管理層發言
Good afternoon. My name is Hillary, and I'll be your conference operator today. At this time, I would like to welcome everyone to the Viavi Solutions Fiscal Third Quarter 2026 Earnings Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Vibhuti Nayar, Head of Investor Relations. Please go ahead.
Thank you, Hillary. Good afternoon, everyone, and welcome to Viavi Solutions Fiscal Third Quarter 2026 Earnings Call. My name is Vibhuti Nayar, Head of Investor Relations for Viavi Solutions. And with me on today's call is Oleg Khaykin, our President and CEO; and Ilan Daskal, our CFO. Please note, this call will include forward-looking statements about the company's financial performance. These statements are subject to risks and uncertainties that could cause actual results to differ materially from our current expectations and estimations. We encourage you to review our most recent annual report and SEC filings, particularly the risk factors described in those filings. The forward-looking statements, including the guidance that we provide during this call, and our expectations regarding the end market and acquired business are valid only as of today. Viavi undertakes no obligation to update these statements. Please also note that unless we state otherwise, all results discussed on today's call, except revenue, are non-GAAP. We reconcile these non-GAAP results to our preliminary GAAP financials and discuss their usefulness and limitations in today's earnings release. The release as well as our supplemental earnings slides, which include historical financial tables, are available on Viavi's website at www.investor.viavisolutions.com. Finally, we are recording today's call and will make the recording available on our website by 4:30 p.m. Pacific Time this evening. With that, I would like to now turn the call over to Ilan. Ilan?
Thank you, Vibhuti. Good afternoon, everyone. Now I would like to review the results of the third quarter of fiscal year 2026. Net revenue for the quarter was $406.8 million, which is above the high end of our guidance range of $386 million and $400 million. Revenue was up 10.2% sequentially and on a year-over-year basis was up 42.8%. Operating margin for the third fiscal quarter was 21%, above the high end of our guidance range of 19.2% and 20.2%. Operating margin increased 170 basis points from the prior quarter and on a year-over-year basis was up 430 basis points. EPS at $0.27 was also above the high end of our guidance range of $0.22 to $0.24 and was up $0.05 sequentially. On a year-over-year basis, EPS was up $0.12. Moving on to our Q3 results by business segment. NSE revenue for the third fiscal quarter came in at $321.5 million, which is above the high end of our guidance range of $304 million and $316 million. Revenue from Spirent product lines was $54.2 million, which was in line with our expectations and included a few opportunities that were pushed out from the prior quarter. On a year-over-year basis, NSE revenue was up 54.4% primarily driven by the acquisition of Spirent product lines. We also saw strong demand for our lab, production and field products driven by the data center ecosystem as well as for our aerospace and defense products. NSE gross margin for the quarter was 65.3%, which is 220 basis points higher on a year-over-year basis and was primarily driven by higher volume and favorable product mix. NSE's operating margin for the quarter was 17.2%, an increase of 680 basis points on a year-over-year basis. NSE's operating margin was also above the high end of our guidance range of 15% to 16% as a result of higher fall-through. OSP revenue for the third fiscal quarter came in at $85.3 million, also above our guidance range of $82 million to $84 million. On a year-over-year basis, OSP revenue was up 11.4%, primarily driven by strong demand for 3D sensing and anticounterfeiting and other products. OSP gross margin was 50.3%, down 130 basis points on a year-over-year basis, and it was mainly due to unfavorable product mix. OSP's operating margin was 35.3%, an increase of 140 basis points on a year-over-year basis. OSP's operating margin was in line with our guidance range of 34.8% to 35.8%. Moving on to the balance sheet and cash flow. Total cash and short-term investments at the end of Q3 were $508 million compared to $772.1 million in the second quarter of fiscal 2026. Cash flow from operating activities for the quarter was a use of $26.3 million versus $7.8 million that we generated in the same period last year. The cash flow was mainly impacted by the earn-out payments to Inertial Labs, timing of working capital and employee variable costs. CapEx for the quarter was $5.9 million versus $6.8 million in the same period last year. During the quarter, we successfully paid $49 million in cash for the remaining principal of the convertible notes due in March 2026, and we issued about 1.8 million shares for the conversion premium above par. We also prepaid during the quarter $150 million of the Term Loan B. We currently have $450 million remaining for that loan. The prepayment is in line with our capital allocation priorities. During the quarter, we did not purchase any shares of our stock as we prioritized our capital allocation towards debt management. The fully diluted share count for the quarter was 249.5 million shares, up from 226.9 million shares in the prior year and versus 245 million shares in our guidance for the third fiscal quarter. Moving on to our guidance for the fourth quarter of fiscal 2026. We expect the fourth fiscal quarter revenue for Viavi to be up sequentially, driven by continued strength in many of our end markets across NSE and OSP. For NSE, we expect quarter-over-quarter revenue to be higher as a result of continued strong demand for our lab, production and field products driven by the data center ecosystem as well as for our aerospace and defense products. For OSP, we expect quarter-over-quarter revenue to be higher, driven by strength across all of the product lines. For the fourth fiscal quarter of 2026, we expect Viavi revenue in the range of $427 million and $437 million. We expect NSE revenue between $340 million and $348 million. OSP revenue is expected to be in the range of $87 million and $89 million. Operating margin for Viavi is expected to be 22.7%, plus or minus 50 basis points. NSE operating margin is expected to be 18.7%, plus or minus 50 basis points. OSP operating margin is expected to be 38.4% plus or minus 40 basis points. And EPS is expected to be between $0.29 and $0.31. Our tax expenses for the fourth quarter is expected to be about $10 million, plus or minus $500,000 as a result of jurisdictional mix. We expect other income and expense to reflect a net expense of approximately $12 million, and the share count is expected to be around 256 million shares. With that, I will turn the call over to Oleg. Oleg?
