Q3 2025 Earnings Call Transcript

Good day, everyone. And welcome to the Lumentum Holdings Third Quarter Fiscal Year 2025 Earnings Call. All participants will be in listen mode only. Please also note today’s event is being recorded for replay purposes. At this time, I would now like to turn the conference over to Kathy Ta, Vice President of Investor Relations. Ms. Ta, please go ahead.

Thank you. And welcome to Lumentum’s fiscal third quarter 2025 earnings call. This is Kathy Ta, Lumentum’s Vice President of Investor Relations. Joining me today are Michael Hurlston, President and Chief Executive Officer; Wajid Ali, Executive Vice President and Chief Financial Officer; and Wupen Yuen, President, Cloud & Networking.

Today’s call will include forward-looking statements, including statements regarding our strategies, trends and expectations for our products and technologies, including demand, our customers, our end markets and market opportunities, our expectations and beliefs regarding recent acquisitions, including Cloud Light, macroeconomic trends, including the impact of tariffs and other trade regulations, and our expected financial and operating performance, including our guidance, as well as statements regarding our future revenues, financial model and margin targets. These statements are subject to risks and uncertainties that could cause actual results to differ materially from our current expectations, particularly the risk factors described in our SEC filings.

We encourage you to review our most recent filings with the SEC, particularly the risk factors described in our most recent 10-K and in our 10-Q that will be filed soon. The forward-looking statements provided during this call are based on Lumentum’s reasonable beliefs and expectations as of today. Lumentum undertakes no obligation to update these statements, except as required by applicable law. Please also note that unless otherwise stated, all financial results and projections discussed in this call are non-GAAP.

Non-GAAP financials are not to be considered as the substitute for or superior to financials prepared in accordance with GAAP. Lumentum’s press release with the fiscal third quarter results and accompanying supplemental slides are available on our website at www.lumentum.com under the Investor section. With that, I’ll turn the call over to Michael.

Thank you, Kathy, and good afternoon, everyone. Before diving into our third quarter results, I want to reflect on my first 90 days as CEO of Lumentum. I joined because of the immense cloud and AI opportunity. And the more I’ve seen, the more confident I am that we’re poised for success.

Having said that, we have opportunities to accelerate revenue growth, improve margins and focus spending. As many of you know, part of my core philosophy is that gross margins reflect the value customers place on our products. As we indicated at OFC, we see a path to take gross margins above 40%. Meanwhile, we can and should be able to improve operating margins by cutting spending in non-core areas.

At OFC, we outlined a path to drive revenue to $750 million a quarter, gross margins above 40% and operating margins greater than 20%. Based on my initial observations and the plan we have in place, these targets are all achievable. Our markets are growing at unprecedented rates, greater than 25% compound annual growth rate over the next five years, driven by an accelerating convergence of optics and electronics. Our strength in optical components is unmatched.

We build for virtually every type of network. At OFC, we showcased our 400-gig lane speed lasers alongside our proven 100-gig and 200-gig solutions. We also demonstrated differential drive EMLs that boost signal integrity and power efficiency. We’re building momentum in the transceiver market with three cloud transceiver customers already on Board and more wins expected in the pipeline.

What sets us apart is that our components are embedded across the ecosystem, even in competitor’s transceivers. That means we often win, regardless of who supplies the module. We’re one of the few companies with true end-to-end optical capabilities, from long-haul, to metro, to inside the data center. Our scale and breadth make us a truly differentiated optical solutions provider.

Let’s now turn to our third quarter results. In Q3, we exceeded the high end of our guidance for both revenue and EPS, driven by strong demand from cloud customers and a recovering networking market. Despite ongoing macroeconomic volatility, growth in cloud continues to be a key element of our financial performance. Revenue in our Cloud & Networking segment grew 8% sequentially and 16% year-over-year, fueled by robust demand from hyperscale cloud customers.

As mentioned earlier, when we execute well, the opportunity ahead is significant and our components business is performing at an exceptional level. We set another record for EML chip set shipments this quarter and remain on track to more than double this business by the end of calendar 2025, relative to our June 2024 baseline. We continue to ship 200-gig lane speed EMLs to multiple customers. With our set of design wins, we’re well positioned in the next-generation of 800-gig and 1.6T transceivers supporting AI workloads.

Our wafer fab expansion remains on track, supporting higher volumes of EMLs and other indium phosphide lasers and photodetectors. In addition, we are ramping production in CW lasers for silicon photonics transceiver applications in the quarter. As our indium phosphide capacity grows, we expect to ship an increasing mix of CW lasers. In addition to supplying components into the transceiver market, we took an early lead in co-packaged optics or CPO.

