area, or PPA. One such example was Tenstorrent, which expanded the deployment of Arteris NoCs for their next generation of chiplet-based AI solutions for high-performance, energy-efficient RISC-V computing for AI and HPC data centers. Similarly, Menta deployed Arteris for their edge IP chiplet to ensure better performance and area efficiency for edge AI and IoT computing. As we look back on 2024, we witnessed accelerating industry demand Arteris technology, which we believe was fueled by increased penetration of AI into not only high-end data centers and autonomous driving but a wide range of new products, including edge devices. Complexity has, and we believe, will continue to impact high-end compute and traditional lower-end technologies, including MCUs, driving demand for efficiency that is enabled by Arteris Network-on-Chip technology. This has resulted in additional 14 new customers and increased wallet share of Arteris products in customers ranging from top 5 technology companies down to new innovative start-ups, with our technology now being part of nearly 850 designs to date. Last year, we also saw increased adoption of the physically aware FlexNoC 5, which leverages advanced node and placement information to enable up to 5x faster physical coverage while supporting best-in-class PPA. We are happy to highlight that in the fourth quarter, over 75% of FlexNoC interconnect IP customers chose this more advanced version, which was introduced just 1.5 years ago. Also noteworthy, last year's addition of tiling and expanding mesh technology in FlexNoC and Ncore product lines, along with Armv9 support helped to advance Arteris as the right partner to support the most innovative chip designs. Moreover, I'm very excited to announce today our FlexGen smart NoC IP, which has the potential of -- to revolutionize semiconductor designs by delivering up to 10x engineering productivity and lowering power consumption and improving overall PPA. FlexGen builds upon the silicon proven and physically aware FlexNoC 5 IP to automate the creation of high-performance Network-on-Chip, NoC designs. Supported by AI-driven automation, FlexGen reduces manual iteration by over 90%, providing expert level NoC topologies in hours or days instead of weeks as demonstrated by Dream Chip under ADAS SoC as well as multiple other designs. FlexGen is now ready for production deployments and has been delivered for evaluations to over 10 companies, some of which have been working with this technology for more than 6 months. FlexGen is the combination of years of groundbreaking innovation and multiple patents with the goal of boosting productivity while improving quality of results to overcome extreme design challenges semiconductor and system companies face when creating today's chips or chiplets, which often contain 5 to 20 NoCs each. We expect FlexGen to have a positive impact on our customers and on our business going forward. Lastly, our long-standing position as a neutral IT provider was illustrated in our continued success with Arm-based designs, customers using RISC-V and x86 CPU IP architectures. To further support this expanding processor IP ecosystem, last quarter we announced a partnership with MIPS to provide a pre-verified RISC-V reference platform to support mutual customers. The goal is to improve interoperability and shortened SoC integration for chip designs for automotive, enterprise computing and edge AI applications using Arteris as their essential connectivity backbone. We believe that the scale and scope of our long-term opportunity remains robust and is supported by our current products and strong product pipeline of new system IP technologies as well as growing relationships with some of the largest and most advanced electronics companies in the world. Our customers continue to innovate in exciting high-growth areas such as generative AI, autonomous driving, using Arteris technologies and global support. With that, I'll turn it over to Nick to discuss our financial results in more detail.