Thank you, Ilan. The results of the third quarter of fiscal '26 exceeded our expectations and came in above the high end of our guidance. The strong year-on-year and quarter-on-quarter performance was driven by strong growth in many of our end markets. NSE revenue in Q3 grew approximately 54% year-over-year, primarily driven by strong demand from the data center ecosystem and aerospace and defense customers. The data center ecosystem, which includes high-performance semiconductors, optical modules, NAMs, and the hyperscalers, drove strong demand for lab, production and field instruments in support of AI data center build-out. We are seeing strong demand across all data center segments: scale up, scale out and scale across. Acceleration of industry investment in ever greater communication speeds and chip-to-chip interconnect technologies are the principal drivers of strong demand for our optical transport, silicon photonics and communication protocol and high-speed Ethernet test equipment. The Q3 growth was also helped by a recently acquired Spirent high-speed Ethernet product lines, which gave us access to a large installed base of enterprise customers. HSE performance came in line with our expectations. Given strong and growing customer demand, we expect the data center ecosystem revenue momentum to continue through calendar 2026. Our Aerospace and Defense business also showed another strong quarter-on-quarter growth, driven by continued demand for our positioning, navigation and timing products. We expect this trend to continue through the calendar year. The service provider business, which includes field instruments, wireless and service enablement, was in line with seasonality. As you may recall, the service provider business is seasonally weaker during the March and September quarters and seasonally stronger during the June and December quarters. Some notable service provider dynamics during the March quarter included early orders from cable operators relating to the new DAA architecture and continued weak but stable demand for wireless test products. We do not expect wireless to recover to growth in the near term. Turning to OSP, OSP saw strong year-on-year growth, driven by strong demand for 3D sensing and anticounterfeiting products. Looking ahead to Q4, we expect NSE revenue to be up quarter-on-quarter, driven by continued strong and growing demand from the data center and aerospace and defense customers and seasonally stronger service provider spend. We expect OSP to be up also quarter-on-quarter, driven by strength across all product lines. In conclusion, we expect our data center and aerospace and defense end markets to be strong drivers for the foreseeable future. I would like to thank the Viavi team for its continued strong innovation and execution and thank our customers and shareholders for their continued support. With that, I will now turn it back over to the operator for Q&A.
分析師問答
Your first question comes from the line of Ruben Roy from Stifel. Your line is now open.
Great. Thank you. Hi, Oleg and Ilan. Congrats on the momentum here in the business. I guess to start, Oleg, maybe we could just drill into the data center momentum. And if you think about sort of the first half of the year and what you're seeing here with the beat here in the March quarter and the guidance for June, can you provide a bit more detail around the drivers by lab, production and field, and maybe what you're seeing in terms of visibility from your customers? Obviously, a lot going on with AI infrastructure and networks. I'm trying to get a little more detail around lab, production and field and how you see that trending from here as you look ahead.
On the lab side, it's the classical optical transport and PCIe/express test products. As companies develop new AI chips for inference and training, they require very high speeds across all ports and overall traffic. Everyone developing products for these next-generation systems, whether for AI training or inference, is buying our optical transport and protocol test solutions. That's primarily lab, but the same equipment is being bought by NAMs building optical switches and other gear. On production, we are seeing momentum in co-packaged optics, which plays very well to Viavi's traditional strengths with products that go into production lines where you measure spectral and optical performance of various optics. We're also selling a lot of that equipment to semiconductor vendors as they develop integrated packaged optical solutions. For field instrumentation, as data centers come online, they invest heavily to ensure peak performance. We've never seen so much demand for our fiber monitoring solutions. Data centers are buying more equipment than regular service providers in many cases, driving the field instrument side of the business. I would say data center-driven demand is now approaching 40% to 45% of field instruments and could soon be close to 50%. In that respect, the market needs align very well with our portfolio.
