Pdf Solutions Inc Q1 FY2022 Earnings Call
Pdf Solutions Inc (PDFS)
Call artefacts
Call audio is not captured yet.
A slide deck is not captured yet.
Transcript
Auto-generated speakersGood day. Thank you for being here. Welcome to PDF Solutions' First Quarter 2022 Conference Call. I will now turn the call over to Joseph Diaz of Lytham Partners. Please proceed.
Thank you, Christian, and thanks to all of you for joining us today on the call. I appreciate your time and your ongoing interest in PDF Solutions. As the operator indicated, my name is Joe Diaz I'm with Lytham Partners. We are the Investor Relations consulting firm for PDF. If you do not yet have a copy of today's press release, it's available on the company's website at pdf.com. Some of the statements made during this conference call will be forward-looking within the meaning of the Private Securities Litigation Reform Act of 1995, including statements regarding PDF's future financial results, performance, growth rates and demand for its solutions. PDF's actual results could differ materially. The forward-looking statements and risks referred to on this call are based on information available to PDF today. The company has no obligation to update them. You are advised to refer to the section titled Risk Factors on the company's annual report on Form 10-K for the fiscal year ended December 31, 2021, and similar disclosures in subsequent SEC filings. With that said, I'd like to introduce John Kibarian, PDF Solutions' President and Chief Executive Officer. He will be followed by Adnan Raza, Executive Vice President and Chief Financial Officer. At the conclusion of management's prepared remarks, we will open the call for your questions. Let me now turn the call over to John Kibarian, President and CEO of PDF Solutions. John?
Thank you, Joe. The first quarter was a great start to our year. Before Adnan discusses the financials in detail, I have some comments about the nature of our business in the quarter and our perceptions of the market. The first quarter bookings built a strong trend in 2021 with many of the same themes that we experienced last year continuing. For the industry in the first quarter, supply constraints, COVID lockdowns and sanctions as a result of the Ukraine war made for uneven availability of equipment, wafers and consumables. Customers expressed to us increased uncertainty in meeting shipment requirements and increasingly noticed our customers' executives spending effort to ensure supply. Despite this backdrop, and in some cases, because of these trends, we had a very strong bookings quarter. We did see some customers ship fewer tools than we had anticipated, which impacted the contribution from runtime licenses to analytics revenue in the quarter. At the same time, the contribution from gain share to integrated yield ramp revenue was improved modestly as a result of increased wafer shipments. We also experienced an uptick in Exensio process control perpetual licenses as customers build out new capacity. Overall, in the first quarter, the macro trends, while not all positive for our business, tended to break in our favor from a revenue perspective. Beyond these puts and takes that were a function of the macro challenges and opportunities, our business activity and results were very strong in the quarter. I will briefly touch on some highlights in bookings and partnership activities. First, building on the Quick Start contract that was signed with a leading-edge customer in the fourth quarter of last year and as anticipated, we signed a large follow-on contract. This makes it possible for this customer to use PDF characterization vehicle test chips, DFM software and Exensio Analytics to optimize its PD case and foundry interface for a broad range of chip designs. Second, and consistent with recent history, the largest Exensio contract in the quarter was for a cloud deployment. This contract was for one of our first cloud renewals as the customer was one of the early adopters of Exensio Cloud in 2019. The renewal grew ARR from this customer well over 50% as they took advantage of tiered storage and expanded usage. There were other Exensio Cloud contracts in the quarter as well as contracts for Exensio Process Control and test modules. These were at both new and existing customers. Third, we continue to experience strong customer adoption of our SDKs for Cimetrix's connectivity products. And while impacted by supply chain, still strong shipments of runtime licenses on an absolute basis. We ended the quarter with record runtime license backlog, which speaks to our customers' challenges in making equipment shipments, but also gives us confidence in the future contribution from runtime licenses. Finally, collaborating with other industry leaders continues to be an important part of our business. As we discussed last quarter, we announced our partnership with Siemens EDA to link Exensio with Tessent diagnostic products. Our first webinar in Q1 was highly attended, and the follow-up customer interest is high. In April, we announced a collaboration with Kulicke & Soffa to link Exensio with their assembly manufacturing equipment. This partnership builds on our existing OEM relationship, where K&S includes our Exensio analytics for assembly operations to enable traceability. Like our Siemens collaboration, this expanded engagement starts off with an early customer access program. Our relationships with