Q4 2023 Lumentum Holdings Inc Earnings Call
Participants
Alan S. Lowe; President, CEO & Director; Lumentum Holdings Inc.
Christopher W. Coldren; Senior VP and Chief Strategy & Corporate Development Officer; Lumentum Holdings Inc.
Kathryn Ta; VP of IR; Lumentum Holdings Inc.
Wajid Ali; Executive VP & CFO; Lumentum Holdings Inc.
Alexander Henderson; Senior Analyst; Needham & Company, LLC, Research Division
Ananda Prosad Baruah; MD; Loop Capital Markets LLC, Research Division
David Vogt; Analyst; UBS Investment Bank, Research Division
George Charles Notter; MD & Equity Research Analyst; Jefferies LLC, Research Division
Jeffrey Robert Koche; Senior Research Associate; Raymond James & Associates, Inc., Research Division
Meta A. Marshall; VP; Morgan Stanley, Research Division
Michael Edward Genovese; Senior Comm and Cloud Infrastructure Analyst; Rosenblatt Securities Inc., Research Division
Samik Chatterjee; Analyst; JPMorgan Chase & Co, Research Division
Thomas James O'Malley; Research Analyst; Barclays Bank PLC, Research Division
Vivek Arya; MD in Equity Research & Senior Semiconductor Analyst; BofA Securities, Research Division
Presentation
Operator
Good day, everyone, and welcome to the Lumentum Holdings Fiscal Fourth Quarter and Fiscal Year 2023 Earnings Call. (Operator Instructions) Please also, note today's event is being recorded for replay purposes. (Operator Instructions)
At this time, I would like to turn the conference call over to Kathy Ta, Vice President of Investor Relations. Ms. Ta, please go ahead.
Kathryn Ta
Thank you, and welcome to Lumentum's Fiscal Fourth Quarter 2023 Earnings Call. This is Kathy Ta, Lumentum's Vice President of Investor Relations. Joining me today are Alan Lowe, President and Chief Executive Officer; Wajid Ali, Chief Financial Officer; and Chris Coldren, Senior Vice President and Chief Strategy and Corporate Development Officer.
Today's call will include forward-looking statements, including statements regarding our expectations and beliefs regarding synergies of recent acquisitions, including NeoPhotonics; financial and operating results, macroeconomic trends; trends and expectations for our products and technology, our end markets, market opportunities and customers; and our expected financial 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 the quarterly report on Form 10-Q for the quarter ended April 1, 2023, and those in the 10-K for the fiscal year ended July 1, 2023, to be filed by Lumentum with the SEC.
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 a substitute for or superior to financials prepared in accordance with GAAP. Lumentum's press release with the fiscal fourth quarter and full year 2023 results and accompanying supplemental slides are available on our website at www.lumentum.com under the Investors section.
With that, I'll turn the call over to Alan.
Alan S. Lowe
Thank you, Kathy, and good morning, everyone. We are extremely optimistic about the long-term secular demand drivers in the markets in which we participate and lead. Additionally, our efforts and investments to grow our share in our existing markets and adjacent markets are generating positive traction that will benefit us for years to come. Our focus on technology and product leadership will ensure our growth and differentiation in support of our customers' needs now and into the future.
In the near term, we are facing significant headwinds as both our direct and end customers actively work to reduce their elevated inventory levels. We believe that the current customer inventory correction cycle will continue through the balance of the calendar year, and therefore, our shipments will be well below end-market demand. Even at these depressed shipment levels, we believe we are continuing to grow our market share outside of the consumer market.
We are seeing meaningful demand strengthening for our Datacom chips as hyperscale customers prepare to ramp AI capacity. We believe that our Telecom and Datacom revenue will be up in calendar '24 compared to calendar '23, as inventory levels should return to more appropriate levels at our customers and their customers.
Despite the inventory headwind we experienced in the second half of fiscal '23, our full year revenue was up 3% from fiscal '22. Also, fourth quarter revenue and EPS were both above the midpoints of our guidance ranges we announced last quarter.
Our acquisition integration is going very well. And in fact, we have completed our ERP consolidation and are now running on one company-wide ERP system. We are tracking ahead of our previously announced synergy plans, while we continue to deliver on our new product and technology road maps. At the same time, we continue to focus on our customers to drive an even stronger partnership and a differentiated level of satisfaction.
ROADM revenue was especially strong in Q4 with increased sequential shipments across all ROADM product categories. Also, our commercial lasers business grew sequentially, particularly in new applications of ultrafast lasers for the solar cell market.
As I indicated earlier, long-term demand trends for photonics products continue to be extremely favorable. Generational upgrades to C+ L-band as well as extended C and extended L-band architectures are underway in the backbone of networks where our transmission and transport products are highly differentiated and enabling for our customers.
We are designing products for the next generation of our customers' Photonics road maps, which will use 130 gigabaud and 200 gigabaud data rate coherent technologies. We are developing these high-speed products in both discrete and integrated form factors, paving the way for enhanced performance in metro and long-haul applications as well as new applications at the edge of the network.
With the addition of the teams from our NeoPhotonics and IPG acquisitions and the capabilities to develop DSPs and RFICs, we believe that our vertically integrated approach to these high-speed transmission products will give us the lowest product cost in the industry.
In addition, we demonstrated our coherent 800G ZR technology earlier this year, which we believe is the industry's first which will provide high-speed connectivity with extended reach for data center interconnect within metropolitan areas.