We have a highly differentiated ultra-high power laser that was announced as a key component in a CPO solution at GTC last quarter. We have already shipped early production units of that product and expect meaningful revenue in the second half of calendar 2026. At OFC, we introduced the R300, a 300x300 port optical circuit switch or OCS engineered to significantly improve the scalability, performance and efficiency of AI clusters in intra-data center networks. The device replaces traditional switches, providing customers significant power benefits by keeping signaling in the optical domain. Our OCS solutions build on decades of engineering expertise and the successful deployment of high performance MEMS technology in demanding telecom environments.

With over 1 trillion mirror operating hours in the field and a robust patent portfolio, our optical switches are designed for high reliability and energy efficient performance in AI driven data centers. At present, we have beta samples being qualified by multiple hyperscaler customers and expect to see early production volumes at the end of the calendar year. We’re accelerating our optical transceiver production at our Thailand manufacturing campus and remain on track to begin shipments to our second announced hyperscale data center customer in June as previously communicated. At the same time, we continue to ship to our third announced customer and we are expanding our product offerings and ramping in shipment volume to our largest cloud hyperscale customer.

In Q4, we expect our overall cloud transceiver revenue to grow over 50% sequentially. We’re also experiencing growing demand for our DCI and long-haul transmission solutions. We achieved another sequential increase in shipments of narrow line with lasers essential to ZR and ZR+ module deployments and in transponders for cloud customers. This marks the fifth sequential quarter of shipment growth for narrow line with lasers.

Even as we ramp additional capacity in Thailand, our shipments will not be able to satisfy demand for the balance of the calendar year. In addition, pump lasers and line subsystems were up sequentially driven by cloud infrastructure demand. Looking ahead to Q4, we anticipate strong sequential growth in our Cloud & Networking segment driven by new capacity coming online across our global operations, the continued ramp of customer programs and strengthening demand from network equipment manufacturers. Now let me move to our Industrial Tech segment.

Industrial Tech segment revenue decreased 5% sequentially but was up 14% from the same quarter last year. Industrial laser revenue declined sequentially but the drop was less than anticipated while 3D sensing revenue followed expected seasonal trends. In Q3, ultrafast laser shipments held steady at near record levels driven primarily by growing demand from a leading tool supplier supporting high volume solar cell manufacturing. We’re also actively collaborating with customers on new ultrafast laser opportunities as the minimal thermal impact of ultra-short pulse technology continues to gain traction in advanced packaging, display technologies and next-generation semiconductor processes.

Consistent with earlier remarks, we have taken actions to rationalize the Industrial Tech portfolio, closing two R&D sites and stopping development activities in three exploratory product areas. With these actions and more focus on the core business, we expect to see increasing profit in this segment over the next handful of quarters. Looking ahead to fiscal Q4, we expect a sequential decline in Industrial Tech revenue reflecting both the ongoing macroeconomic headwinds impacting Industrial laser demand and the typical seasonal decline in 3D sensing revenue. As I reflect on my first 90 days at Lumentum and look ahead, I’m energized by the momentum we’re building.

Our strong Q3 performance combined with the strategic positioning we shared at OFC reinforces that we’re on the right path. We’re executing on a focused strategy, investing in high growth, high impact areas where our differentiated technologies provide a lasting competitive advantage. From delivering record EML chip shipments to partnering with AI hyperscalers on connectivity innovation, we’re demonstrating that our products are essential to powering the future of cloud and AI. While macro uncertainty, tariff dynamics and export controls present near-term challenges, Lumentum has taken recent actions and delivered steps over the years to build resilience through a globally diversified manufacturing footprint, a flexible supply chain and active engagement with customers.

We remain focused on what we can control, pricing, disciplined spending and flawless execution. These fundamentals are positioning us to succeed across a variety of market environments. Finally, strong demand for cloud and AI continues to support our confidence in achieving the medium- and long-term financial targets we outlined at OFC. We remain on track to exceed a $500 million quarterly run rate as we exit the calendar year.

There’s still work ahead, but the opportunity in front of us is substantial. Our ability to deliver differentiated optical solutions across our portfolio provides a strong, resilient foundation for sustainable growth. Now, I’ll hand the call over to Wajid.

Thank you, Michael. Third quarter revenue of $425.2 million and non-GAAP EPS of $0.57 were both above the high end of our guidance ranges. GAAP gross margin for the third quarter was 28.8%, GAAP operating loss was 8.9% and GAAP net loss per share was $0.64. Turning to our non-GAAP results, third quarter non-GAAP gross margin was 35.2%, which was up 290 basis points sequentially and up 650 basis points year-on-year. Due to better manufacturing utilization and favorable product mix due to increased Datacom laser shipments.