That's great detail. For a follow-up, you had started to see some hyperscaler activity around 800 gig last year and some of the things you're describing today, like 1.6 and co-packaged optics, are actually just starting. Is that the right way to think about it? In terms of your mix, is it still more weighted toward older generation technology, or are the 1.6 and other new products layering in now?
Both are true. 800G is still a high-volume driver. A lot of new development uses 1.6. We released 1.6 about 1.5 years ago. Initially, NAMs and optical equipment vendors deployed it in development, and now it's spreading to semiconductor vendors and co-packaged optics. So 1.6 is ramping in the lab, and we also have early production products to module vendors. Both older high-volume technologies and newer higher-speed technologies are contributing: one is ramping in volume while the other continues to grow as it matures.
Your next question comes from the line of Mehdi Hosseini from SIG.
Yes. Oleg, congrats on execution. I want to get a longer-term view. I see the midpoint of your June quarter guide implying annualized earnings of about $1.20, which is much higher than the prior peak from FY '22. With that context, how should we think about the company's earning power over the next one to two years? I'm not asking for a guide, I just want a better picture of how the new demand vectors that are materializing will enable longer-term earnings power. I have a follow-up after that.
Great question. I think our NSE business is getting very close to 20% operating profit, and much of that is volume-driven. For every incremental dollar in NSE, you're getting roughly $0.40 to $0.45 dropping to the bottom line. Ilan can provide more detail, but clearly that's driving operating margin expansion. OSP has typically had high operating margins, and with higher revenue its gross margins are improving as well, which flows to operating margin. Already at the blended level, we are in the low-20s percentage for operating margin, and as NSE continues to strengthen that should move into the mid-20s and potentially higher. There's significant operating leverage with volume. We've been able to manage pricing and component cost pressures, and to the extent we pass through pricing increases to customers, that will help gross margins. So the combination of maintaining pricing, sustaining healthy gross margins and volume growth drives substantial operating leverage. Fundamentally, from this point on, the business scales well; once fixed costs are covered, every incremental dollar has strong contribution to operating income.
For modeling, Mehdi, NSE's current incremental contribution to operating income is around the 40% level, and as the top line continues to grow, you should assume that operating income will grow proportionally as well.
Also, on taxes and NOLs: now that we are generating more profit and a lot of it falls in the U.S. jurisdiction because of where our IP and R&D is, incremental profit comes in at a low effective tax rate in North America, which is another benefit.
This quarter, for example, the effective tax rate is about 12%. So that is a clear benefit.
One other note: with convertibles, you do get some dilution as your stock price goes up, so you should factor a somewhat higher share count in your calculations.
Okay. My follow-up: you're scaling now and the midpoint of the June quarter implies about a 25% growth from the prior peak in mid-2022. Back then, 5G wireless was a big factor. It seems wireline — especially as 1.6 and subsequent 3.2 speeds roll out — offers a much bigger TAM. Given that, you have scaled revenue through both acquisition and organic growth. Where do we go from here? Could you help me understand growth from here, and whether opportunities like 1.6 and 3.2 could help you reach, say, $500 million of quarterly revenue at some point? I'm not asking timing, just whether $500 million quarterly is realistic.
I think it's entirely realistic. The intensity of test requirements increases as you go to higher speeds and with chip-to-chip interconnects. Co-packaged optics and near-package optics broaden where our equipment is bought. For example, when manufacturing multi-mode and hollow-core fiber, manufacturers now need our instruments in production lines. Likewise, co-packaged and integrated optics require extensive testing at many insertion points because yield is critical; you may not be able to rework a sealed module, so every component needs testing, then testing again after mounting, and final tests after ASIC integration. This drives a tremendous amount of test requirements in manufacturing. So the market scope is expanding. Also, our Aerospace and Defense business is growing nicely and contributes meaningful revenue. Wireless is down now but won't be down forever; its recovery could add $20 million to $30 million of quarterly revenue. Taken together, these trends make the $500 million quarterly target realistic over an upcycle.
To clarify: I wasn't sure whether Mehdi referred to $500 million for NSE alone or Viavi overall. If he meant Viavi overall, that is a different context. In any case, this is not necessarily immediate next fiscal year; it's over an upcycle.
Your next question comes from the line of Ryan Koontz from Needham.