Turning to cloud data centers. As I indicated earlier, we are seeing increased customer activity for AI in the data center and expect this to translate to increased shipments of our chip level products for 800-gig transceivers. We have broadened our Datacom product portfolio with continuous wave or CW lasers for silicon photonic applications that connect server racks and AI clusters, which require higher data rates while consuming less power. Starting in fiscal Q1, we expect a return to sequential growth in our Datacom revenue.
The data center optical component market is projected to grow sharply over the next 4 to 5 years to accommodate the increased traffic associated with AI as customers employ ever higher bandwidth interconnects between racks, within racks and between servers and storage. We also believe that Datacom VCSEL growth will be meaningful in the next several years as copper is replaced by short-reach multimode optical links.
As stated earlier, we expect Telecom and Datacom revenue to be up in calendar '24 from calendar '23 as customers reduce their inventory levels of our products and our shipment rate is more in sync within market demand.
Before I provide additional detail on the fourth quarter results, I would like to address the topic of China's export controls placed on gallium and germanium. We have determined that our existing supply is sufficient for the medium term, and therefore, we expect that these controls will have little to no impact on our manufacturing output. We will continue to monitor the situation and work with our suppliers to source material outside of China to mitigate any long-term impacts of these controls.
Now let me turn to the fourth quarter and full year results. Telecom and Datacom revenue was down 2% sequentially but up 2% year-on-year. As expected, we saw sequentially lower shipments of tunable access modules in the quarter. As we expand our customer base and current customers complete near-term product transitions and reduce inventory levels, we expect this business to return to growth in fiscal '24.
The lower revenue in tunable access modules was partially offset by sequential increases in narrow line with tunable lasers and ROADM shipments across several leading customers. In fiscal '23, our tunable access module product line achieved new record revenues, growing 57% year-over-year with strength in metro access and fiber-deep applications. These products enabled cable MSOs and wireless network operators to improve network performance while avoiding the cost of replacing existing infrastructure.
In fiscal '23, we doubled our manufacturing capacity for tunable access modules in our wafer fab and our back-end assembly and test factories to address the anticipated growth in our shipments to these customers. Our ultra-narrow line with tunable lasers and our advanced ROADMs are key enablers of our customers' next-generation network architectures that are just starting to be deployed. We saw sequential growth in narrow line with tunable lasers and across all major categories of ROADMs, including low-port count, high-port count and contentionless MxN platforms.
Also, fiscal '23 ROADM revenue grew 22% from fiscal '22, driven by the adoption of these advanced ROADM architectures. Cloud data centers are being redesigned to support the high bandwidth requirements of AI workloads. These workloads require several times more bandwidth than traditional cloud computing.
At this early stage of AI hardware deployment, 800G transceivers can provide the bandwidth while also reducing latency. The new 800-gig transceivers utilized 8 different wavelengths at 100 gig per lane, triggering orders for our EML products and driving a return to growth for our EML product line.
Additionally, we are seeing strong demand for our high-power CW lasers for customers utilizing silicon photonics to build 800G transceivers. In calendar '24, our 200-gig per lane EMLs will enable the next generation of transceivers with capacity of up to 1.6 terabits. We expect to start ramping shipments of 200-gig EML products in calendar '24 and customer qualifications of 800G and 1.6-terabit transceiver designs are well underway. We expect our 200G per lane optics to be the workhorse of hyperscale data centers for years to come.
To further address the connectivity requirements for AI and machine learning clusters, we have been developing high-speed VCSELs for short-reach connections between servers and switches in these systems, and we expect to begin to ramp these shipments meaningfully in calendar '24.
In the longer term, we also expect to supply even higher-power CW lasers for leading AI hardware architectures to provide the high bandwidth, low-latency optical interconnects essential for training and inference applications.
Turning to Industrial and Consumer. Fiscal Q4 was down from Q3 and down year-over-year as expected due to smartphone seasonality and end-market demand. We continue to expect our fiscal '24 3D sensing revenue will be lower than that of fiscal '23 due to our assumption around 3D sensing end-market demand, pricing and an additional competitor on a certain socket, as discussed previously.
In the fourth quarter, Commercial lasers revenue was up 4% sequentially, but down 2% from the same quarter last year. Overall, fiscal '23 Commercial lasers revenue was up 8% from fiscal '22. We achieved a 35% sequential growth in ultrafast laser revenue and over 25% sequential growth in fiber lasers, which was partially offset by sequentially lower solid-state laser shipments primarily for semiconductor applications. Our growth in ultrafast lasers is being driven by new applications, particularly in solar cell processing.
We expect that as demand for these new types of applications grows, we will continue to gain share in ultrafast lasers. Based on our latest customer forecast, we expect overall Commercial lasers demand to be softer over the next several quarters due to customer inventory digestion and macro factors impacting end markets. We expect continued rapid growth in new applications for our ultrafast lasers to partially offset these near-term headwinds.
Although we expect our shipments in the near term to be soft, I'm very confident about Lumentum's mid- to long-term prospects given the current softness is primarily driven by high inventory levels, the fundamental end market and technology trends driving our growth expectations are strong and unchanged, and Lumentum is investing in R&D to capitalize upon the long-term growth drivers and is uniquely positioned to serve our customers at scale with financial and structural resilience built into our business model.
In the near term, we are focused on expense controls while maintaining crucial R&D to continue to drive the forefront of innovation as we partner with our customers.
Before turning it over to Wajid, I would like to thank our employees and our customers around the world for their focus and dedication as they continue to collaborate and partner with Lumentum as we execute upon our strategy.