Third quarter non-GAAP operating margin was 10.8%, which was up 290 basis points sequentially and up 1100 basis points year-on-year, and driven by improved Cloud & Networking profitability. Third quarter non-GAAP operating profit was $46.1 million and adjusted EBITDA was $71 million. Third quarter non-GAAP operating expenses totaled $103.4 million or 24.3% of revenue., an increase of $5.1 million from the second quarter and a decrease of $2.4 million from the year ago quarter. This demonstrates our disciplined cost management across the organization, even as we ramp investments in our growing cloud opportunities and absorb the typical payroll fringe reset that takes place each year in our third fiscal quarter.

Q3 non-GAAP SG&A expense was $40.1 million. Non-GAAP R&D expense was $63.3 million. Interest and other income was $2.9 million on a non-GAAP basis. Third quarter non-GAAP net income of $40.9 million and non-GAAP diluted net income per share was $0.57. Our fully diluted share count for the third quarter was 72.2 million shares on a non-GAAP basis.

Turning to the balance sheet. During the third quarter, our cash and short-term investments decreased by $30 million to $867 million. Our inventory levels increased sequentially to support the expected growth in our Cloud & Networking revenue. In Q3, we invested $59.5 million in CapEx, primarily focused on expanding cleanroom capacity at our Thailand manufacturing site and increasing equipment capacity for indium phosphide wafer production to support EML chip manufacturing.

Nearly all of our CapEx investment was directed toward our Cloud & Networking business. Turning to segment details. Third quarter Cloud & Networking segment revenue at $365.2 million increased 8% sequentially and 16% year-on-year. Cloud & Networking segment profit at 20% increased 380 basis points sequentially and increased 540 basis points year-on-year on higher revenue and favorable product mix.

Our third quarter Industrial Tech segment revenue at $60 million was down 5% sequentially and up 14% year-on-year. Third quarter Industrial Tech segment profit of 4.3% decreased sequentially on lower revenue and increased year-on-year on higher revenue. Now let me move to our guidance for the fourth quarter of fiscal 2025, which is on a non-GAAP basis and is based on our assumptions as of today. I want to acknowledge that the impact of tariffs on our supply chain, along with broader geopolitical dynamics remains highly uncertain.

The guidance we’re providing reflects our best assessment based on the current environment. This outlook includes an estimated 100-basis-point reduction in overall company gross margin, primarily driven by higher material costs and tariffs on shipments to U.S. destinations where we serve as the importer of record. For clarity, despite this 100-basis-point headwind, we expect a sequential improvement in gross margins from Q3 to Q4. We expect net revenue for the fourth quarter of fiscal 2025 to be in the range of $440 million to $470 million. This Q4 revenue forecast includes the following assumptions.

Cloud & Networking to be up sequentially with strong growth in products addressing cloud and AI applications, and Industrial Tech to be down sequentially with declines in both Industrial lasers and 3D sensing. We project fourth quarter non-GAAP operating margin to be in the range of 13% to 14% and diluted net income per share to be in the range of $0.70 per share to $0.80 per share. Our non-GAAP EPS guidance is based on a non-GAAP annual effective tax rate of 16.5%. These projections also assume an approximate share count of 72.7 million shares.

With that, I’ll turn the call back to Kathy to start the Q&A session. Kathy?

Thank you, Wajid. To allow everyone an opportunity to ask questions, please keep to one question and one follow-up. Jasan, let’s begin the Q&A session.

[Operator Instructions] Our first question is from Samik Chatterjee with JPMorgan. Your line is now open.

Hi. Good evening. Thanks for the question. This is Joe Cardoso on for Samik.

I guess first question, and maybe just a big picture question. I -- yeah, I noticed you guys didn’t necessarily talk about the $500 million revenue target that you initially laid out shooting by the end of the calendar year. I guess in a recent performance in the June quarter guidance, how are you thinking about achieving that target by the December quarter? Is there the possibility that you can achieve that a bit earlier and what are the percentage to your expectations there?

And then I have a follow-up.

Yeah. Let me take that one. This is Michael Hurlston. So we did outline that we’re still very much on track for the $500 million by the end of the year. We’re obviously not guiding, we’re guiding a quarter at a time, but we feel pretty pleased about where the company is setting up, and still, as we said in the remarks, are very much on track for the $500 million quarter.

Did you have a follow-up, Joe? And then maybe just a…

Yes. Yes. Definitely have a follow-up. So maybe just talk, maybe just touching on the Datacom chip business.

I think you were initially targeting a 40% increase in volumes by the June quarter. I guess just wanted to get an update there in terms of how it’s tracking relative to your initial outlook and whether you’re outperforming your initial expectations, both on the volume and ASP front. And similar question, but obviously more related to kind of the calendar 2025 outlook, it does look like you’re kind of outperforming your initial expectation in terms of where you were going to track towards the end of the calendar year, but any kind of color in terms of how you’re thinking about ASPs trending there, and then sorry for the long question, but just curious, what’s the contribution that you’re expecting from CW lasers now that you’re talking about ranking those up, as well as EML?