Thank you, Charlie, and good afternoon, everyone. As I review our fourth quarter results today, please note, I'll be referring to GAAP as well as non-GAAP metrics. A reconciliation of GAAP to non-GAAP financials is included in today's earnings release, which is available on our website. Also, as a reminder, I will be referring to 4Q 2024 earnings presentation, which can be found in the Investor Relations section of the company's website under the Events and Presentations tab. Turning to Slide 5 of the presentation. Total revenue for the fourth quarter was $15.5 million, up 24% year-over-year, and above the midpoint of our guidance range. At the end of the fourth quarter, annual contract value, or ACV plus royalties, was $65.1 million, slightly above the midpoint of our guidance range, and a record high for the company. Remaining performance obligations, or RPO, at the end of the fourth quarter were $88.4 million, representing 22% year-over-year increase and growing to the highest level we have ever reported. Non-GAAP gross profit for the quarter was $14.2 million, representing a gross margin of 91%. GAAP gross profit for the quarter was $13.9 million, representing a gross margin of 90%. For the full year, non-GAAP gross profit was $52.7 million, representing a gross margin of 91%. GAAP gross profit was $51.8 million, representing gross margin of 90%. Now moving to Slide 6. Non-GAAP operating expense in the quarter was $16.9 million, flat sequentially and only 1% higher year-over-year. This reflects the team's continued focus on prudent management of our operating expense. Total GAAP operating expense in the fourth quarter was $21 million, representing a 4% year-over-year increase. For the full year, non-GAAP operating expense was $67.6 million, a decline of 2% from prior year. Total GAAP operating expense of $83.4 million, a slight decline from prior year. As we look ahead, we plan to continue to limit spending to strategically critical areas while investing in profitable revenue growth. Non-GAAP operating loss in the quarter was $2.8 million, which came in above the top end of our guidance range. This represents a $2.7 million improvement compared to a loss of $5.5 million in the prior year period, and a $0.6 million improvement sequentially. GAAP operating loss for the fourth quarter was $7.1 million compared to a loss of $9.2 million in the prior year period, and $7.9 million in the third quarter. For the full fiscal year, non-GAAP operating loss was $14.8 million, representing a $5 million improvement compared to the prior year. GAAP operating loss for the fourth quarter (sic) [ full year ] was $31.6 million, representing an improvement of $3.5 million from the prior year. Non-GAAP net loss in the quarter was $3.9 million, or diluted net loss per share of $0.10, based on approximately 40.2 million weighted average diluted shares outstanding. GAAP net loss for the quarter was $8.2 million or diluted net loss per share of $0.20. For the full fiscal year, non-GAAP net loss was $16.9 million or diluted net loss per share of $0.43, based on approximately 38.9 million weighted average diluted shares outstanding. GAAP net loss for the year was $33.6 million or diluted net loss per share of $0.86. Moving to Slide 7 and turning to the balance sheet and cash flow. We ended the quarter with $52.3 million in cash, cash equivalents and investments, and we have no financial debt. Free cash flow, which includes capital expenditure, was negative $2.7 million in the fourth quarter and negative $1.0 million for the full year. This was below our guidance due to short-term working capital timing changes at the end of the year with some customer payments that were forecasted for the fourth quarter being received shortly after the fourth quarter close. This along with strong order growth resulted in an increase in our accounts receivable balance of $11.9 million from the prior quarter end. I would now like to turn to our outlook for the first quarter and the full year and refer now to Slide 8. For the first quarter of 2025, we expect ACV plus royalties of $65.5 million to $67.5 million; revenue of $15.7 million to $16.1 million with non-GAAP operating loss of $4 million to $3 million; non-GAAP free cash flow of negative $2 million to positive $2 million. For full year 2025, our guidance is as follows: ACV plus royalties to exit 2025 at $73 million to $77 million; revenue of $66.0 million to $70 million; non-GAAP operating loss of between $12.5 million to $8.5 million; and non-GAAP free cash flow of positive $1 million to positive $7 million. We are encouraged by the strong deal flow exiting the year and our effective cost management that resulted in better-than-expected performance in non-GAAP operating income in 2024. And this positions us for further improvements in our key financial metrics in 2025. With that, I will turn the call over to the operator and open it up for questions. Operator?

[Operator Instructions] Your first question comes from the line of Gus Richard from Northland.

I guess, Nick, first for you, by my calculation, your bookings were over $30 million in the quarter, and that would kind of help explain the increase in deferred revenue and balance in accounts receivable. Am I in the right zip code?

Gus, welcome to the call. Nice to speak to you again. So we don't actually specifically talk about bookings, it's not one of our metrics that we publish. But we have characterized the fourth quarter as a strong deal flow quarter. Fourth quarter is always our strongest deal flow quarter of the year, and this was no exception.

Okay. And Charlie, for you, just -- could you talk a little bit more about FlexGen and a 10x increase in productivity is a lot. Is that place and route? Or is it just doing the NoCs? And just can you talk a little bit more about how the product works and how it's different from prior versions? You still there?

I wonder whether Charlie might be on mute. Let me just...

Sorry I apologize, I was on mute, yes. So FlexGen is based on FlexNoC 5, so you have all of the manual editing capability that you had before. But what FlexGen does, is you basically feed it a connectivity map, a list of all the connections, locations of all the IP block exit ports and a floor plan, and it gives you basically a NoC, a NoC topology in minutes -- or hours to minutes versus days. And this has been validated by a large number of benchmarks. So the productivity increase is huge, but it's not enough. So we've also been able to achieve superior wire length. So sometimes, depending on the benchmark, it can be from relatively modest for something like a small microcontroller, all the way to 30% for -- up to 30% for very complex large SoCs. And that gives you improvements in latency and gives you improvements in power. So we think that the 3 years of work that we have put into this is generating some very good results. And as we had in the script, we've basically shipped this to about 10 companies. And I think 3 more in the last 1.5 months as well. So I think we're up to 13 or so. And the take-up of the product has been very strong.