With that, Wajid?
Wajid Ali
Thank you, Alan. Net revenue for the fourth quarter was $370.8 million, which was down 3% sequentially and down 12% year-on-year. As Alan mentioned, this revenue level was not unexpected and was driven primarily by the customer inventory digestion we have seen and expect to persist through the end of the calendar year.
During the quarter, we had 3 greater-than-10% customers, all in the telecom market with no 10% customers in the Consumer market. GAAP gross margin for the fourth quarter was 24.2%, GAAP operating loss was 15.1% and GAAP diluted net loss per share was $0.88.
Fourth quarter non-GAAP gross margin was 36.7%, which was down sequentially and year-on-year, primarily driven by product mix, factory underutilization and lower revenue. Fourth quarter non-GAAP operating margin was 9.1%, which decreased sequentially and year-on-year. Fourth quarter non-GAAP operating income was $33.7 million, and adjusted EBITDA was $59.4 million. Fourth quarter non-GAAP operating expenses totaled $102.4 million or 27.6% of revenue. Non-GAAP operating expenses were down $2.5 million from Q3 due to tight expense controls. Q4 non-GAAP SG&A expense was $40.7 million. Non-GAAP R&D expense was $61.7 million. Interest and other income was $13.3 million on a non-GAAP basis due to higher interest rates on our cash and investments. Fourth quarter non-GAAP net income was $40.2 million, and non-GAAP diluted net income per share was $0.59. Our fully diluted share count for the fourth quarter was 68.6 million shares on a non-GAAP basis. Our non-GAAP tax rate remains at 14.5%.
Turning to the full year results. Fiscal '23 net revenue was $1.77 billion, which was up 3.2% from fiscal '22. GAAP gross margin for fiscal '23 was 32.2%, GAAP operating loss was 6.5% and GAAP diluted net loss per share was $1.93. Full year fiscal '23 non-GAAP gross margin was 43.2%, which was down relative to fiscal '22. Fiscal year '23 non-GAAP operating margin was at 19.2%, down from fiscal '22. Fiscal '23 non-GAAP operating income was $339.2 million, and adjusted EBITDA was $431.7 million.
For fiscal '23, our fully diluted share count on a non-GAAP basis was 69.1 million shares. And non-GAAP net income was $315.3 million and non-GAAP diluted net income per share was $4.56.
On to the balance sheet. Cash and short-term investments increased $346 million sequentially to $2 billion, primarily driven by our convertible note offering. During fiscal '23, we generated $179.8 million in cash from operations, of which $49.2 million was generated in fiscal Q4.
During the quarter, we purchased 2.67 million shares for $139.8 million, which includes 2.34 million shares repurchased concurrent with the issuance of our 2029 convertible notes. As we make progress on the integration of NeoPhotonics products into our global manufacturing footprint and attain synergies without impacting customer deliveries, we plan to carry elevated inventories over the short term. However, we expect inventories to decline by approximately $30 million exiting calendar year '23 as we continue to focus on cash generation.
Turning to segment details. Fourth quarter Optical Communications segment revenue at $320.5 million decreased 4.4% sequentially and down 13.6% year-on-year. Optical Communications segment non-GAAP gross margin at 36.1% decreased sequentially and year-on-year.
Our fourth quarter Lasers segment revenue at $50.3 million was up 4.1% sequentially and down 1.8% year-on-year. Fourth quarter lasers non-GAAP gross margin of 40.6% was up sequentially, but down year-on-year.
Now let me move to our guidance for the first quarter of fiscal '24, which is on a non-GAAP basis and is based on our assumptions as of today. We expect net revenue for the first quarter of fiscal '24 to be in the range of $300 million to $325 million. Within this Q1 revenue forecast, we anticipate Telecom and Datacom and Commercial Lasers to be down sequentially, driven primarily by customer inventory dynamics we have discussed. We expect Industrial and Consumer to be approximately flat sequentially. Based on this, we project first quarter non-GAAP operating margin to be in the range of 1% to 4% and diluted net income per share to be in the range of $0.20 to $0.35.
Our non-GAAP EPS guidance for the first quarter is based on a non-GAAP annual effective tax rate of 14.5%. These projections also assume an approximate share count of 67 million shares.
In terms of expectations beyond Q1, as Alan mentioned, we do expect a return to growth in Telecom and Datacom shipments in calendar '24 compared to calendar '23, as customer inventory levels are reduced and our shipment rate is more in sync with end-market demand.
Our synergy plan that we communicated at our March investor event at OFC is proceeding ahead of schedule in terms of operating expense reductions. We will exit certain manufacturing facilities at the end of this calendar year, which will deliver significant cost of goods sold synergies over the subsequent quarters. Overall, we remain on track to the total synergy plan of $80 million in annualized savings that we articulated previously, and we have achieved over half of the savings in fiscal year '23.
With that, I'll turn the call back to Kathy to start the Q&A session. Kathy?
Kathryn Ta
Thank you, Wajid. (Operator Instructions). Now let's begin the Q&A session.
Question and Answer Session
Operator
(Operator Instructions) Your first question comes from the line of Meta Marshall from Morgan Stanley. Again, Meta Marshall from Morgan Stanley.
Meta A. Marshall
Sorry. Just having the buttons day. Just in terms of how you guys are evaluating how much of this is inventory correction versus potential for share loss or just not being involved in certain platforms, just how are you kind of evaluating that given kind of repeated kind of step downs and maybe more particularly on the Telecom business, maybe as the first question.