Thanks.

Yeah. I’m going to take the first part of the call -- question and then turn it over to Wupen. So what I would say, look, we’re obviously tracking very well. We’ve increased our capacity in the indium phosphide fabs.

We feel like demand is still outstripping supply on our EMLs. So we -- even as we ramp supply, we’re still very much sold out. That being said, as we increase the capacity, we feel like there’s room to add CW into the mix and the reason for that is it really extends our serviceable market. So we feel like we want to do that to get a foot in that door.

Our design we think is differentiated and gives us an opportunity to meaningfully participate in CW, but the predominant focus is and remains EMLs. From an ASP standpoint, again, given the supply constraints, obviously, there’s opportunities there to continue to maximize our price and we’re looking at that. We have opportunities as we think about now transitioning from 100-gig to 200-gig lasers, and obviously, going up to 400-gig. But maybe, Wupen, you want to add some color on the tech and how we think about it?

Yeah. Yeah. So I’ll just comment a little bit on the CW laser ramp [ph]. We’re at the very beginning, right?

We just won a couple of customers that CW lasers, as Michael pointed out, plus our fab capacity continue to be ramping up, we would like to be able to play in that market, which increased our TAM. So we’re at the very initial phase of the ramping, not huge numbers yet, but we will continue to see that volume going up as time goes by.

Thanks, Joe.

Thanks, guys. Appreciate it.

Our next question is from Simon Leopold with Raymond James. Your line is now open.

Thank you very much. I wanted to see if you could maybe unpack what’s in your assumptions around the tariff headwind. You quantified it around 100 basis points, which we appreciate. I guess what I’m trying to get at is trying to get a sense of how much business or manufacturing do you still have in China and that assumption.

I understand you’re moving more to Thailand and what are the other underlying assumptions you’ve made? Then I’ve got a quick follow-up.

Yeah. Sure, Simon, I’ll start off and then certainly, Wupen and Michael can jump in. Yeah, so like you mentioned in our prepared remarks, we did talk about the 100-basis-point headwind to gross margins, and therefore, operating margins during the quarter. A lot of that headwind is coming from increased component costs, in-feed costs, where we are having to pay the tariff associated with sub-mount products as well as capacitors.

And so wherever we see that, we’re having to pay a tariff associated with that. For most of our customers, we are trying to move our production from China into our Nava facility in Thailand. We’re starting off with our data center interconnect products, but we are eventually going to also be moving products from our Dongguan facility for cloud modules into Nava as well. So what we’ve baked into the forecast right now is our current production ramp for the quarter, but we’ll see improvements in that in our fiscal Q1 and Q2 as more of our Dongguan production moves to our Nava facility for cloud modules.

So certainly the impact from that perspective will diminish.

Thanks. And then just as my follow-up, longer-term question is, I’d like to get your latest thinking on what contributions and what role you can play in CPO, right? It’s nice to see that you were announced within the ecosystem back in March. I assume you’re providing products like CW lasers, but we can get an idea of maybe some dollar content and forecasting of how you see this affecting your business over the intermediate or longer term?

Thank you.

Hey, Simon. This is Wupen speaking. I think the CPO, frankly speaking, there’s still some times away, right? We talk about ramp volume in second half of the next calendar year, number one.

So therefore, in the meantime, the module is going to continue to be a ramping business for us. That’s going to be a foreseeable strong growing business for us and for the industry to come. On the other hand, in addition to the laser being an important component in the CPO ecosystem, we’re also looking at adding other components in the longer term. There’ll be other optical chips that’ll be needed in the CPO world in the longer term.

That’s very much what we’re actually looking into as well. But I would say that’s really a multiyear prospect. In the near-term, it’s going to be the laser components plus the cloud module ramping in the cloud interconnect world.

Thank you.

Thank you so much, Simon.

Our next question is from Blayne Curtis of Jefferies. Your line is now open.

Hi. Ezra Weener on for Blayne. Thanks for taking my question. Just a quick one and then probably a follow-up. Just look at the transceiver business. 50% plus growth for next quarter is great and clearly a lot of traction. Can you talk about what you’re seeing there from an EML versus CW laser perspective and the plan for insourcing versus externally sourcing there?

Yeah. Near-term, all of our products are using CW lasers. So we’re using silicon photonics approach for all of that, including all the revenue growth that you just cited. And all of it actually is externally sourced.

So we have not yet incorporated our own lasers in our transceivers, but we would expect to do that sometime early in the calendar year next year. That’s kind of the plan. We obviously have a number of big ramps in front of us with multiple customers that we talked about, but we -- our next phase of growth will incorporate our internal lasers. Wupen anything to add?

Okay. Good.

Thanks for the question.

Just a quick follow-up in terms of…

Ezra, do you have a follow-up.