Advantest, Siemens and K&S all speak to our customers' desire to add more analytics to the back-end assembly and test. This is particularly true for chip and system companies, employing system in package processes to implement 2.5D and 3D chip systems. Advantest's Voice User Conference is coming up next week, and we anticipate announcing new products as a result of our continued collaboration in conjunction with this conference. IBM, SiView's user group conferences also soon, and we'll be working with IBM to outline our collaboration for SiView customers. Overall, customer and partner activities in the first quarter were strong and consistent with our expectations entering the year. Now one quarter into 2022, the semiconductor environment remains robust and demand for integrated circuits is broad-based. We anticipate continued demand for our analytics platform, particularly on the cloud from a broad cross-section of customers. Given the industry tailwinds I just summarized, and in spite of the macroeconomic uncertainty, we remain confident in the outlook we provided earlier this year. Before I turn the call over to Adnan, I would like to thank all of the PDF employees and contractors for their efforts during the first quarter. We managed to work in tight concert, navigating many of the challenges to have the strongest revenue quarter in the history of the company. Now I'll turn the call over to Adnan who will review the financials and provide his perspective on the business. Adnan?
Thank you, John. Good afternoon, everyone. Good to speak with you again today, and I hope all of you and your families are doing well. We're pleased to review the financial results of the first quarter of 2022. As mentioned, our earnings release and unaudited report are posted in the Investor Relations section of our website. Our Form 10-Q was also filed with SEC today. Please note that all of the financial results we discuss in today's call are on a non-GAAP basis, and a reconciliation to GAAP financials is provided in the materials on our website. We are off to a strong start in 2022. Total revenues for the first quarter came in at $33.5 million, up 38% over last year's first quarter and up 12% versus the prior quarter of Q4 2021. Analytics revenue came in at $30.4 million, an increase of 57% year-over-year and 12% quarter-over-quarter. The increase versus the prior quarter was driven primarily by the start of a new leading-edge booking, which John spoke about, and increased Exensio software license sales. Analytics represented 91% of total Q1 revenues. For the quarter, Integrated Yield Ramp revenue was $3.1 million, a year-over-year decrease versus $4.8 million and a quarter-over-quarter increase versus $2.6 million. This quarter-over-quarter improvement was primarily due to higher wafer volumes driving gain share. We believe that our transition to a leading analytics provider to the global semiconductor supply chain is progressing well and is widely recognized within the industry. John also spoke about the progress of our ongoing and future plans for collaborations via engagements with Advantest, Siemens, Kulicke & Soffa and IBM, all important influential leaders in our industry. We're pleased that non-GAAP gross margin for the first quarter came in at 69% versus 61% year-over-year and 65% quarter-over-quarter. We improved our margins as we started to reap the benefits of scale in our cloud business, allowing us more efficient cloud spend and the ability to apply application engineering resources to presales and product management rather than customer support. We also benefited from increased perpetual software license sales during the quarter, both of which together contributed to getting closer to our 70% long-term gross margin target. Bookings for the quarter increased more than 90%, and our backlog at the end of the quarter grew to a healthy $196.8 million. Non-GAAP net income for the quarter totaled $3.7 million or $0.09 per share versus a non-GAAP net loss of $1.9 million or $0.05 per share loss in the year-ago period, year-over-year increase of $0.14 per share. Turning to the balance sheet. We have carefully managed our cash position and carry 0 debt. In Q1 '22, we purchased approximately $5.8 million worth of shares. After the conclusion of the quarter, we purchased an additional $16.7 million worth of shares in a privately negotiated transaction when a block of approximately 715,000 shares became available. The total number of shares purchased this year totaled 933,458 at an average price of $24.07 per share for a total year-to-date buybacks of $22.5 million. Our latest share count of 36.9 million shares as of May 6, 2022, as noted on the cover of our 10-Q, is now lower than the number of shares outstanding when we filed our Form 10-K for the year 2020. As we look to the next quarter and the rest of the year, we expect to increase costs in Q2 as we wrap investment again to continue delivering on the increased interest in our products and solutions. We expect Q2 total revenue to be similar to Q1 levels. And for the full year 2022, we expect total revenue growth to be between 20% to 25% on a year-over-year basis. All in all, it was a solid first quarter. We are pleased with the organic growth of our Analytics business and are making good progress to meet or exceed our target model gross margin of 70%. With that, let me turn the call over to the operator for Q&A. Operator?
Your first question comes from Christian Schwab from Craig-Hallum.
This is Tyler on behalf of Christian. So first, I guess I wanted to ask about the announcement of the partnership with Kulicke & Soffa during the quarter. That's multiple back-end customers we're partnering with now. I was wondering, is that what we should expect going forward, more back-end equipment partnerships? Or how do you feel about the possibility of partnerships more on the front end?
Yes, we see that we are engaged with customers on the front end as well. The back end has historically had very little analytics and simple processes. However, complexity has increased, and there is a strong desire for more automation and analytics. We are hearing from customers involved in both analog and sensor systems as well as complex computer chip systems. There are many opportunities and needs in this area. While it appears to be more balanced at the back end compared to the front end, we also have front-end customers making progress.
That sounds great. And then maybe a little bit on the model, the gross margin is obviously really strong in Q1. That's great to see. It sounded like I just want to confirm, I guess, that there wasn't anything onetime in nature in that? And would it be fair to expect that kind of 69% to go forward as you trend towards the 70% target?
Yes, I think that's fair. We've become increasingly confident in the way we've been able to manage our business, getting comfortable with the scale we're achieving and also what we're seeing in the outlook for the rest of the year. I think you can pick that up also in our comments that we expect the revenue to be 20% to 25%, which is a little bit of a more positive change compared to the last earnings release as well.
Your next question is from Tom Diffely from D.A. Davidson.
Let me ask a question or two questions. So I guess on the collaborations with Advantest, Kulicke, Siemens, IBM. John, maybe just talk a little bit about what your investment going into those collaborations is? What are your projected timelines before they drive your end market revenue and just kind of how you view those collectively?
Sure. Yes, I think we're gaining insights as we progress, Tom. It definitely takes at least a few quarters. Typically, we've been working on these projects for quite some time before announcing them. For instance, with K&S, we have been engaged in initial analyses and prototypes of Exensio modules for over a year to utilize their data effectively, capitalizing on our traceability and exploring what machine learning can identify in manufacturing. By the time you see them, there has usually been a substantial year of investment, sometimes even more. As I mentioned earlier, we aim to provide early access to our programs. While there may be some minor revenue linked to these, it's primarily to ensure a commitment from our customers for deployment. These programs will take a few months to quarters to begin generating revenue. For example, with Advantest, we started seeing revenue above the minimum level about nine months after the contract announcement, and we anticipate a similar timeframe, which might sometimes be around six months, but generally, within that range.
Okay. And are these contracts that these 4-plus companies are going to have with their customers and you'll get a percentage of that? Or how do you think the contracts will be ultimately structured?
Yes, that's a work in progress. Some of these are sell-through situations where our partners sell directly to the customer and manage the contracting. We've done this with Advantest, and there have been specific cases where IBM has taken the lead in sales. In those instances, the contract is between IBM and the customer for a bundle of Exensio and their software systems, allowing us to receive certain revenue amounts. There are also early access programs where each party sells their portion. As we observe the patterns emerging from these programs, we might decide to continue selling separately with connectors or leverage the channel. We are interested in utilizing our partners' channels. Given PDF's history, we have not developed a large sales channel, but we believe there is significant untapped interest in our platform. Therefore, we aim to utilize their channels whenever possible, which sometimes involves them maintaining the direct relationship with the customer.
Okay. Great. That's helpful. And then, Adnan, when you look at the cost of these programs, is there any dramatic change to your cost structure once they start, and once they start to produce revenue in the sense that are there certain costs that like nonrecurring engineering costs that get lumped into the later date?
No. I mean the perspective you take on all these engagements is, like John said, right, either they are incremental revenue contributions that we're getting from the revenue that they're getting or it's revenue directly to us. But in all those cases, we want these to be positive marginal contributions. So nothing that would tend to make our model head in a regressive direction. We feel pretty good about some of the engagements. And frankly, some of these have also been yielding results and are already part of our results over the past couple of periods. So we feel pretty good about these engagements.
Okay. And then, John, just finally, maybe a little update on the progress on the DFI tool in the field?
Yes, we are quite optimistic about that. We are currently operating in full manufacturing with tens of wafers a week, which doesn't allow time for any research and development work or setup, but I believe everything is progressing well. At the SPIE conference, several semiconductor engineers shared their firsthand experiences with the capabilities we demonstrated in our keynotes. They discussed their observations regarding the machine's performances and our overall approach, and we were very pleased with the positive feedback received in that public forum.
Great. And then do you think this is essentially augmenting, I guess, call it, the optical inspection work? And are they working closely together? Or is just one set up to replace it ultimately at...
I think it's generally, the devices have gotten many, many years ago, we involved in this whole thing was a focus on the things you can't see. I think one of the technology leaders from NVIDIA, when he spoke at our user conference that, expect the invisible. If you look at the early applications, it's a lot to look at open contacts and vias, which you can't see optically because the surface of the wafer looks great. The machine's capabilities are scanning billions of those per hour. And when you look at a complex chip, even a 50 square millimeter chip, there's so many billions of those features on a chip that you need to have visibility to 5 billion or 10 billion minimum per wafer just to see if you're at your target yields. So we feel we have the only ability to see opens on an in-line inspection case that's very much dependent on all the analysis system does on the design. Database as a software, our DFM software for that. And then all the analysis on the back end with Exensio to look for the trends and the layout features that are there. So I think it's quite complementary to optical inspection. And our industry for decades has had a problem of not being able to see open now that we're trying to do more and more with 3D open across layers is an increasingly important problem. So we feel like the DFI program is skating to where the puck is going in terms of inspection problems.
Your next question is from Gus Richard from Northland.
Nice Quarter. Just on the top line, 90% is now the analytics business, which I would imagine is pretty predictable. Can you give me a sense of how much of your guidance between '20 and '25 is sort of already in backlog, if you will?
Yes. The color we give on our backlog that we put in our 10-Q as well, consistent with the last 2 quarters of this backlog more than half is within the next 2 years. I think within that, it's also fair to say that the majority of that happens within this next year. Look, we're starting to get increasingly confident every year. We don't break out the specifics, but I will tell you in our internal board decks over the last 2 years, we've been putting every year as part of our annual operating plan, what percentage of next year is booked for. And I can tell you that percentage has been going higher. So it's all a function of having this recurring revenue across a variety of streams of our business. We keep working on other areas. Today, for example, on the Cimetrix side that are run time to also explore how to make those recurring as well, and parts of that are headed there, but again, very small and lots of opportunities, but increased confidence.
Yes. From a product standpoint, the Exensio and the leading-edge stuff, Gus, is all ratable, right? There's some perpetual licenses on the Exensio side, but in any given quarter, it's under 10% of the analytics revenue. The piece that we're still working on, I think, the runtime licenses on Cimetrix come down to when customers ship, as I said in my prepared remarks, we had anticipated them shipping a little bit more this quarter than they actually shipped. We saw our backlog go up to records, I mean, really super level, which gives us some predictability about runtime licenses over the next couple of quarters. But we're still learning that piece of the business, Gus. In that piece, it's kind of a shadow backlog. We know they're going to ship something, but we don't really know each quarter how much it's going to be.
Okay. That was super helpful. And then R&D came in a lot higher than I thought, and gross margin came in a lot higher than I thought depending on what an engineer is doing, they're either allocated to COGS or R&D? Was there a little bit of that in the quarter? Do we think about R&D at this run rate? Or is that going to go up a little bit more?
Yes. So a couple of comments, right? One that I think Tyler already asked. So we feel pretty good about our gross margin targets going forward. So that should be a proving point about how we're feeling about the cost and the management of cost. Second, specifically within the quarter, look, it's also that time of the year starting with Q1 where we start to accrue some personnel-related bonuses and things like that. So that's part of the reason why you're seeing a little bit of a jump as well. Most of our jump was related to some of these accruals and some onetime things related to the headcount. But overall, on the R&D as a percentage of margin and our gross margin target is feeling pretty good about the rest of the year where we are.
Very helpful.
To provide some context on that question, when we initially moved Exensio to the cloud, we didn’t fully utilize many of the features available in the cloud systems. As I mentioned regarding tiered storage, we are now leveraging capabilities offered by cloud hardware that our on-premise customers don't access. This allows us to tell customers that managing backups can be costly and inconvenient, making it difficult to retrieve them when needed. Engineers often request backups during field returns. We now provide a feature called tiered storage in Exensio, and we are introducing an additional tier this year, enabling the software to automatically move data to progressively cheaper, colder storage. For engineers, this process still functions like an SQL query, so they no longer need to ask IT for access to backup tapes. The data is always accessible. From a cost perspective, this is very beneficial for the customer, and these enhancements are improving the margins on Exensio cloud deployments by giving customers more features that help reduce their overall analytics costs while increasing their spending with us. This is a key factor behind the significant growth in ARR associated with such features, and we plan to introduce more of these improvements.
Got it. Got it. All right. Very helpful. And then I think my last one is the quick-start contract expanded to a full-blown contract, and you're talking about helping your customer develop PDKs. And I'm just wondering if you could talk a little bit more about that. I don't believe I've ever heard you work in that arena before. And just kind of wondering what it is you're doing for the customer and sort of how you're helping them along with their customer enablement.
Yes. With advanced technologies, the interaction between Foundation IP and modeling support must be validated with Silicon across a wide range of layout usages. While you can identify your design role, that design win will behave differently in various density and neighborhood environments. We have consistently maintained the highest density of information per unit time and per unit area in the industry. When evaluating the analytics from PDF systems, an engineer once noted that we test about 30% of the test chips we design and analyze a similar percentage. This means we are looking at just 10% of our data, while the PDF systems allow us to examine everything. This thoroughness is crucial when transitioning from a limited set of IP to a wider array of designs as we expand the diversity of designs in cutting-edge technology. This typically occurs when moving from a single product family to multiple families or enhancing factory availability. Our capability has always existed, but it was initially offered as part of gain share. Now that we've unbundled the gain share, we can create and license a range of applications on a subscription basis. Previously, we essentially gave away much of this to help customers meet yield targets, which we realize now may have been insufficiently capitalized or monitored. We can now directly sell it to the segments of the organization that need it and charge separately, thereby delivering additional capabilities through that application over time.
Okay. I understand what you're doing. Is this capability per node? Is it based on a time-based licensing that allows use anywhere? How...
It's a time-based license, I mean it's a subscription across node or a family of notes depending on where the contract is structured. It comes with a set of vehicles set of Exensio Cloud, analytics and part of that is additional vehicles. So as they bring on new products, new product families, sometimes associated with third-party customers that are able to use the vehicles in conjunction with those IP families. So it's really around that design manufacturing interface.
Got it. Got it. And then if they need additional CDs, is that incremental charge for them? Or is...
There's incremental charges on top of a base level. There is an assumption associated with MPWs for a certain amount of customers that are coming into the product to a technology node and then there's upcharges if they want to use more than that as it's designed.
Got it. Got it. And then if you can help me out, is this as the size of an old IYR contract? Or is it smaller? Can you size it relative to prior opportunities or business models?
Yes. These contracts typically fall into the range that, as shown in our announcements, the quick-start was in the single-digit million range. They generally fall in the double-digit millions recognized over a few years, so they are relatively significant. Our goal is to extend the subscription period to ensure that the system continues to provide value over a longer time frame. Monitoring the renewal process will be crucial for us, and the more we see these renew, the more satisfied we will be. We are still in the early stages of this.
Your next question is from Blair Abernethy from Rosenblatt Securities.
Thanks. Nice quarter, guys.
Thank you.
I have a couple of follow-up questions. John, I'm curious about the DFI. First, are you experiencing any supply chain issues with components that might be causing delays for you now or in the near future? Secondly, regarding the backlog, the sequential growth of 10% is quite strong. Is that entirely driven by Exensio, or does DFI contribute to that as well? I'd like to understand more about the reasons behind that backlog growth.
Sure. To address your question about the supply chain, we've encountered some requests from customers, particularly those using computers who want to implement Exensio process control on-premise and have asked us to acquire the necessary computing configurations for Exensio. We've noticed that lead times for various components have increased, similar to industry trends. However, there are often ways to prioritize and expedite orders if needed. We're confident in the availability of resources for the DFI eProbe program for the rest of this year. Our forecast for this year includes a 20% to 25% growth, which isn't hindered by the availability of any specific part. We don't see shortages as a significant issue for us at this time. While prolonged shortages could impact us over several years, we believe that for this year, they won't restrict our business. Particularly for our cloud customers, I often highlight that we can deploy Exensio on the cloud almost immediately. In contrast, ordering and shipping equipment for on-premise installation might take 6 to 12 weeks. Therefore, cloud availability is generally much more rapid. For the DFI programs, we have sufficient capacity to meet our needs for the rest of this year. That addresses the supply chain question. I'll now hand it over to Adnan for your second question.
Yes. I think on the second question, look, I mean, if you look within the analytics business, again, as others have pointed out, 90% plus of revenue. And we feel pretty good. I mean within that, when we look at the three components that John also talked about, right, the Exensio piece, the leading-edge pieces or the CCG piece. We track progress of each one of them within our businesses. And we're pretty pleased with how they did quarter-over-quarter and obviously, year-over-year as well. Then in terms of a booking color, which you asked about as well, yes, when we do sign these larger deals, leading-edge deals will contribute to the total booking growth, of course. But at the same time, we're also pleased with the booking that we saw in the Exensio business and then as well as the CCG business. I'll give you two colors, right, John already talked about the leading edge. But with an Exensio, it's the type and the quality of the booking that we're starting to focus much more on. The comment that John made about the ARR growth for that customer that came up for renewal of greater than 50% ARR, that's precisely the kind of thing we like to drive. In the CCG business, yes, it is mostly perpetual and we book and we ship, but what's important is how much more booking there is that we weren't able to ship or might be for booking in the future. So the backlog even for that business starting to reach very strong levels is another positive indicator. So all in all, tool pretty good. Of course, focusing on making sure that we continue to deliver growth from all the three components of analytics.
Okay. Great. And just shifting to your margins, gross margins were solid this quarter. As you look at Adnan, as you look at some of these partnerships that you're supporting, you're taking on versus your R&D, internal R&D spend, how are you looking at capital allocation, if you will, or you have limited resources, so how are you sort of deciding where you're going to spend this year? And what's sort of your hiring outlook for the rest of this year?
Yes, that's a good question. One reason is the increase in headcount spending, especially in R&D and overall operating expenses. We were pleasantly surprised by our ability to hire more staff than we initially anticipated, and we managed to expedite the hiring process, which is encouraging in this environment. This contributed to the rise in R&D costs. Additionally, when business performance is strong, sales bonuses and commissions also increase, which is another factor. At this time of year, we begin to accrue bonuses as well. Regarding capital allocation for new business, we evaluate each opportunity based on whether it will generate revenue through partnerships or involve new deals. We assess resources on a case-by-case basis and sometimes hire contractors as needed. We currently utilize a significant number of contractors and will adjust our usage based on business demands. Our focus remains on ensuring we maintain and gradually improve our margins.
All right. I'm showing no further questions at this time. I would now like to turn the call back to CEO of PDF Solutions, Mr. John Kibarian for closing comments.
Thank you for participating in our Q1 call. We look forward to talking with you again soon. Have a great day.
Ladies and gentlemen, this does conclude PDF Solutions first quarter conference call. Thank you for participating. You may now disconnect.