Alan S. Lowe
Yes. Thanks, Meta. As we said in the remarks earlier, we believe we're not losing any share other than in the consumer space, where we've been talking about that share normalization and the third competitor.
I'd say in Telecom, we're very confident in our ability to continue to hold share, if not gain share, especially as we introduce new products. So I firmly believe that the slowdown in revenue is really, mostly inventory correction and has really nothing to do with share.
Meta A. Marshall
Great. I mean just as a follow-up, just how do you examine kind of what -- or get a sense of what the inventory positions are? And is kind of the belief that revenue would resume kind of in the revenue growth would resume in the calendar '24 period based on when people are telling you that they want to start to get the equipment again? Or is it a matter of that's when you think that budgets will open up from your customers and they're expecting. I guess just is that an expectation? Or is that backed by kind of conversations you're having with customers?
Alan S. Lowe
Yes, I wouldn't say it has to do with releasing the budgets. I think it's primarily due to the inventory levels on their balance sheets and their desire to lower those, given the availability of components and the availability of us to supply. So I would say that as we look at the next few quarters, Telecom is going to be tough through the balance of the calendar year. And I'd say that we're seeing some signs of inventory absorption faster than expected at some of the hyperscalers where we expected that to take longer. So I think AI is helping with that, both inside the data center as well as in the data center interconnect space.
So I think from that perspective, that's why we're confident about calendar '24 being higher than calendar '23. And those are really in-depth conversations with customer executives and really viewing our inventory in their locations and at their contract manufacturers. And that's why we believe that the next couple of quarters are going to take to get rid of that inventory.
Operator
Your next question comes from the line of David Vogt from UBS.
David Vogt
And just two for me. One, Alan, going back to the inventory question, I guess, can you kind of help us understand, you made a comment that you think you're undershipping to industry demand, where that demand might be today? And how you see that demand sort of progressing as we move through this sort of more challenging period of time in '23 into '24? That kind of colors your view about recovering back to growth in calendar '24.
And then maybe a longer-term question, I'll just give you both at the same time. When you think about mix of the business today, obviously, Consumer and Industrial is a lot smaller than it was last year. And presumably, that's a relatively strong gross margin business. How should we think about the operating leverage as growth recovers in Telecom and Datacom in calendar '24 from a margin perspective? Obviously, it's unlikely that you're going to get back to the high 40s gross margin, but I want to get a better sense for maybe gross margin trajectory as we move through the balance of this -- the next 4 quarters into maybe fiscal '25.
Alan S. Lowe
Yes. Thanks, David. I'd say on the undershipping question, I'd say that clearly, at the revenue levels that we're having today, our customers are shipping out more than we're shipping in. And I think that's really due to the fact that they've built up inventory over the last few years when there was fear that components would not be available. So I do believe that, that's the case.
That said, there are some North America carriers that I've talked about lowering their CapEx, but not significantly. So I'd say that the demand for bandwidth continues to be robust, and there's nothing that's going to slow that down.
I would say that, again, on the hyperscaler side, AI is really consuming a lot of the inventory that we had shipped over the last couple of years. And now we're starting to see signs that things could pick up before the end of the year in the hyperscalers. But in the normal carrier space, I'd say that's probably premature until calendar '24. Wajid, do you want to take the operating leverage question?
Wajid Ali
Yes, sure. So on operating leverage at the revenue levels we're currently seeing for the back half of this calendar year, you can appreciate that we're having a lot of underutilization charges within our internal factories due to the lower revenue levels and our desire to bring down our company inventory to a more normalized level.
As the back half of the fiscal year moves on, our expectation is that we'll really have 3 things working for us. One is, is that we're expecting to continue with our synergy plans that we've done that we've executed on quite well to date on NeoPhotonics. And the consolidation of the factories in the back half of this calendar year should start to show through the P&L in calendar '24.
The second thing is that our Datacom business is a chip business. And so we're expecting to see improved demand in that part of the business through calendar '24 and even actually in the back half of this calendar year as well. And so that should give us some uplift. And then like Alan talked about, once the telecom business becomes more normalized, the underutilization charges should reverse themselves, and we should see improvement. So those are kind of the 3 tailwinds from this point in time that can help us.
Operator
Your next question comes from the line of Alex Henderson from Needham.
Alexander Henderson
Great. wanted to start off with the comment on 3D sensing that the business would be flat into the September quarter. Historically, the June quarter is a seasonally softer quarter. The September and December quarters are seasonally stronger. Can you parse a little bit between, is that a late start because of some production issues? And therefore, we make up a little bit of the 3D sensing in the fourth quarter? Or is it truly a -- as steep a decline as the numbers would indicate? I think if it's flat sequentially, that's 60-some-odd percent decline year-over-year. So it's a pretty steep number.
Christopher W. Coldren
Alex, this is Chris. Thanks for the question. Yes, I think a couple of things go into our outlook. I think as we've alluded to on prior calls that we've anticipated the impact of having an additional competitor in the mix. And so that and current demand environment were both factored into our guidance, and that's what's impacting the sequential and year-over-year comps you're asking about.
Alexander Henderson
Yes. So again, the question is, is there a shift between September and December? Historically, September has actually been a little stronger than the December quarter. Is this a late start to some extent, and therefore, a little bit more in the December quarter? Or should we be using that as the new level of a seasonally strong period?
Christopher W. Coldren
I would say it's a little early to guide that product line for that time frame, but I would think you should think about it being flattish between the 2 quarters.