Yeah. Thanks. So that would be three customers all using CW lasers, just kind of wondering what that means for the industry in terms of what you’re seeing from people choosing CW or EML?

Yeah. I mean, obviously, as we talked about a second ago, our EML are very much sold out. In fact, it really is a chase for us to keep up with the demand. We are basically choosing to allocate our supply toward EML; one, because of differentiation; two, because of pricing.

So we’ve really focused our capacity as much as we possibly can on the EML product. As we said, we’re really increasing the capacity quarter-on-quarter, year-on-year. And as we sort of hit a space where we think we have a bit of room, we think there’s an opportunity now to reallocate just a bit towards CW so we can start participating in that market first externally, right? We’ll source those to third-party transceiver customers, but then strategically, to your point, we will begin insourcing those.

So it’ll be mostly CW that we end up insourcing, and that’s why strategically it is important for us to develop that muscle.

Awesome. Very helpful.

Thanks, Ezra.

Thank you.

Our next question is from Meta Marshall with Morgan Stanley. Your line is now open.

Great. Thanks. Would you maybe just some clarity on the margin improvement that you saw on the Cloud & Networking business? How much of that is from some of the yield improvements on Cloud Light versus just kind of pricing mixed advantages on EMLs? And then I have a second question.

Yeah. Sure, Meta. I mean, we’re seeing across all the product lines a nudge of improvement, I would say. On our data center interconnect products, we are moving a lot of that production from CMs to our own internal manufacturing in Thailand.

And so the margin dollars that used to accrue to them are now accruing to us. So that’s certainly helping us both in Q3 and more substantially in Q4. And actually, we expect to increase that capacity moving into the back half of the calendar year. So that will continue to benefit us. On the Datacom chip side, Michael and Wupen both mentioned earlier that demand is ahead of supply.

And so as that production has improved year-over-year and sequentially, that’s also providing a tailwind for us. To your earlier question on modules -- to the first part of your question on modules, yes, yields have improved. And so the yield issue that we had in our December quarter, which caused a bit of a headwind on our gross margin line, that situation has improved into Q3 and that certainly helped us. And into Q4, that’s also helping us, especially as revenues are expected to increase 50% sequentially, the benefit of improved yields are accruing to our gross margin.

So I’d say kind of across the Board, we’re seeing a nudge of improvement across multiple product lines and that’s helping us.

Meta, I mean, just from a perspective, we said this in the prepared remarks that we’re, sorry, Meta, we’re going to continue to focus on gross margin. I think there’s opportunity on pricing and on our cost line to continue to improve that. So we’d expect a pretty nice ramp toward the 40% target that Wajid outlined at OFC.

Okay. Got it. And just as a second question…

Thank you.

…obviously a very fluid situation with tariffs currently, but assuming kind of the 100-basis-point headwind that you guys are talking about is within that kind of 90-day exemption period, kind of considering current tariffs as of that and just kind of wondering what your thinking is, if there are kind of greater tariffs that come from Thailand just or thoughts about eventually moving production out of Thailand, it sounds like you’re moving more production there for now, but just kind of overall thinking if there ends up being tariffs on Thailand? Thanks.

Yeah. I mean, obviously, if there are tariffs on Thailand that will have a headwind that we’ll have to work through. Yes, we are moving more of our production to Thailand, whether it’s from our CMs to us, or whether it’s from our cloud module business in Dongguan, China to us in Thailand. So we are making that manufacturing shift that’s been in place for a number of quarters now and so I think from an infrastructure and strategy standpoint, that certainly is there.

If there’s tariffs that accrue post kind of July 8th because of that, then we’ll have to work through that with our customers. And as you can appreciate, as there’s been noise on tariffs, we’ve been speaking to our customers about how we work through the pricing impact to them that is associated with those tariffs. But up until now, for the large part, most of our shipments that come to the U.S., we are not the importer of record. And because we are not the importer of record, those conversations have been a lot simpler, I’d say.

So that’s really how we’re thinking about it.

Great. Thanks.

Thank you, Meta.

Our next question is from Christopher Rolland with SIG. Your line is now open.

Hi. This is Dylan Ollivier on for Chris. Thanks for taking my question. So for my first question, I wanted to follow up on the 50% sequential growth in the optical transceiver business.

So which of your transceiver customer programs will be responsible for that growth and how should we expect transceiver shipments to trend for the rest of the calendar year? Do you guys think that level is sustainable? And are you guys expecting to grow from there as you add customer programs? Thank you.

Yeah. On the transceivers, I mean, I don’t want to necessarily outline specific customers that are driving it, but it’s predominantly the customer that we’ve had shipping for quite some time will be most of the contribution. We would expect this business to continue to grow. I think we’re really well positioned.