Terrific results, guys, just excellent. Maybe a quick clarification on the data center customer mix. It sounds pretty broad and diversified. As you think about the segments — semiconductors, optical module/system makers, NAMs, operators — can you give an order of which segments are largest within the data center mix today?
It varies quarter-by-quarter. If you take one single segment, the hyperscalers are the biggest bucket because they buy for data centers and run R&D developing processors and modules. Within hyperscalers there's variance by company, but overall they are a large buyer. The next big bucket would be module makers and system makers, people making optical modules and optical systems. The next is the silicon vendors. Overall, it's a fairly balanced distribution across those buckets.
That's great color. Switching to Spirent: it's clearly a big part of your success in data center momentum. Can you talk about synergies you're seeing between the businesses on both product and sales sides?
Spirent brings an established customer base and strong application-hardened software for Ethernet traffic. We're upgrading hardware performance to make products more competitive and combining our fast-hardware development with their software strength. The first truly integrated product between us will be our combined 3.2-terabit product. We're leveraging our 800G position with their 800G Ethernet test solutions and expanding into customers they didn't have while gaining access to their customers, which drives broader strategic discussions across high-speed Ethernet and our other products.
Sounds like cross-selling is already starting. One last quick area: defense is an emerging area. How would you characterize progress in winning share for positioning, navigation and timing with growth in drones? And what are you doing in wireless and satellite — do you see LEOs as an opportunity?
For positioning, navigation and timing (PNT), initial revenue was driven by designs won before we acquired the business, but since acquisition we've seen continued ramp and an improving funnel of wins. Drones are a big driver, but essentially anything autonomous — air, land, sea, undersea — creates opportunity. We're now engaging U.S. Tier 1s more effectively after implementing processes around ITAR and secure access, enabling us to play in larger programs. On wireless and satellite: while wireless has been weak, we are involved in 5G-plus and 6G efforts, including non-terrestrial networks. Many of our advanced wireless applications focus on AI RAN and ground-to-satellite communications, so we are positioned for opportunities there as markets evolve.
Your next question comes from the line of Andrew Spinola.
I wanted to ask about component shortages and supply constraints. During the quarter, were you able to meet demand, or did supply constraints limit you? Also, we're seeing long-term supply agreements in the industry to meet hyperscaler demand and increases in visibility a couple of years out. Have you started to enter long-term supply agreements with bigger customers? What's evolving on that side?
We don't typically enter into long-term supply agreements because our volumes and product mix require early access to alpha and beta silicon so we can develop products before silicon is fully qualified. In test and measurement, customers pay higher ASPs and that generally provides access to components; the issue is lead time and giving vendors sufficient notice rather than absolute availability. If we get upside orders with too little lead time, that's when we can be constrained. For mature products like field handhelds, cost matters more and we generally maintain inventory. We increased inventory ahead of anticipated shortages a few quarters ago to secure parts, and that has proven beneficial. Memory, for example, is expected to be in deficit through 2030 per some studies, but given our volumes it's manageable. Early access to components and managing lead times is more important than long-term supply contracts for our product set.
If you look at our balance sheet for March, inventory is up by a single-digit million. The majority of that increase was to secure additional components for the production ramp and expected demand.
We look two to four quarters out and make bets to ensure lead times are covered. The last thing you want to do is be pennywise and pound-foolish when it comes to availability.
Appreciate that color. One follow-up: you talked about incremental margins earlier of 40% to 45%, which makes sense. For Q4 in your guide and thinking about fiscal '27, I'm assuming the 40% to 45% incremental on the core business as you scale. You announced about $30 million of restructuring last quarter, and you mentioned some of that will be reinvested. How much of that will drop to the bottom line versus be reinvested? My assumption is a good chunk will hit the bottom line, potentially pushing incrementals closer to 50% in fiscal '27. Any color?
We'll implement most of the restructuring by the end of our fiscal year in the June quarter, with some remainder through the end of the calendar year. A portion of the savings will be reinvested. Ilan, you said roughly one-third gets reinvested?
Yes. The current fall-through is about 40%. There's reason to assume it can increase. The restructuring actions will take until the end of the calendar year to fully materialize. With top-line growth and the restructuring realized, we expect the fall-through to be more visible in the second half of fiscal '27, meaning the first half of calendar 2027.
To clarify seasonality: the March and September quarters are seasonally down for the service provider segment. That pattern remains, but it is now on a smaller scale because data center and aerospace/defense growth offsets much of it. You will see a bigger jump between March and June because of the service provider tailwind, and a smaller increase into September because of the seasonal headwind, then stronger again into December.