Alexander Henderson
Great. And then second question I had for you is, looking at the broader context of the trajectory of demand, I think we started back in the September time frame saying that we were going into an inventory correction at least in Datacom. Telecom obviously started later than that. But you had initially thought that by the end of the June quarter and into the end of the summer, Datacom would start to recover. Then you pushed it out to the end of the fourth quarter calendar. At this point, it sounds like that's holding, but it may be a little bit steeper initial decline. Can you talk a little bit about what's going on relative to the trajectory from the expectations that you gave last quarter to the expectations you gave this quarter on that business and within Telecom as well? Has it steepened as a result of excess cutbacks in shipments at the service providers? What's the linearity of the demand structure there? Did it fall off towards the end of the quarter?
Alan S. Lowe
Well, let me take them one at a time. I'd say on the Datacom market, as you said, in September, we talked about it getting better in the summer, and we're seeing that in a lot. And I'd say that, in fact, we're seeing it so much that we probably ratcheted back our capacity more than we should have on Datacom. That's now in full force to accelerate the output of our Datacom chips. And so I think that's why we talked about Q1 being up from last quarter, and then we expect to see sequential growth in our Datacom chip business through the balance of this year as well as into calendar '24. So Datacom, I think, is on the right trajectory.
On Telecom, I'd say that our customers are telling us that they want to bring down the inventory to normal levels. And I think the confidence that they have in our ability to produce what they need when they need it, and that the component suppliers, including semiconductors are going to be there when they need it gives them confidence that they can live with even less inventory than originally anticipated. So I'd say nothing really changed there other than inventory levels need to come down further. And that, therefore, we're shipping into our customers less than they're shipping out to the carriers, which in turn should take care of that problem and more normalize as we get into calendar '24.
Operator
Your next question comes from the line of Samik Chatterjee from JPMorgan.
Samik Chatterjee
I guess from -- first one, if I can just ask you to delve a bit more into the AI-related demand that you're seeing? How much of that are you seeing in terms of interest on the VCSEL side versus EML? And when you think about also the customer set there, how much of the engagement is hyperscalers versus other newer sort of companies coming into the ecosystem? Any thoughts in terms of what you expect that demand to look like in a couple of years would be useful. And I have a follow-up.
Alan S. Lowe
Yes, let me address that, and I'll ask Chris to add on. I'd say you're right, it's beyond the hyperscalers. And so we've collaborated with a lot of the new customers for us that are leading the way in AI and figuring out how do we then customize a product laser, high-power laser or external laser source to satisfy the future needs of AI in a unique way. So we are doing customization of our products to meet the needs of those unique situations where the data bandwidths are just huge.
And so I'd say that, that's really more of an impact in calendar '24, but that work has been going on now. And those products will come to market in the calendar '24 stage. I'd say that what we're seeing today is more EML-based-driven 800-gig transceivers that both use our CW high-power lasers for silicon photonics as well as 8 EMLs for each transceiver. Because those are the products that are available today that can use -- that can satisfy that bandwidth needs that the hyperscalers are going through. So that's what we're seeing.
As far as VCSELs are concerned, we're going through the qualification work on those 100-gig VCSELs I talked about, and we see that really starting in a meaningful way to contribute to our revenue in calendar '24.
Christopher W. Coldren
Yes. And I'd also add that in addition to the VCSELs, which are just ramping, and EMLs, we're a market leader. And we also -- there are CW lasers that power certain silicon photonic solutions. So we've really focused, as we've commented on prior calls, of broadening our product portfolio in Datacom to be able to address different parts of the data center and different approaches to the same parts of the data center.
And we do anticipate significant uptick in the market in Datacom in the coming 12 months relative to kind of the -- obviously, the down year that we have at being a chip supplier over the last year. And this is going to be a multiyear phenomenon that we definitely believe that -- you're talking tens of percent CAGRs for AI-related deployments over the next few years. So very exciting opportunity for us.
Samik Chatterjee
Got it. Got it. And for my follow-up, just rotating back to the results and the guide for the fiscal first quarter. I think with 3D sensing in the past, you've obviously had a certain seasonality to the business through a year. Any thoughts of how you're thinking about seasonality this year? Should investors expect, sort of as you get to beyond 1Q, to see sort of sustained sequential growth in the business? Or given some of your comments about the sort of calendar year being inventory digestion impacted, it's really more of a step-up into the second half of the fiscal year?
And I think what I'm trying to get to is we all understand the earnings power of the business at this run rate is depressed because of the lower demand or the inventory digestion. But in terms of revenue increasing, your OpEx savings coming through, what do you see as the normalized run rate for the business where you want to exit the year?
Alan S. Lowe
Well, normalized, it's hard to say. I'd say in the short term, the next couple of quarters, Telecom is going to be tough. And as we said, 3D sensing, you can think about being flat. Lasers is going to be down sequentially from Q1 -- in Q1, and you can think of that as being flat with fiber lasers going down in Q2, but being offset by our ultrafast lasers. So I think that then, the real mover between Q1 and Q2 is the Datacom demand, and that's really going to be gated by our ability to meet that -- supply that demand. And so I would say that there's some small uptick in Q2.
And then assuming that the inventory is taken care of over the next 5 months, then we should start seeing some pickup in the first half of calendar '24. And then normalized demand is -- happens when we're shipping into our customers what -- exactly what they're shipping out. And I think we're confident we can get back to the levels of revenue we've had in the past. And then with our model, the operating leverage is pretty immense. And we should see significant bounce back when we get up to that $400 million to $450 million, even $500 million in revenue. And I think that's doable in the not-too-distant future.