Obviously, generally speaking, our share is relatively modest. We think that we’ve put some of the issues behind us relative to some issues transferring products into manufacturing and getting those into production. And we would expect from here to continue to grow. Will we grow at this growth rate?

Hard to say, but certainly we would expect to continue to grow this business and it’ll be a major growth driver for us in 2026. Wupen, any comments from you?

Yeah. And also, as Michael said, as the more programs start ramping, we will see this to be more broader based than this one major customer ramp. Other programs will be gradually brought into the ramping phase as well. So we’re going to see that going forward.

Did you have any follow-up?

Got it. That’s great. Yes. So for my follow-up, I wanted to ask about 200-gig EMLs. So it sounds like you’re getting solid traction there already. When should we expect that to ramp and do you expect it to quickly cannibalize 100-gig EMLs or just to layer on top initially?

Yeah. This is Wupen here speaking. I think that timing-wise, it’s going to be toward the end of this calendar year. That’s when 200-gig EML will start to ramp.

And we don’t see this being a cannibalization of 100-gig EMLs, because there’s still a very large, a growing demand for 100-gig EML for 800-gig modules. 1.6-gig modules or solutions will be on top of the 100-gig demand for quite a while to come. So this should be additional opportunities than a cannibalization opportunity.

Thank you.

Great. Thank you.

Our next question is from Ryan Koontz with Needham & Co. Your line is now open.

Thanks for the question. Maybe stepping back a little bit on the Datacom transceiver market here relative to the strong Chinese players. What are you hearing from your customers, your major cloud-oriented customers about their interest in moving away from Chinese suppliers and interest in bringing those products in-house? Thanks.

Yeah. No. There’s definitely an interest to move away from Chinese supply. I mean, I think the problem right now is that collectively we can’t meet the demand.

So they are going to continue, I think, to buy from Chinese suppliers until U.S. suppliers can ramp to a level to meet what looks like a relatively insatiable demand. Generally speaking, there is some design insourcing going on. Some of the customers are doing their own designs, but the outsourcing is by far the dominant pole. Working with third-party transceiver manufacturers and we obviously are a small player at the moment.

We expect, given the number of designs that we have in flight, we expect to be a more meaningful participant in the market starting next quarter, but then really growing into the quarters beyond that.

That’s great. Thank you for that.

Ryan, do you have a follow-up.

And then maybe -- I do. Yeah. Thanks. On the telecom side of the business.

Yeah, what kind of big shifts are you seeing there? Obviously, DCI is driving the boat there in terms of growth. As you think about that makeshift, does it kind of enlarge your expectations? How should investors think about the shift toward DCI and how it affects your business, whether it’s measured on gross profit or revenue per unit, these sorts of things, but wallet share, but you can just step back and explain how investors should think about these shifts in telecom, would be helpful?

Thank you.

Yeah. I mean, I think that the traditional telecom business targeted to operators and target to MSOs is still obviously meaningful, but the long-haul now is the hyperscalers. So DCI, very important. Even these guys are now participating in our long-haul business.

So we talked in the call and the prepared remarks about pump lasers. We’ve seen really an increase in that business. A lot of the backbone now is being supplied by the hyperscalers and we’re participating in that with our traditional telecom products, but now more targeted toward the hyperscalers. So that’s been a surprising growth driver.

I think, quite frankly, we haven’t seen this business perform at these levels in quite some time. So it’s really a nice surprise. I mean, technology-wise, Wupen, any other things that you’re seeing from a trend line?

Yeah. Certainly. I think the, overall, we should think about the AI driving force is not just about data center internal. Data has to be transmitted across data centers.

So all these actually have the same driving source, right? We see the VR being growing very strong and therefore, tunable lasers doing really well. Even the long distance transmission coherent components are doing very well because of the terrestrial long-haul requirements and also the pump lasers for fiber capacity. So, all in all, we’re seeing a very strong DCI kind of traffic growth and that’s driving our so-called telecom part of the portfolio as well.

I’ll just add an additional comment, Ryan, around your question on gross profit dollars. So DCI…

Yeah.

… products are accretive to us from a gross margin standpoint. Not only where we’re currently operating from a gross margin standpoint, but also versus the 40% target that Michael laid out earlier, it’s even accretive to that. And so the combination of increased demand and our increased capacity internally at Thailand is certainly a tailwind for us that is driving a lot of the confidence that you’re hearing from the team around our gross margin abilities is coming from the secular demand along with internal manufacturing, along with the pricing power that we believe we have on our data center interconnect products.

So it’s all adding up together.

Thank you, Ryan.

We really appreciate it.

Our next question is from Karl Ackerman with BNP Paribas. Your line is now open.

Yes. Thank you. I have two. I believe you said that your cloud transceiver revenues would grow 50% sequentially on a contribution from all three of your cloud customers.