On quarter-to-quarter trajectory overall, December is stronger versus September, and September is generally more muted relative to June.
That's right.
Your next question comes from the line of Michael Genovese from Rosenblatt Securities.
Thanks. Exciting times, guys. Congratulations for being right in the middle of it. Oleg, I keep hearing that as we go to silicon photonics and co-packaged optics, bottlenecks are moving to packaging and test and measurement for electronics, optics and ASICs. It seems like there's a lot of testing required for co-packaged optics. First question: do you agree testing is a bottleneck? If so, how will you address that over time to take advantage of it?
Amen — you're absolutely right. Test and packaging used to be a back-end afterthought; it's now strategic. Leading semiconductor companies consider packaging expertise a core capability. You're integrating glass substrates, photonic integrated circuits next to electronic integrated circuits, embedding optics into substrates, and it all becomes extremely complex. That is positive for us because our technologies and capabilities are being pulled into the value chain: from individual optical components to wafer-level packaging to heterogeneous integration packaging and into major test platforms. Additionally, many custom rack-mounted test systems being built by leading players use our hardware as the guts. So the test market is expanding in scope and importance, and we are well positioned to serve that expanding need.
Great. As a follow-up: for newer areas like OCS and co-packaged optics, are these represented in current numbers at all, or are they still mostly future opportunities? And when you say 'foreseeable future' you feel great about growth, over what time horizon are you referring — multiple years or quarters?
These are in the early sales today; you can see early innings now. By 'foreseeable future' I typically mean the next three quarters. We're already seeing early demand and ramps that will start to show more materially in the coming quarters.
Your next question comes from the line of Tim (Timothy) Savageaux from Northland Capital Markets.
Congrats on some pretty spectacular results. Quick question: Spirent was $54.2 million in the quarter, is that right?
That's correct.
What do you expect for Spirent next quarter?
Spirent benefited from a few orders that got pushed from the prior quarter into this quarter, which made this quarter a bit stronger seasonally. On an annual basis, the Spirent run rate is still around the $200 million we indicated, with a split of about 45%/55% between different lines. Normalizing for the timing, expect Spirent to be roughly $48 million for the June quarter.
That suggests higher levels of organic growth in NSE. Could you see organic growth rates approaching 30% to 40% for NSE over the next few quarters or into next year?
Percentages are tricky as the base grows — 30% on a small base is different from 30% on a larger base. We focus on absolute dollar growth, and maintaining strong dollar growth should be realistic given current market dynamics and the early innings of multiple trends.
Also keep in mind service provider seasonality impacts comparisons; September is traditionally down for that segment. So year-on-year growth can be strong, but quarter-to-quarter seasonality will still be present.
Understood. Could you provide an updated breakdown for NSE among data center, defense and service provider? Previously you suggested something like 45%, 15%, 40%; is that still in the right area?
Exit velocity this year is moving data center toward the high 40s percent of NSE revenue. Service providers are toward the mid-30s and aerospace and defense a little over 15%. I wouldn't be surprised if data center reaches about 50% of NSE revenue in the not distant future.
Your next and final question comes from the line of Andrew Spinola with a follow-up. Andrew, your line is now open.
I wanted to ask about the performance of the Inertial Labs / inertial navigation business that you acquired. There's a lot of demand and new programs in that space. How is that business performing? Do you have the approvals and ability to sell into all of the customers, and how should we think about that opportunity over the medium term?
Good question. The fact that we paid out a significant earn-out indicates the acquired business exceeded forecasts. In my experience, acquisitions rarely exceed first-year forecasts; this one has done exceptionally well. They produce a range of products from basic sensors to full inertial navigation systems with sensor fusion (GNSS, LiDAR, and others). We engage with many drone and subsystem vendors globally, including U.S. vendors. We have clear boundaries regarding export controls: products meeting certain accuracy thresholds require export approvals, while many commercial products for surveillance, agriculture, mining, etc., are commercial and can be sold without the same restrictions. We've implemented procedures to ensure compliance and to define product grades and pricing accordingly.
One follow-up: given strong growth in lower-cost drones, are your inertial modules applicable to the lower-end drone market? For example, would the inertial sensors be suitable for a $3,000 drone versus a $30,000 drone?
For very low-cost drones around $3,000, typically no; for drones in the $30,000 range, yes our modules are applicable. There are clear distinctions in required performance and accuracy that determine the applicable product and export controls.
There are no further questions at this time. I will now turn the call back to Vibhuti Nayar for closing remarks.
Thank you, Hillary. This concludes our earnings call for today. Thank you for joining everyone, and have a good afternoon.
This concludes today's call. Thank you for attending. You may now disconnect.