Operator
Your next question comes from the line of George Notter from Jefferies.
George Charles Notter
I wanted to ask about the gross margin performance in the quarter, down, gosh, roughly 400 basis points sequentially on a pretty similar revenue number. I guess I'm wondering if there's -- certainly, the lower 3D sensing is an element of that, I suppose. But I'm wondering if there's some excess and obsolete inventory charges there or anything else that drove that big sequential step down?
Wajid Ali
Yes. So you're right. So the 3D sensing number did come down into Q4. Excess and obsolescence was pretty normal from a run rate standpoint. It was really underutilization. So as -- we've been working through where we really want our inventory to be outside of some of the prebuilds that we're doing as part of our consolidation strategy. We had a pretty hefty impact within our fiscal Q4 with underutilization. That hit both our OpComm and our lasers business pretty significantly. And we're seeing that really flow through into the back half of the calendar year.
George Charles Notter
Got it. So is it fair to say that a good portion of the product that you shipped in the June quarter was then manufactured in prior quarters? Is that the way to read that?
Wajid Ali
Yes, that's fair. Yes, that's fair. That's right.
George Charles Notter
Okay. And then is there some strategy here to kind of, I don't know, I guess, produce a softer gross margin in the June quarter? Take that underutilization hit all at once then, and then you can kind of clean that effect up going into the September quarter? Or is it -- is there some other strategy in terms of how you run your utilization?
Wajid Ali
Yes. I mean, really, the only way to do it is to produce more, and so -- or consolidate facilities. So we are consolidating facilities, and we'll start to see the benefit of that probably in the January time frame because we'll be out of the facilities in the November, December time frame this year. So we'll start to see the impact of that in our fiscal Q3. So that will be pretty sizable. And then it's really production coming back up at all of our facilities to a more normalized level. So that's -- those are really kind of the only 2 ways that you can improve that.
Alan S. Lowe
Yes. Just to add to that, though, I'd say that the June quarter Datacom was below on Datacom, and those chips that are produced in our fabs absorb a lot. And as that ramps back up, we should see a better absorption and utilization of the fabs that we have in Japan, making Datacom chips.
Wajid Ali
Yes. That's fair.
George Charles Notter
Got it. Okay. And is that -- I was going to say...
Kathryn Ta
Please, go ahead.
George Charles Notter
Is that inventory -- let me follow on here. Is that inventory that you built up from a manufacturing perspective in the March quarter, is that inventory then consumed heading into the September period?
Wajid Ali
Well, no, not all of it because if you take a look at our inventory turns, it doesn't turn that fast. So yes, some of it will flow through into -- that already flowed through our fiscal Q4, and then some of it will flow through our Q1. A lot of it is very product level dependent. So like Alan said, in Sagamihara where our Datacom chips are produced, that inventory is turning very fast. So as the finished goods are coming out, it's being shipped. But at our Rose Orchard facility, where we have internal wafer fabrication, that's moving a little bit more slowly and as well as what we've got in Thailand, that's built up for lasers as well. So it is a little bit dependent. And because BOM costs are lower on Datacom chips, the impact at a consolidated level doesn't show up as well as it does when you take a look at by BU inventory turns.
Operator
Your next question comes from the line of Michael Genovese from Rosenblatt.
Michael Edward Genovese
I want to dig in on this guidance for Telecom and Datacom to be up for the year, I guess, my -- in fiscal '24. My first question is, is that more of a Datacom comment that Datacom will be way up? Or are we seeing that Telecom will also be up as well year-over-year?
Alan S. Lowe
Yes, Michael, I think what we said was that calendar '24 would be up from calendar '23 given that we believe the next 2 quarters of Telecom shipments are going to be depressed given the inventory reductions that are happening at our customers and at the end customers. So not fiscal year-over-year, I'd say calendar year, we're pretty confident that Telecom will be up in calendar '24 from calendar '23.
That said, again, I'd say Datacom is going to grow sequentially each quarter between now and the end of calendar '24.
Michael Edward Genovese
Okay. That makes sense. That's a calendar comment. And I guess my other question is, when I look at the Datacom 800G and above AI opportunity, you guys are in EMLs and it looks like they're trying to prioritizing VCSELs as well, I guess my question is, are there any other parts of the Datacom market now with this improvement? I'm not...
(technical difficulty)
Alan S. Lowe
Sorry, Michael, we couldn't hear your last question. I think the question though was around are there any other parts other than EMLs and VCSELs that are showing signs of demand growth? Is that your question? Or have we lost you?
Michael Edward Genovese
Yes. No, I'm sorry, here. I hope this isn't bad.
Operator
Mr. Genovese has disconnected. Is it okay to proceed to the next question?
Alan S. Lowe
Yes. Maybe we'll try to answer the question if we understood it right, and we can talk about different kinds of lasers, CW lasers and such, Chris?
Christopher W. Coldren
Yes, certainly, our focus is on serving a hyperscale cloud market, primarily to our transceiver customers that we supply to also certainly supply into the enterprise end markets. So again, supplying EMLs as the largest product line, and that's primarily playing now into the transition to 800 gig. And we will also have 200-gig per lane EMLs coming up in calendar '24 ramping, and that's for either a next generation of 800-gig transceivers and then the eventual transition to 1.6 terabits per second. We highlighted the VCSELs ramping so that we have a broader product portfolio as well as CW lasers to intersect the 800-gig and 1.6-terabit silicon-photonic-based approaches.
Alan S. Lowe
Yes. Just the only other thing is we're providing a series of different types of lasers to address the antenna band as well.
Operator
Your next question comes from the line of Ananda Baruah from Loop Capital.
Ananda Prosad Baruah
Yes, just a couple if I could as well. I mean, I guess, just sticking with Datacom chips and lasers. Is it -- I guess, I've been trying to frame for myself how meaningful this could become as a part of your business. And I think I'd just sort of clarify this. Correct me if I'm wrong. I believe that on the company's sort of prior, prior rev, let's say, [believes], 12 months ago before things really started -- before the hyperscales really started to work down their inventory, the run rates at which you were thinking could occur back then, which I think got to the kind of [2 40, 2 50, 2 60] level annually, would sort of suggest if you could still achieve those, if you have more of a tailwind on current run rates, you could begin to see sort of 15%, 16% of the company, once achieved, the beyond -- and I think those are primary EMLs.
Now you have what's going on gen AI related, you're doing some new VCSEL work. And I know a bunch of [cities] now at OFC but the CW laser were talking about InfiniBand laser work as well. Could we be in a situation in 8 quarters where like 20% of the company is gen AI Datacom chip laser related? And I just have a super quick follow-up after that. Let me get your thoughts there.
Alan S. Lowe
Yes. Good question. I think it depends on how fast the other parts of our business grows. But I think your numbers are not too far off. I'd say that we have gone down significantly over the last year from Datacom revenue standpoint, both from the standpoint of unit shipments or way down, and average prices have gone down in this past year. So we're in the midst of growing that.
I'd say that certainly within the next 8 quarters, could we get back to the kind of [2 40 to 2 60] annually? Yes, absolutely. And I think we're putting capacity in place to do that. We have capacity. As you remember, 3 years ago, we've been continuing to add capacity. We're in the midst of going to larger wafers to address the demand we've seen in the long term. So that certainly could be a significant growth driver from where we are today.
Ananda Prosad Baruah
Yes, Alan, that's really helpful. And I guess the quick follow-up is, like longer term, what's your view on sort of NeoPhotonics VR technologies' role in data center for AI related at all? And that's it for me.
Alan S. Lowe
Yes. We're very, very happy with the acquisition of NeoPhotonics and the technology and the team that came along with that. I'd say that there was a buildup of inventory of ZR and ZR modules at certain hyperscalers. I think that, that is being consumed, and we're starting to see signs of life, frankly, of hyperscalers needing more ZR modules. We're also providing a lot of the components to go into the world's ZR market. And so from that perspective, we're happy with the whole AI-driven data center demand. But also data center to data center, there's a lot of bandwidth going between them. So I don't know, Chris, any other thoughts?
Christopher W. Coldren
Yes. I mean I think that one of the thesis of the acquisition with regard to ZR was that we would provide a sort of stronger company, if you will, to give customers confidence. And that certainly is bearing out. Unfortunately, as Alan highlighted, that the acquisition closed at about the point where they also realized they had a lot of inventory. So now that inventory is beginning to clear at hyperscalers, we do expect ZR to be one of the products that recovers earlier in our Telecom portfolio.
And then with the level of vertical integration we have, customers are very excited about us as a supplier because they view us as being able to both evolve, obviously, from a cost and volume standpoint given our vertical integration, but also lead the transition to 800-gig and next-generation solutions that follow on beyond the current 400-gig products.
Operator
Your next question comes from the line of Tom O'Malley from Barclays.
Thomas James O'Malley
I just wanted to square just a couple of comments you've made. So I think that in response to Ananda's question, you talked about getting back to a greater than $200 million run rate in the Datacom business. But then on a question about Q2 in the fiscal year in December, you talked about just a moderate sequential increase just given the fact that you don't have the capacity and you said, "Hey, maybe in retrospect, we would have not taken down capacity as much."
So just to kind of put those 2 in perspective, if you look at this coming fiscal year, are you guys going to be able to get back to the levels that you saw in fiscal year '23? I think you started the year around $50 million. Is that something you'll be able to get to? Or is capacity going to hold you back really until the back half of calendar year '24, which would be your fiscal year '25?
Alan S. Lowe
Yes. I would say, it's certainly not a lack of demand that would get us there. So I'd say that in calendar '24, there's certainly enough demand. The question is, can we get the capacity back up, given that ASPs are down? So we have to actually produce a whole lot more chips to get to that $50 million run rate.
But that said, the gross margin on those chips are still very, very solid. So we're anxiously driving the team to grow the output. And you're right, we took down capacity more than we should have. But at the time, it was the right thing to try to drive the underutilization. But I'd say that the cycle time on Datacom chips is such that starting wafers today, it doesn't really impact the next 4 months. So it is a long cycle time. We're working on that as well. And that's why we're trying to dampen the expectations of rapid growth in the December quarter for Datacom, but we are expecting to see an uptick in Datacom in the December quarter.
Thomas James O'Malley
Got you. That's helpful. And then just on the technology side, and this may be just a broader question, but I just wanted to hear your take here, is if you look at the early days of AI here, there's obviously some drivers that are pointing more towards EML. But clearly, there's a lot of drivers that are porting more towards VCSEL. And you guys are looking to ramp that product.
I guess the question is, if you look at the market for lasers today in AI, how would you split out the percentage of lasers that are being used between VCSEL, CW and EML? And do you think that just given that maybe there's some more VCSELs early on, that you're missing out on some of that opportunity? Or if it's the other way around, just any color would be helpful there.
Christopher W. Coldren
Yes, I would say from the end-market opportunity and unit quantities, the VCSEL arrays and basically short reach and long reach, that's really what we're talking about, are not that different from a volume standpoint. So therefore, that's -- and being a leader in the EML, so that indicates why that VCSEL is such a good opportunity for us to grow in addition to just market growth to grow share.
In terms of CW -- CW and EML sort of compete with each other a little bit in that certain silicon photonic architectures are able to do the needed distance or reach that EMLs at the lower end do. So there is a little bit of cannibalization or zero sum between those two, but that's why we introduced and have both sets of products. Now that said, EMLs will lead the transition to the 200 gig per lane. And so we expect EMLs to really be the workhorse of the longer-reach AI links.
Alan S. Lowe
And just to note that while these units are about the same, the VCSEL cost that are priced are significantly lower than the EMLs.
Christopher W. Coldren
Yes. So the market is much larger for the longer-reach EMLs.
Operator
Your next question comes from the line of Vivek Arya from BFA Securities.
Vivek Arya
I was hoping you could help us quantify what percentage of your Q4 sales were related to AI? Were they all in Lasers or Telecom, Datacom segment? Just where do the AI-related sales show up? And how large were they in the quarter?
Christopher W. Coldren
Yes, I think the primary products going to AI would obviously be our Datacom business. We don't break that out, but it's sub 10% of company revenue. That said, we don't know when we ship an EML to a customer, whether that's going into an enterprise customer or a cloud AI customer at this point. But what we do expect, and as Alan has highlighted, that the growth we are seeing going forward is primarily AI-driven.
Vivek Arya
So sub 10% of your Datacom segment is AI or EML related? I just wanted to...
Christopher W. Coldren
No. No, I would say a very high percentage of our Datacom revenue is EML-related. It's just that the Datacom business is a smaller percentage of the overall company revenue given it's a chip-based business. So lower ASP, high margin but lower ASP.
Vivek Arya
All right. And for my follow-up, I'm just trying to think about the path back to growth, right? It seems for the overall company, it seems like Q2 will probably -- could be in the same range as Q1 and Q3 has in the past tended to have seasonal headwinds, but I don't know whether the mix now is different. So is it really fair to think that Q4 is the earliest we should be thinking about meaningful sequential growth? Or do you think seasonality and mix could play out differently this year?
Alan S. Lowe
Yes, I would say seasonality probably doesn't play as much as it has in the past. And when you look at the March quarter, just given how much less we're shipping into our customers today than they're shipping out. So I think it all comes down to their inventory levels. And when does that equalization happen, we think that it's going to take until at least December. But that said, we're going to see some strength in demand for new products. And I talked about the 130-gigabaud, 200-gigabaud type products as well as some new products we have in ROADMs and other, that should drive demand growth in Q1, regardless of inventory situation because they're new products, and they don't exist in the inventories today. So I'd say it just depends.
Operator
Our last question will come from the line of Jeff Koche from Raymond James.
Jeffrey Robert Koche
Jeff with Raymond James in for Simon. So one of your big competitors put out a forecast for the Datacom transceiver market basically calling for roughly 5 billion increase from 2023 through 2028. I just want to know what your thoughts on that forecast are. Do you think that's reasonable? And I guess your strategy and your milestones that you're targeting for that business would be helpful.
Christopher W. Coldren
Sure. I think that's not -- I'm not familiar with the exact numbers you referenced, but it is not inconsistent with our assumptions around growth in the overall Datacom market, particularly driven by AI.
Kathryn Ta
Simon, do you have a follow-up? Go ahead.
Jeffrey Robert Koche
Yes, yes. I'm just wondering -- so if I got this right that the -- did you say that NeoPhotonics was up or down sequentially in the quarter? And is it safe to say that the majority of the downtick in September and December is going to be in the legacy Telecom business? And that Neo is going to do -- is more, I guess, that inventory digestion with the ZR is more played out? Is that fair to say?
Alan S. Lowe
Jeff, the delineation between Neo products and legacy products after a year is pretty much gone. So we're still seeing strength, as we talked about in narrow line with tunable lasers. I think I said we're seeing growth again in expectations for ZRs as we move forward. But I'd say that Neo products in the June quarter were down from previous, and we expect, as the inventory of whether it's the narrow line with tunable lasers or other products that we got through the acquisition, as those burn down, we're going to see growth across the board. So I wouldn't say it's anything specific to legacy versus Neo products, it's an inventory issue at our customers that we're trying to take care of.
Kathryn Ta
Lara, I think we'll turn the call back over to Alan for some closing remarks.
Alan S. Lowe
Thanks, Kathy. Thank you, everyone. I'd like to leave everyone with a few thoughts as we wrap up the call. As I said before, mid- to long-term fundamentals are solid for our business as we serve the exponential growth in network bandwidth in the artificial intelligence, machine learning, mobile carrier and cloud computing markets. New industrial applications are emerging for our imaging and sensing products, and our commercial lasers are expanding into high-growth applications beyond our traditional markets. We are committed to investing deeply in innovation to deliver on our customers' needs today and in the future.
With that, I would like to thank everyone for attending, and we look forward to talking with you again at the investor conferences and upcoming meetings in the coming weeks. Thank you all for attending.
Operator
Thank you, sir. Ladies and gentlemen, this concludes your conference call for today. We thank you for participating and ask that you please disconnect your lines. Have a lovely day.