I have to ask, though, is this tied to -- it sounds like it’s tied to a new GPU or ASIC program, ramping at least at one of them, but I guess, how much of this, if at all, is from customers pulling forward transceiver demand ahead of potential tariffs? And if not, how do you manage that supply and demand balance with these three customers? And I have a follow-up, please.

Yeah. Hey, Karl. Michael Hurlston. So, first of all, like, we don’t see any pull-forward.

I mean, it’s been a race for us to keep up. We kind of alluded to this in one of the previous questions. We definitely had an issue ramping. We went through a ramp phase and then a qualification phase.

And really this is a product that’s driving a lot of the increase that was something that’s been in our hopper now for a couple of quarters. So we certainly don’t see a pull-forward effect. And again, with that layering in with the additional two customers beginning to ramp this quarter, we feel -- we don’t see any effect on pull ahead. I’m going to have Wupen talk a little bit about what those are tied to relative to kind of the customer tech.

Yeah. I think from the product release point of view, definitely I think those ramps are tied to the dominant GPU ramp and also some other accelerator ramp. There’s some loose association with those kinds of introduction of the accelerator products. But also, though, I think, the demand is pretty broad based.

It’s not specifically only to those two or three accelerators that are being introduced now into the market.

Karl, did you have a follow up?

I did one, if I may. Wajid, you spoke of moving production away from contract manufacturers to in-house solutions. I’m curious, is that for your Industrial Tech products or does that relate more toward Cloud & Networking? And similarly, do you need to add substantial capacity to move that production in-house, please? Thank you.

Yeah. So it’s revolved around our data center interconnect products and so that -- those CapEx decisions were made six months to nine months ago, and actually the CapEx is showing up this fiscal quarter and it’s expected to improve our capacity for data center interconnect products by 50% exiting this calendar year. And so that is the production that instead of providing to our contract manufacturers, we’re still keeping them going at full speed, but the incremental supply is coming from our Nava facility for CapEx that is being put in place this quarter and that will fully ramp by the end of this calendar year and give us another 50% more capacity on data center interconnect products, specifically tunables.

Thanks, Karl.

Our next question is from Mike Genovese with Rosenblatt Securities. Your line is now open.

Great. Thanks. I’m not sure if I missed this in the prepared remarks, but did we get an update on telecom performance in the quarter and the sort of product issue last quarter with just getting supply, if that has been solved yet?

Yeah. Sure. So I’ll start off and I’ll let Wupen also jump in. So telecom performance came in exactly as expected.

We had shortages on pumps. We had shortages on tunables. We had shortages on CDM products from our classic transmission product lines. And so all of those products, we were working through supply constraints throughout the quarter.

We are expected to see improvements in supply for all three product lines moving into Q4 and into the back half of the year. The earlier question was around our tunables products where we mentioned that we have CapEx that we’re putting in place in our Thailand facility to increase the production capacity by 50% over the remaining part of this calendar year. So with that increased capacity, we’re going to see the telecom part of our business improve as we move through the quarters for the rest of this calendar year. And maybe you want to talk about pumps a little bit more, Wupen.

Yeah. I think the, overall, the -- from the demand and supply perspective, we don’t expect that our capacity can really fulfill all the demand for the next several quarters, because of the demands being really strong. Although we’re doing better than we were able to. So we’ll continue to run capacity, continue to get more supply into the picture.

But we’re going to be in a pretty tight supply situation for next -- at least the next couple of quarters.

Did you have a follow-up, Mike?

Great. Yeah. I guess I want to revisit the question whether there’s any caution in the guidance for the back half of the year related to macro or tariffs. If I just make the transceiver comment of 50% sequentially, I mean, to me, that gets us $30 million more revenue.

That gets us the whole growth in the guide. I know Industrial is supposed to be down, but you’ve got EMLs growing, you’ve got telecom that should be up. Is there an extra dozen conservatism in here or am I overthinking this?

I’d say you’re not overthinking it, Mike. I mean, obviously, we’re in an environment where there’s a lot of uncertainty. We’re quite confident in the guide that we’ve provided. But if you just listen to some of the capacity improvements that we’ve got, there is certainly opportunity to deal with any type of headwinds that come our way during the fiscal quarter.

As far as the back half of the calendar year looks, across all of our product lines, we continue to see an uptick in demand, whether that’s Datacom chips, whether that’s data center interconnect products, or whether that’s modules. Even our Industrial lasers product line that is going to -- that is expected to come down in fiscal Q4 has new ultrafast products that are ramping in the back half of the calendar year. Demand for those products is quite strong. But offsetting that, we’re seeing some headwinds from customers that take kilowatt products.

But overall, the trend is up and our supply plan is up. So we should see improvements from here to the $500 million target that Michael spoke about. But again, I think the macro uncertainty, none of us have a great crystal ball on that.

Thanks so much.

Thanks, Mike.