And then last one for me. If I recall, the uplift on this product is about a 30% increase in ASP. Is that right?

That's right.

Your next question comes from the line of Kevin Garrigan from Rosenblatt Securities.

Congrats on the solid results and solid 2024. Charlie, just kind of going off of Gus's last question. Can you just kind of talk about ASP trends per project? I think previously, you were looking to hit kind of $1 million in 2026. So where did they kind of finish out for 2024? And is that $1 million ASP still kind of on track?

Yes. It is. But you asked a good question on the third quarter earnings call is we've announced that we're going to enter the microcontroller business. And we've been able to demonstrate success with 2 customers, Infineon and Giga, and you asked a good question is what's the ASP on the microcontrollers. And the answer is that while we're going after entire product lines in microcontrollers, on an individual basis, the ASP there is going to be lower because those do not really need FlexGen. They may or may not need physical awareness and the interconnects, while getting much more complex, a little bit simpler. But for the complex SoCs, we're definitely on track for $1 million ASPs on the average. And we know this because we're getting $1 million per project deals now, it's just not necessarily the average. And if you were to buy everything from us right now, you'll be looking at kind of a reasonable industry discount, about $1.5 million. So -- if you were to buy everything from us. So as we deliver these new products and as the amount of system IP that's being used increases per project, we're on track for that $1.6 million -- I'm sorry, $1 million ASP. But the caveat is that you have to exclude the microcontrollers.

Okay. Got it. That makes sense. So that actually leads me into my next question. And you're looking at complex designs in the MCU market, but maybe not as complex as those in enterprise computing or AI machine learning. So now that you kind of have Infineon and Giga under the belt, is -- the time between design starts production, is it similar to your typical average? Or is it kind of maybe accelerated because they're not -- they may not need as much?

So our target is to essentially establish relationships with the large microcontroller vendors. And they typically build generations of microcontrollers every 3 years. So from that perspective, the design start windows are longer. However, once they start designing a generation of microcontrollers, they may design 6 to 10, maybe even more, maybe up to 15 microcontrollers per generation. And so the time between those design starts is actually very, very short, right? So what you have to do is you have to hit the time where the customer is doing a new generation of microcontrollers and then the design start window gets relatively short of that particular generation of microcontrollers. It's a pretty different dynamic than you see in the automotive or AI SoC space.

Congrats on the results.

[Operator Instructions] Your next question comes from the line of Ethan Potasnick from TD Cowen.

Congrats on the great results. I wanted to kind of dig into licensing and royalty results. Kind of considering the broader macro backdrop and particularly what's occurring in the automotive space. So maybe to get some expanded thoughts there would be helpful.

Nick, this is for you -- yes, you take this one. You're the royalty king.

Yes. So yes, so the astute observers among you will have spotted that the royalties and other line was slightly lower year-over-year. There's 2 real causes for that. One is that, as you'll recall from previous calls, we had some onetime benefits, particularly around royalty audits, fairly substantial in 2023, and those were later in 2024. So we tend to look at variable royalties as a better sort of trend guide. And variable royalties, which is -- that's still the bulk of that total income line, were 20% up year-over-year. Remember also that Mobileye had a sort of fairly major inventory correction particularly at the beginning of 2024, as I'm sure you'll remember. And so if you exclude Mobileye from that trend, which is quite a big royalty contributor, the overall variable royalty growth year-over-year was in excess of 30%. So we are seeing some -- if you look at the concentration of royalties, in 2024 versus 2023, the proportion of total variable royalties that came from automotive was, of course, a little lower than it was in previous years because of the Mobileye impact. But it's still around half of the total.

Okay. Got it. Got it. Understood. And then to kind of piggyback on a prior question about pricing and the incorporation, I guess, of new products and the foray into MCUs, with FlexNoC 5 carrying a higher ASP, which is great to see. But how should we think about kind of the profitability trajectory for this year and perhaps the free cash flow positive target this year kind of exiting 2025? How are things going there?

So let me take that one real quick. So the -- just first on free cash flow. We're right at that point where we're flipping from negative to positive. So we mentioned in the call that we had a small number of customers who just missed the cutoff at the end of the year. Relatively minor impact, but it turned the plus sign into a slightly minus sign, but very small either side. The reason we're so confident about the full year is that we're growing the top line, as you know, in the high teens, low 20s percent, and that's basically our cash inflow, and we're constraining the OpEx and the cost of revenue growth, which is essentially our cost base, at half of that. So we naturally grow cash flow just automatically by our control of that metric, the OpEx being 50% of the top line growth and the growth at the top line is growing at high teens, low 20s percent. Now with -- one additional piece of color for you on that, Ethan, is that the free cash flow is always weighted towards the second half or generally speaking, last quarter -- the fourth quarter was a bit of an anomaly because of this small working capital shift. But generally speaking, we received most of our cash inflow in the second half. And we have some reasonably substantial sort of nonlinear cash outflows, particularly around management bonuses, for example, and around the high commissions and the annual accelerators all happened in the first quarter. So there's a little bit of seasonality around it, which basically pushes most of the cash generation into the second half, and that's traditionally what we've seen year-over-year. Does that give you a good feel?

Yes.

Your last question comes from the line of Gus Richard from Northland.

Great results, guys. Just, Charlie, you mentioned your architecture-agnostic RISC, Arm, x86. And I'm kind of surprised to hear you list x86. Can you talk about the reasons you might be involved with that particular architecture?

PC chipsets.

I'm sorry, say again?

PC chipsets. So we are in some PC chipset designs.

Got it. And then -- you also mentioned multiple NoCs per design. Do all of the fabric or the network-on-the-chip have to be from the same vendor? Or can you mix and match NoCs? For example, some of the x86 guys have their own fabrics.

Absolutely. That's a great question. So our products are designed for mix and match. I would say there are some designs where Arteris is 100% of all the interconnects. But I would say a majority of them have some other types of interconnect in them either because of legacy reasons or other reasons, right? So for example, one common -- reasonably common configuration would be the Arm, CMN cache coherent interconnect with FlexNoC, the non-coherent one. So that would be fairly common. Another one would be an internal fabric with Ncore or FlexNoC. So we are designed for mix and match, and this is where frequently -- this is frequently what occurs. But of course, over time, we would like Arteris to be more and more 100% of all the interconnects, but we're designed for mix and match.

Got it. And then the last one for me. Again, you mentioned a number of chiplet designs. And I'm just wondering if you could talk a little bit about -- as people chop up chips and use chiplets, how does your opportunity change? And how does that change the competitive landscape for you all?

So it makes the interconnect a lot more complex, right? Because now you no longer -- you're not just in a die, but you're sending data back and forth between different pieces of silicon. So that raises the ASP. We're also finding out that in these chiplet projects, multiple -- there are multiple companies involved, so that involves multiple licenses. And of course, we treat each die as a separate project. So when there were questions about the $1 million ASP, I think Kevin asked that the chiplets are going -- and I said the microcontrollers may lower the ASP, but the chiplet projects will increase it because there's multiple dies and even multiple companies involved in those projects. So that increases the revenue opportunity for Arteris significantly.

Got it. And then there's alphabet of standards connecting chiplets together. Does that have any -- do you ride on top of those? Or do they ride on top of you? How do those different protocols impact your opportunity? Does it make it more -- again, add to complexity or -- and again, sort of if you can help explain what layer you live in and what layers those live in and sort of how they play together?

So there's multiple layers, but the layer that we're involved with is a data transport layer. And so we basically interface to physical layer IPs, such as those made by Synopsys. And so we're kind of in the digital domain, and we ride on top of the physical layers, which are heavily analog. And as far as the different protocols, it's a trade-off, right. We tried to conform to protocols that are used by our major customers, but standards are beneficial to the industry. And so we're -- and one of the -- somebody said that the good thing about standard, there's so many to choose from. What we're trying to do working with our partners and the ecosystem to kind of create major standards that the industry can rally around and lower costs. And in the chiplet companies, that would be UCIe, it would be CHI over UCIe, those kinds of things. So it's kind of a trade-off. And the more standards you have, the higher your costs, right, because you have to do fairly expensive developments to conform to a large variety of standards.

There are no further questions at this time. I'd like to turn the call over to Charlie Janac for closing remarks. Sir, please go ahead.

Yes. Thank you for your time and interest in Arteris. I think it was quite a positive quarter and a year, and we look forward to meeting with you at the upcoming investor conferences we're participating in during the next couple of months. And we look forward to updating you on all our business progress in the quarters to come. Thank you for your support.

Ladies and gentlemen, this concludes today's conference call. Thank you very much for your participation. You may now disconnect.