Good afternoon. I will be your conference operator. At this time, I would like to welcome everyone to the Applied Optoelectronics second quarter 2019 earnings conference call. All lines have been placed in mute to prevent any background noise. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After the speaker's remarks, there will be a question and answer session. To ask a question, you may press star, then one on your telephone keypad. To withdraw your question, please press star, then two. Please note, this event is being recorded. I now turn the call over to Maria Riley, investor relations for AOI. Ms. Riley, you may begin.
Thank you. I'm Maria Riley, Applied Optoelectronics investor relations, and I'm pleased to welcome you to AOI's second quarter 2019 financial results conference call. After the market closed today, AOI issued a press release announcing its second quarter 2019 financial results and provided its outlook for the third quarter of 2019. The release is also available on the company's website at ao-inc.com. This call is being recorded and webcast live. A link to that recording can be found on the investor relations page of the AOI website and will be archived for one year. Joining us on today's call is Dr. Thompson Lin, AOI's Founder, Chairman, and CEO, and Dr. Stefan Murry, AOI's Chief Financial Officer and Chief Strategy Officer. Thompson will give an overview of AOI's Q2 results, and Stefan will provide financial details and the outlook for the third quarter of 2019.
A question and answer session will follow our prepared remarks. Before we begin, I would like to remind you to review AOI's Safe Harbor statement. On today's call, management will make forward-looking statements. These forward-looking statements involve risks and uncertainties, as well as assumptions and current expectations, which could cause the company's actual results to differ materially from those anticipated in such forward-looking statements. In some cases, you can identify forward-looking statements by terminology such as believes, anticipates, estimates, intends, predicts, expects, plans, may, should, could, would, will, or thinks, and by other similar expressions that convey uncertainty of future events or outcomes. Forward-looking statements also include statements regarding management's beliefs and expectations related to the expansion of the reach of our products into new markets and customer responses to our innovations, as well as statements regarding the company's outlook for the third quarter of 2019.
Except as required by law, we assume no obligation to update forward-looking statements for any reason after the date of this earnings call to conform these statements to actual results or to changes in the company's expectations. More information about other risks that may impact the company's business are set forth in the Risk Factors section of the company's reports on file with the SEC, including the company's annual report on Form 10-K for the year ended December 31, 2018. Also, with the exception of revenue, all financial numbers discussed today are on a non-GAAP basis unless specifically noted otherwise. Non-GAAP financial measures are not intended to be considered in isolation or as a substitute for results prepared in accordance with GAAP.
A reconciliation between our GAAP and non-GAAP measures, as well as a discussion of why we present non-GAAP financial measures, are included in our earnings press release that is available on our website. Before moving to the financial results, I'd like to announce that AOI management will attend the D.A. