Penguin Solutions, Inc. (PENG) Earnings Call Transcript & Summary
January 5, 2021
Earnings Call Speaker Segments
Suzanne Schmidt
executiveThank you, operator. Good afternoon, and thanks for being with us on today’s earnings conference call, where we’ll discuss SMART Global Holdings’ first quarter fiscal 2021 results. Joining me are Mark Adams, Chief Executive Officer, and Jack Pacheco, Chief Operating and Financial Officer. We’re webcasting this call from our site at smartgh.com. Our website also has an accompanying slide presentation and the earnings press release. We encourage you to visit our website throughout the quarter for the latest information about the company, including details on the various financial conferences we’ll be attending. Before we begin, I want to point out that today’s remarks and the responses to questions may contain forward-looking statements. Anything that talks about what we expect, what we project, or how we describe things that haven't happened yet — including financial forecasts and where the market might go — counts as a forward-looking statement. What actually happens may be materially different from what those forward-looking statements say. To learn more, take a look at the forward-looking statements disclosures in our earnings press releases, and also the risk factors covered in the documents we file with the SEC from time to time, among them our latest Form 10-K and Form 10-Q. We're not obligated to update these forward-looking statements, and they speak only as of today. Also, on this call we'll be talking about non-GAAP financial measures. You can find reconciliations to the comparable GAAP financial measures in today's earnings press release. We'll start with Mark Adams, our CEO, for a business update. Jack Pacheco, COO and CFO, will then go over the financials and forward guidance, and following that we'll open it up for questions. Mark?
Mark Adams
executiveThanks, Suzanne, and happy 2021 to everyone. I'd like to use this moment to express appreciation to our team members around the world for their dedication and resilience while we navigate these uncertain times. In my first full quarter at SGH, the team's work ethic has really stood out to me, and I feel optimistic about what we can achieve in carrying out our strategy of growth and diversification; from my perspective, success comes mostly from people, purpose, planning and process. That's precisely what I've concentrated on during these initial months with the company. Is the organizational structure positioned for success? Is the team unified around purpose—around what we need to accomplish and, equally critical, what we should avoid doing? Once purpose is aligned, do we possess the correct plan for addressing the company's priorities, or are there situations in which our ambitions require more clarity, more investment and more commitment? With the proper structure established—purpose aligned and the correct plan in place—do we have the proper process for executing and measuring performance, enabling us to hold ourselves accountable? Although we remain a work in process in some of these areas, I am more excited today than I was 120 days ago. The future at SGH appears very bright. From a strategy standpoint, our focus stays on providing distinctive offerings across every business line. Going forward, we're targeting expansion into growth areas — high-performance computing, artificial intelligence and cloud via Penguin Computing; Edge Computing via our embedded unit, previously Artesyn Embedded Computing; IOT solutions via smart wireless, formerly Inforce Computing; and advanced package memory, low density storage and in-memory computing inside our memory solutions business, which covers specialty memory and the Brazil memory business. Financially, each business line is charged with boosting profitability, and on this call as well as future ones, we'll offer evidence showing how we're advancing along the way. Turning to how we did in the first quarter: we brought in $292 million in revenue, which is 7% higher than the same quarter in fiscal year 2020. Our gross margin finished a little above the middle of the range we had guided to, and non-GAAP earnings per share came in at $0.78, which was above the top end of our guidance; on top of that, the balance sheet keeps getting stronger — cash and equivalents grew 9% from the previous quarter and now total $164 million. Overall, I felt good about these results, since we continue to make progress on becoming a diversified growth company. Now I'll walk through a little more detail on how each of our businesses did. To start with specialty, compute and storage: revenue and gross margin were essentially unchanged from the quarter before, coming in at $66 million and 27%, respectively. On our last call, I mentioned that we are conducting a detailed review of each business line, and where it makes sense, we will either find ways to improve gross margins or exit lines that don't meet our margin targets. One example was the recent decision to shut down our battery operation in Brazil. We will move aggressively to pursue opportunities, but we'll be disciplined about investment and about monitoring results. In specialty compute and storage, we're focused on building higher-margin business by making use of software and services in the solutions we offer. That value-add emphasis gained confirmation through the formal introduction of the Penguin Computing solution strategy, comprising four specialized solution practices aimed at helping customers adopt artificial intelligence, high performance computing, cloud and data analytics. For each of these four practice areas, we'll bundle software and services tuned specifically. Although expansion in the high performance computing and artificial intelligence market is largely propelled by newcomers to these fields, current enterprise clients are gaining from access to these emerging technologies as we assist them in bringing such capabilities in-house. Confirming this fresh direction, Penguin earned the HPCwire Reader's Choice Award for best HPC Solution and Financial Services. Our key partners also acknowledged us. Intel gave Penguin Computing its Executive Summit Award for exceptional platform innovation, and WekaIO picked Penguin as its partner of the year for 2020, recognizing software defined storage. SMART's embedded business secured a $30 million order from the U.S. government for a ruggedized ATCA system built to operate under severe shock and vibration conditions. On the technology side, our embedded team introduced a new advanced telecom computing architecture memory blade, or ATCA memory blade, targeting industrial and military edge computing applications. One customer usage model example for this technology involved a leading semiconductor company looking to use our memory blade product in an edge computing solution to support onsite maintenance predictability and an AI application. While some strategic programs and specialty compute take time to develop, I feel good about the pipeline in front of us over the near term, which demonstrates that customers endorse what we are building. We expect specialty compute revenue to grow over 20% versus our first quarter. I'm excited about how focused our team is on solutions and value-added services, and I see that as an engine of future growth for the business. Turning now to specialty memory, which recorded $120.7 million in revenue for the quarter, up 16% from fiscal Q1 2020. In the first quarter, our DDR-3 lineup turned in a solid performance, driven by stronger demand and DRAM prices that were leveling off. We also saw increased interest in our persistent memory, or NVDIMM, products aimed at storage uses. Since we want to broaden both our customer base and our end markets, specialty solid state storage, or SSD, is a major growth focus for us. Our latest SSD product is now being sampled by customers, and it relies on a controller that SMART Modular built in-house. The team is concentrating on moving into additional vertical markets, among them surveillance and transportation. On the strategic side, our specialty memory group continues to look at how we can provide more valuable memory solutions to customers whose focus is enterprise, cloud, AI and industrial applications. In Brazil, our total revenue came to $105.2 million, which is up nearly 12% from fiscal Q1 of 2020. Higher memory density fueled robust mobile sales during the quarter, rising almost 30% year over year; we also saw stronger memory sales for notebooks, up 27% versus the first quarter a year earlier, helped by the growing trend of more people working from home. We keep speeding up new product introductions to support our Brazilian customers. We qualified several high density products for mobile applications, among them a 64 gigabyte and 128 gigabyte eMCP. Our team right now is working through qualification of SSD production within the country, and that effort draws on our sophisticated packaging know-how, our key customer ties, and local manufacturing resources. Next, a quick status note on the Cree LED deal we have not yet closed. We’re still genuinely enthusiastic about bringing Cree LED into the SMART family. Its leading position across specialty LED segments fits nicely with the broader specialty solutions approach we pursue — one centered on expansion, better margins, and a more varied portfolio. As we jointly develop a smooth integration blueprint for Cree LED within SGH, I keep coming away impressed by that company’s leadership group, its culture, and the discipline with which it runs operations. Regulatory developments have been encouraging, and our teams are putting in strong effort toward integration milestones. We believe that should lead to possible flows around late February or early March. At this point I'll hand things over to Jack, who will walk through the financials and Q2 guidance in more detail. Jack.
Jack Pacheco
executiveAppreciate that, Mark. Net sales for fiscal 2021's first quarter came in at $291.7 million, which beat the midpoint of the guidance range we'd given, and the driver was combined memory sales—specialty memory in Brazil included—climbing 14% compared with last year's same quarter. Non-GAAP gross margins landed at 18.6%, while non-GAAP EPS finished above the top end of our guidance at $0.78 per share. Mark briefly mentioned this in his remarks, and it's worth repeating: our balance sheet continues to strengthen—cash and equivalents rose $13 million over the quarter to $164 million—and we're still concentrating on improving inventory turns, which climbed to 10 this quarter. For the first fiscal quarter, end-market net sales were split this way: mobile and PCs accounted for 34%, network and telecom 19%, servers and storage 17%, and industrial, defense and other 30%. As a share of sales, both mobile and PCs and servers and storage increased from Q4 of fiscal year 20, and together they made up 51% of revenue in Q1 compared with 43% in Q4. Networking and telecom fell 6% quarter over quarter, which reflects softer enterprise spending during the quarter just ended. Turning to the remainder of the income statement: non-GAAP gross profit in the first fiscal quarter came to $54.1 million, equal to 18.6% of net sales, compared with $57.8 million, or 19.5% of net sales, in the prior quarter. Here's how the non-GAAP gross margin broke down across our business groups: Specialty Compute and Storage came in at 27%, Specialty Memory at 15%, and Brazil at 17%. Non-GAAP operating expenses landed at $30.4 million, up from $29.4 million the quarter before. For the first quarter, non-GAAP net income reached $19.6 million, or $0.78 per diluted share, versus $20.4 million, or $0.82 per diluted share, in the prior quarter. Adjusted EBITDA came to $29.5 million, compared with $33 million in the prior quarter. Our non-GAAP effective tax rate for the quarter was 14.1%, which matched our expectations. On working capital: net accounts receivable were $212.9 million, down from $215.9 million last quarter. Days sales outstanding held about steady versus last quarter, at 46 days. At the end of Q1, inventory was $147.2 million, down from $163 million at the close of Q4. Turns came in at 10 times, up from 9 times the prior quarter, reflecting our continued efforts to use materials more efficiently. As in the past, we calculate accounts receivable, days outstanding and inventory turnover using gross sales and cost of goods sold, which for Q1 were $423.2 million and $370.6 million, respectively. Just a reminder: the difference between gross revenue and net sales stems from our supply chain services business, which is recorded on an agency basis—so for a supply chain services transaction, only the net profit shows up in net sales. We closed Q1 with cash and cash equivalents of $164.1 million, up from $150.8 million at the close of the previous quarter. Cash generated from operations for the first quarter more than doubled, reaching $35.2 million compared with $16.2 million in the prior quarter; on a trailing 12-month basis, operating cash flow totaled 88.7 million; for those keeping an eye on CapEx and appreciation, capital expenditures came in at $14.6 million, consistent with our expectations for the quarter, while depreciation was $5 million. We further strengthened our liquidity sources by signing a $100 million ABL with Bank of America on December 23rd. The effective rate on that facility is 2.25%, with a variance of 0.25% either way based on how much is drawn. As of right now, that line has no balance outstanding; we also hold a $50 million revolver, which is likewise undrawn. With a balance sheet that keeps getting stronger, we believe we're well set up to succeed going forward. Now, moving to our fiscal Q2 2021, I'd like to start with some background on our guidance. The guidance we're giving incorporates an accounting change implemented in our fourth fiscal quarter — fiscal year 2020 for Brazil — which reduced our gross profits and operating expenses by equal amounts; keeping that in mind, I'll now walk through our guidance for the second quarter of fiscal 2021. For the second quarter, we right now project net sales between $285 million and $305 million. We estimate gross margin for the quarter at roughly 18% to 20%. GAAP earnings per diluted share should come in around $0.38 per share, plus or minus $0.05. On a non-GAAP basis—leaving out share-based compensation expense, intangible asset amortization expense, and convertible debt discount OID and fees—we anticipate non-GAAP earnings per diluted share in the range of $0.80, plus or minus $0.05. Our second-quarter guidance doesn't factor in any forecast for foreign exchange gains or losses, and it assumes an income tax provision of roughly 10% to 14%. For estimating diluted earnings per share in the second fiscal quarter, we're using 25.6 million shares. We expect capital expenditures in the second fiscal quarter to come in around the same as last quarter, between $10 million and $15 million. For more detail, please see the non-GAAP financial information section, along with the reconciliation of non-GAAP financial measures to GAAP results and the reconciliation of GAAP net income to adjusted EBITDA tables, in our earnings press release. Operator, we're ready for questions now.
Kevin Cassidy
analystThanks for taking my question, and nice work on the strong results. My first question is about the market — Mark, you said the memory market, the DRAM market, is stabilizing. During the last downturn, we saw that a number of companies making phones, smartphones in Brazil, chose to manufacture them outside Brazil. Now that prices are recovering, will that change? Is there a trend toward making more phones inside the country?
Jack Pacheco
executiveHey, Kevin, this is Jack. Yeah, I mean, I think phones in the country will keep being built — those that are in the country will keep building in the country. I don't think we'll see any new entrants at the moment. We're not getting any — news is going to come in Brazil, so I think <<inaudible>> the current, the current companies that keep building the phones in Brazil have been building. We don't think it'll…
Kevin Cassidy
analystAll right, so if seasonality stays the same, then, I suppose, that's another way to answer what I asked.
Jack Pacheco
executiveSeasonality is unchanged at the moment. We aren't seeing — because when you consider the phone, there's a great deal of flash memory inside it — and we aren't seeing flash memory prices rise; flash memory keeps getting cheaper.
Kevin Cassidy
analystAll right, got it. And Mark, perhaps picking up on what you said regarding profitability, especially at Penguin—since you're leaning into software and services—could you share what headway you've made and how much appetite exists among your customers? Also, which customer base would be the one keen on your cloud or software offerings?
Mark Adams
executiveAbsolutely. Thanks for that question, Kevin—I appreciate it. If we examine the pieces that could help us capture these margin opportunities and really deepen our overall ties with federal as well as commercial customers, the way to frame it is that there are essentially three categories. There's what I'd call hardware optimization, then software, and then services—which cover things such as on-demand style offerings plus systems implementation, installation, and full field service. So across each of those, we're investing to make our offering stronger. Regarding software, most of what we mean falls into a handful of areas: workload optimization, resource provisioning, and a concept known as data gravity—which becomes relevant once customers begin wanting a hybrid deployment of cloud services for particular workloads, with the more compute-heavy workloads kept on premise. So we're building opportunities through partnerships with software companies alongside Penguin, so we can deliver those kinds of capabilities on the software front. On the services front, there's clearly presale configuration analysis, plus systems design work carried out to tune our platforms for the end-market applications we're targeting. Next, there are installation services and post-sale capabilities that we provide to the customer. Many of these customers simply don't have the infrastructure and resources on hand to handle the hardware side of data center installation—even after the sale, around break-fix situations. And then, over the longer term—I believe we've mentioned this before, and I just want to underscore it—we're looking very seriously at on-demand services at Penguin and putting investment into business models that let us grow our existing customer relationships and also connect with new customers on the on-demand side, whether that's multi-cloud or again a public-to-private or private-to-public style architecture built on on-demand models. A portion of the infrastructure is already built out, and we're aiming to roll out additional capacity in the POD area—POD being the acronym for Penguin On Demand—and then gov POD for our federal customers. So that's essentially how we view the enhancement model; and it's a partnership that today is mainly with third parties on the software side, while we assess how we want to participate in that part of it. Yet much of the infrastructure we already have in place — the same infrastructure that has helped make Penguin such a formidable force in HPC — lets us deliver that sort of added value around these systems, including the on-demand piece of the equation.
Kevin Cassidy
analystAll right, great. Thanks. And perhaps I'll just tack something on — building on that point. Regarding the sales process for that, is it six months or a year? I suppose, how long does it take from when you land the customer? A — Mark Adams So usually, you're sort of splitting this apart, since some of it is in fact already happening, and I think you caught my remarks earlier today on the call regarding our Q2 — a portion of this is already underway. Frequently it's a systems-level sale within one company involving these new tiers of services that we're providing; so we do believe margins will strengthen in the near term. On your question about sales cycles, the pre-sale phase — identifying a customer and a live project with funding, whether on the federal or commercial side — can run anywhere from three to six months. After that comes the process where we work to understand what the customer needs within a given vertical and which applications they want to optimize around. That is really where Penguin's expertise comes into play, because we have been in high-performance computing for so long that we bring a skillset and competency our customers truly value when it comes to understanding the system and optimizing it for those applications. We can essentially replicate that and validate it across those various verticals. So I'd estimate three to six months for the presale portion, followed by probably another six to nine months for a phase one deployment. As I mentioned during the previous call, I was starting to notice signs that we hold some truly solid prospects within the pipeline — and by pipeline, I mean the deals we are in negotiations over — and those have been converting quite favorably. Kevin, we entered Q2 with a backlog 40% greater than the one we had entering Q1. Handling that cycle is clearly something of a craft, in a manner of speaking, that we have to keep refining, yet as the business scales, the funnel's size and the commitments we secure against the in-quarter forecasting timeline are growing more predictable for us. Great. Thank you. A — Mark Adams Thank you. Operator Thank you. Our following question is from Raji Gill at Needham & Company. Your line is open at this time. Q — Raji Gill Yes. Thank you, and congratulations to you as well. Mark, please correct me if I'm mistaken. You said you anticipated the Penguin Computing business would expand 20% quarter over quarter in February—is that right? And assuming so, what's fueling that as we head into the February quarter? Also, Jack, how do you see the outlook for the Brazil business and the Specialty Memory business—what are the pluses and minuses for the February quarter? A —
Mark Adams
executiveThanks, Raj. To begin with that 20% figure — it represents our Specialty Compute segment as a whole, and Penguin has it exactly right; we hold a really robust backlog within the federal side of the business, plus a number of freshly accepted commitments whose deliveries will kick off during Q2 in our Penguin operations. Federal and commercial clients, then, make up our two main customer categories. We prefer not to disclose particular customer names, for understandable reasons, though what I can share is that every one of the orders I'm mentioning runs into the tens of millions of dollars — and bear in mind that billing for them doesn't all land within a single quarter. Taking into account the healthy backlog I mentioned — which again was up roughly 40% entering this quarter compared with Q1 — plus a funnel that continues to expand, it's fair to expect Q2 will land in a pretty good spot for the Penguin business. I'd also like to highlight that our embedded business landed a $30 million order. That order will ship across the entire calendar year — pardon me, the entire fiscal year — and roughly 20% of it shipped during the first quarter. So with the stronger Penguin funnel combined with some large commitments on the embedded side, we're feeling fairly bullish. This isn't something I see as lasting just a single quarter. We feel quite positive about how our specialty compute business keeps expanding and how well it's being executed; there's still a great deal of room ahead, and I believe substantial upside lies in store down the road. Jack, would you like to speak to that Brazil point on Specialty?
Jack Pacheco
executiveYeah, Raj, if you consider our expectation for where they'll land in Q2 compared with Q1 — we told you Brazil would be seasonally lower. For the memory business, I'd anticipate a decline of low single digits percentage-wise versus Q1, within that range. So quite a few different — not a dramatic change from how we closed the one, but somewhat lower. Q — Raji Gill Alright, so specialty memory is roughly a low single digit sequential decline, and Brazil is seasonally lower? Yeah, somewhat. Q — Raji Gill Somewhat. Okay. Understood. Alright, understood. Turning to specialty memory, you mentioned $121 million in November, up from $103 million in the same month a year earlier. Could you expand on what's driving that? It looks like storage and networking have picked up somewhat, but I'm curious whether broader macro factors are behind the year-over-year gain in specialty memory, and how we should view it going forward — including through this year, which could bring a post-COVID recovery. I'd just like to know how your business might stand to gain if a COVID recovery takes hold. Honestly, the expansion you've witnessed over the last twelve months has come primarily from DRAM. DRAM did nicely. Mark noted that DDR-3 enjoyed a strong quarter. We continue to win new designs with DDR. So the DRAM business is performing — specialty has stayed low. Going forward, I think — as we approach the second half of this year, we anticipate the flash business will improve. Flash got slammed harder by COVID-19. As you're aware, we were working to secure new design wins. Securing design wins has been tough throughout the COVID-19 period, given that customers aren't present in their labs, which has made the process considerably slower. So, as those begin to take effect, I believe that when recovery arrives, you'll see flash improve in the latter half of the year for the specialty memory business.
Jack Pacheco
executiveI think that's going to help us in the back half of the year push our margin up, as we discussed for that business too. So we still need to get that business on track. We've done really well, and we believe the flash business will begin to improve here as we move toward the latter part of this fiscal year. Q — Raji Gill And my final question concerns gross margins: guidance implies roughly 40 bps of improvement at the midpoint on somewhat higher revenue volume. You also broke out the gross margin split for this quarter. So setting Cree aside, what margin levers will you pull for each of these business lines? And how should we think about Brazil improving margin, specialty memory improving margin, and so on? A — Mark Adams I'll begin with — I think Brazil, Brazil, we went — we've discussed for some time that we're working to move toward a more fixed-price model in Brazil and the unit — our value add per unit in Brazil. And so I think — as we — as Brazil grows through this year, you may not see gross margins in Brazil improve all that much. We're trying to be more — make gross margins somewhat more stable, right. If ASPs rise, if revenues rise, we'll capture more value add. I think our gross margin percentage will more or less hold steady. So we're really not expecting major changes from Brazil; in specialty, we said we thought we'd bring specialty back up toward '20 as we exit the year, and we still believe we can achieve that, and it will be driven by flash. A lot of new design wins in flash — and also the return of some additional enterprise customers. On the flash side, we've seen some weakness in enterprise that has really hurt our flash business, which tends to carry higher margins. We see that coming back as well in the back half of the year, even with specialty memory. And then on — but just one clarification before I address specialty compute, on the margin side, during my remarks about profitability, we're not going to enter the flash business simply as a revenue play. And as Jack mentioned, some of our design work and efforts have taken a bit longer than we'd like, but again, the emphasis will be on specialty flash storage solutions, not merely commodity trading of NAND. That's not a business I'm going to let us get into. So as he mentioned, some of the application work we're doing on specialty NAND storage is playing out. It's just a — it isn't on the timing we'd like, but it's still going to be — we're going to be very disciplined in how we, how we compete in the NAND space. I think the NAND business will be significantly more volatile than DRAM, and we'll be very cautious in that area. Even though we anticipate growth and solid results in the back half, we had simply hoped it would arrive somewhat sooner. Regarding specialty compute, I believe I discussed this in relation to software services and, essentially, as we shift toward applications with more value-add that we sell into; and I think that will be, once again, quite comparable for us going forward. We're going to be far more disciplined when we fold these into the deals we want to pursue; we aren't chasing hardware revenue alone; and you've already begun to observe that. Penguin was acquired. At the moment of that acquisition, it was roughly a 15% gross margin business, and I believe in the quarter we're above 20% gross margins. And I think I noted on the prior call, and I'll restate it today: I think our next stop in the near term, across the coming quarter or two, ought to land in the mid-20, and I hold many aspirations for a considerably improved gross margin picture at Penguin across the next three years or so, but we were on a solid path there and we'll remain selective because in this business you want customers who genuinely require your expertise across all those areas, not merely hardware, and we're identifying those customers. And some of these folks in the vertical markets we're discussing, like AI or sophisticated high-end machine learning that leverages high-performance computing, find us becoming far more important than simply a hardware provider, and those are the capabilities we're investing in and building, and I think that will be reflected in the margin as we move ahead. Q — Raji Gill And one final question, regarding Cree, the transition of production to Taiwan, is there any status update? A — Mark Adams So we're on track. I don't want to say too much because, obviously, the deal hasn't closed and that's for them to comment on. But I would simply note that we're very happy with where matters stand in the process and remain very pleased, obviously, with the strategic value they'll bring us, but just as we come to know the team better along with their operating philosophy and culture, we're very excited about them joining the team. Q — Raji Gill All right. Good deal. Thank you, Mark and Jack. A — Mark Adams Thank you, Raji.
Brian Chin
analystHello, good afternoon. Happy New Year, and thank you for giving us the chance to ask a few questions. Perhaps just to keep the conversation focused back on LED for a moment. Mark, I heard the timing you updated regarding closing the deal. I suppose, what are some of the indications you have that make you think you'll receive that regulatory approval over the next month or two? And could you, especially since the key hurdle there is probably in China?
Mark Adams
executiveYes. Actually — I don't believe regulatory approvals are an issue, and we had actually been, I shouldn't even call it an issue. There are just a couple of timing matters around integration with IT systems, because remember this is a carve-out and before the acquisition, Cree LED was part of a wider corporate IT infrastructure, which couldn't be carved out in a nice, tidy way. So I think the long pole in the tent, as they say, will be IT. One of our closing conditions was a third-party audit of the business, again because of the carve-out nature of this transaction. Getting a clear opinion of the carve-out financials was very important to us, and that is on track. But I would say, I don't think — we've received positive indication on the regulatory side, that isn't the issue we're facing relative to the end of February target.
Brian Chin
analystUnderstood. All right, thanks. One more question. Regarding the Cree business. Within Cree, the LED components unit bore a substantial amount of fixed costs. After you close SMART, none of the wafer or wafer fab — none of those wafer and wafer fabrication assets — will remain. So my question is, roughly right out of the gate, and assuming you do close, what would LED gross margins look like on a consolidated basis under SMART if we backed out the related depreciation? As I recall, fully loaded it was around 21% gross margin in our fiscal ‘20, and closer to 22%. A — Mark Adams Yes, from a modeling standpoint, I'd say it's reasonable to assume roughly a 400 basis point improvement. Now, on one hand, those fixed costs disappear, but on the other hand, when it comes to cost of goods in an outsourced model, those fixed costs don't really disappear. They may simply surface in cost of goods, in the pricing your partners charge you. That said, we believe there are considerable efficiencies, and we do see an opportunity for margin improvement with Cree. And again, we're being cautious until we're truly inside and can understand all of the upside potential, but we're fairly confident there will be some gross margin expansion in the LED business. Understood. So just to confirm I have this right, I believe you originally said that a year after close you could be at something like 200 to 400 basis points of improvement. And that includes the fixed cost, perhaps with a bit of shifting from in sourced to out sourced. I'll factor that in, along with other efficiencies. A — Mark Adams I'd say that's the right zip code at this point. Yeah, maybe just one final question, returning to the prior targets you set when you announced the pending acquisition. At that time, I think you said some organic growth was sort of dialed in there. But looking at the Cree business, it's $400 million plus in annualized revenue right now; and if you add that to where your revenues stood last fiscal year, it's kind of zeroish growth. And clearly, HPC, specialty compute sounds like it has a good growth orientation, certainly to start the fiscal year. I'm just wondering to what extent you might reevaluate those, or whether we can view them as a little conservative. A — Mark Adams Fair enough. Though, as I mentioned in my prepared comments, we're continuing to evaluate businesses that we might have been chugging along, and without much upside potential or gross margin. That isn't the kind of level we aim to reach as a company — or should I say, the kind of target we set as a company. So I think it's reasonable that, as we come to understand better what's happening within Cree beyond what our due diligence showed and where things stand today, there's potential for that. There's also a possibility that — if we can't get our arms around the gross margin on some of those lines of business I mentioned earlier — we may exit some of them. I don't expect it to be dramatic; I'd say the range we initially gave is definitely something I'm comfortable with today. And around the end of our calendar quarter — this calendar quarter — I'll probably be able to provide a bit of an update on how we see the operating model, after spending more time inside Cree LED and, obviously, evaluating some of the existing businesses within SGH. So on that revenue, and the margin improvement, some of this addition by subtraction that's contemplated sounds fine. A — Mark Adams But I think your question is a good one. I'm not dodging it. I believe that potential exists. We're not here to comment on it now because I just want more clarity on what's underneath Cree LED on specific product lines, and to see whether there's a need to reinvest in those businesses and scale them further beyond where they are. Because you have to remember, they've been fairly limited at some level in their ability to grow, given the fixed wafer capacity coming out of North Carolina; so there's a lot for us to learn about that business. We see how it operates today. But we want to make sure that — we want to invest properly. We think that — we think it's a really good acquisition for us, obviously. So I'm hedging a little because I want to get those answers done and get back to you, kind of toward the end of the first quarter, and be able to give you — first quarter calendar — a sense of how we kind of view the combined entity going forward. Okay. No, I appreciate the detail. Thanks Mark.
Mark Adams
executiveThank you.
Sidney Ho
analystThanks for taking my question. I want to build on what was said earlier about gross margin. You said that once the LED deals are closed, gross margin would improve by 200 to 400 basis points annually. Could you help us understand how much of that gain would come from bringing that business in versus how much would be organic? Mark, I know you mentioned some of the new profitability initiatives, particularly in Specialty Compute, and I was hoping you could put some numbers around them. One thing I've noticed is that specialty memory today is quite a bit lower than it was a year ago. And I believe during the prior quarter you mentioned that certain new products were creating a meaningful margin headwind; so I'm just attempting to understand how every segment can expand, and what sort of margin profile it might have excluding the LED piece?
Mark Adams
executiveAbsolutely. So to break it out once more, just as you said, apart from the LED topic we just covered, which I think you're comfortable with. On the organic side at SGH as it stands today, I noted that, Specialty Compute in the software and services space, I gave a few of the examples, particularly involving Penguin. My view is that this represents a genuine increase in their gross margin and benefits ours as well. Regarding Specialty Compute, keep in mind that this revenue stream also includes a logistics service operation, which likewise affects margin. So, when you pull that piece out separately—Jack mentioned some of the newer market opportunities we're putting money into for growth, again around specialty flash rather than commodity flash: specialty flash, lower density, our own controller, individual products that are kind of ruggedized or industrial-strength. These products, which are the target of some of our newer memory efforts, do carry higher margins. So I think, if you look at gross margin across the board—he already noted that Brazil is relatively flat—you'll see organic gross margin improvement of roughly 2% to 2.5% before Cree. And then, as I said, Cree has some margin opportunities of its own. My sense is that, taken together, these elements will add up to a fairly strong narrative overall.
Sidney Ho
analystGreat, that helps a lot. Now for my follow-up, I'd like to ask about the Specialty Compute and storage business. Clearly you've completed several acquisitions over the last two years, yet growth has not been—and simply isn't—all that consistent, for reasons beyond your control. So looking ahead, now that you're 120 days into the role, what growth rate do you think this business in aggregate can achieve organically? And what seasonality pattern should we expect going forward?
Mark Adams
executiveYeah, good question. I think you're touching on quite a lot there. Honestly, when it comes to growth in this area—and again, as you mentioned, four months into the job—I believe we need to make the leadership team stronger, and we will do that. I believe we need to continue investing, not only in people but in the company's capabilities, be it software, integration, or service capabilities. That said, I don't view this quarter as an anomaly when it comes to our growth opportunity. It's interesting, and I want to share this with the team on the call. This is a different kind of business. So trying to make it resemble a smart modular memory business simply doesn't work that way. What I mean is that it's a design-in business, and it will grow. Seasonality matters less here, and there certainly are some spending patterns tied to the federal business, but seasonality matters less and design-ins and wins and delivery schedules matter more. By the way, I happen to believe that will get crowded out by our growing backlog and growth opportunities. So—but I think this business isn't a fulfillment model, and it isn't a design-in model like the memory business; it's more of a point-to-point customer-to-vendor relationship design and sale, which is a longer-term sale than, again, the memory model per se. So, to put it in a long-winded way, I think the growth will crowd out. It's not really about seasonality so much as the lumpiness that may exist as we target growth in new vectors and new scale; but coming from where we are—coming from, as you noted in your question—I don't think you're going to see that. I think you're going to see continued growth for the foreseeable future, starting with Q2.
Sidney Ho
analystGreat, I appreciate that. Thanks, everyone.
Mark Adams
executiveWell, thanks once more; fiscal '21 got underway from a position of strength as we set out on our path toward profitable growth while broadening the business with a specialty emphasis. That's happening via wider strategic partnerships tied to growth markets like cloud and edge computing, IoT, high performance computing, enterprise storage and specialty memory solutions; we have reshaped the balance sheet to support future success. And we're currently assembling the leadership team so that this next stage of scaling our business becomes possible. Our attention is on devising fresh approaches that stretch existing business models and will show we can bring customers higher value solutions, with more emphasis on software services and on demand offerings. Among my top priorities is strengthening diversity within the leadership team and throughout the company. Beyond that, we're kicking off a process focused on our environmental, social and governance business practices. I'll have more to say on these during upcoming calls. To wrap up, I'm genuinely enthusiastic about what lies ahead for SGH, and I thank you for your interest in and support of the company. Thank you.
Operator
operatorThat wraps up today's conference call, everyone. We appreciate you joining us. You're free to disconnect at this time.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Penguin Solutions, Inc. transcript — plus 132,000+ transcripts from 7,000+ companies, speaker segments, AI summaries and full-text search — through the EarningsCalls.dev API.
Get the API View API docs →This call discussed
For developers and AI pipelines
Programmatic access to Penguin Solutions, Inc. earnings transcripts and 132,000+ others is available through the
EarningsCalls.dev REST API. Plans from $39.99/month — full transcripts, speaker segments,
full-text search, and the recently-added /api/v1/transcripts/recent polling endpoint for ETL pipelines.
More tools for stock research
OptionsBell: an unusual options activity scanner that tracks options flow across all US stocks and emails you when large trades appear in Penguin Solutions, Inc. or any ticker on your watchlist.
StockMarketScan: a stock screener with alerts: 24 pre-built screens refreshed every 5 minutes, watchlists and email notifications. Both offer API and MCP access and fair monthly pricing.