Our next question is from David Vogt with UBS. Your line is now open.

Great. Thanks guys for taking my question. I just want to go back to the telecom business for a second. I jumped on late and I apologize.

Was there anything in the quarter that was similar to last quarter? What I mean by that, there was some channel normalization last quarter and even in the quarter before, just trying to get a sense for what’s core underlying demand, particularly in the face of tariffs or the risk of tariffs or the potential tariffs. And I’ll give you my follow-up. And I might’ve missed this because I jumped on late.

Did you share with us sort of how the core transceiver business grew in the March quarter relative to either last year or last quarter? Thanks.

Yeah. Maybe, I mean, look, on the telecom business, obviously, we feel pretty good. I mean, I think, there’s a lot of opportunities for us to continue to grow. Wupen talked about the narrow line with lasers.

We actually said that in the prepared remarks. So we think that business continues to grow through the balance of the calendar year and into next year. And it’s not just that. I think the -- we talked about pump lasers.

That really has been a strong product area for us. So across the Board on telecom, which as Wajid said correctly, is a gross margin enhancer. It’s a creative on the gross margin line. We would expect to see that to continue to creep up.

Large portion of growth will be cloud modules and so as you think about us across the next couple of quarters, in order to keep the gross margins up, we are going to need to see improvements in telecom, because we think that that’s an area for us that has a creative gross margins. On the Datacom transceivers, the second part of your question, essentially that was flat on quarter. We didn’t see a lot of growth. I think the growth for us, we talked about it, I think in previous discussions that we’d hit a bit of an air pocket relative to execution on ramping some of our Datacom transceivers.

We’re past that at this point. We would expect to see the growth in this subsequent quarter in the guide. And then we expect to see pickup from there. So on balance, look, we feel really good about the way the business is shaping up.

Wajid gave correctly a right amount of prudence around tariffs and macroeconomic uncertainty. But if we don’t have that, we feel really good about where all of our business is landing right now.

Great. Thanks guys.

David, did you have a follow-up?

No. Those are my two. Thank you.

Okay. Excellent. Thank you. Jasan, I think we have time for just one more question.

Okay. Our last question is from Alek Valero with Loop Capital. Your line is now open.

Hey, guys. Thank you for taking my questions. This is Alek on for Ananda. So I have a question on co-packaged optics. What’s your view on when co-packaged optics truly have a chance to gain traction in the data center and provide competition to transceivers?

Yeah. This is Wupen speaking. I think the co-packaged optics would take some more time. The technology is pretty complex.

A lot of engineering challenges, I think, yet to be worked through. As the leading AI customer talk about in their conferences, this year there will be some, a first system release. And next year, 2026, there’ll be another system release based on the Ethernet switch. We believe that’s the timing we’re going to see some level of adoption of the CPO optics.

However, though, we do believe that the plug-in optic [ph] will continue to be the dominant form factor going forward. I think it will take the next several years, really, to gradually run the CPO optics to be a meaningful share. And the benefit of CPO is clear, but the adoption will take time for people to figure out how to use it very effectively.

Got it. That’s very helpful. And just as a quick follow-up, can you remind us the economic difference between transceivers and co-packaged optics?

The difference?

The difference between transceivers.

Oh! So, basically, if you take any kind of a switch box, for example, the transceiver basically is at the front panels so you have the connection between the DSP, sorry, the switch chip in the signal all the way to the front panel of the box, which can be 20 inches, 30 inches actually away from the switch chip. The co-packaged optics basically remove all that, put the optics directly onto the switch chip, thereby removing all the signal losses and the complexities associated with that transition. That, in essence, is what a CPO optics does.

And then that lowers the power consumption, reduce the cost, reduce the complexity. But in a sense, you actually move the complexity from optical modules onto the switch chip, which then is some engineering work to be done to make sure that it’s a reliable solution.

Got it.

Thank you, Alek.

And -- but the co-packaged optics...

Oh!

Thank you.

Go ahead.

Yeah. No. I was going to ask, does co-packaged optics come with an ASP increase compared to transceivers?

Well, it’s a different pricing model. Today, the consumption model is the switch chip is purchased separately and the box is made by somebody else and the modules are purchased through third-party manufacturers such as us. And it’s a very different consumption model than the CPO optics, which basically means that the switch vendor will control the entire ASIC plus the optics. So it’s very hard to say how they’re going to price it.

And therefore, how to say that which one’s going to be cheaper really depends on the competition and benefit and so on and so forth. Very, very hard to give an estimation at this point.

Thanks, Alek.

Got it. Thank you again.

Thank you, everyone, for your questions. That’s all the time we have for questions. We look forward to connecting with all of you at upcoming investor conferences and meetings this quarter. And with that, I’d like to thank you for joining us today.

That concludes the conference call. Thank you for your participation. Enjoy the rest of your day.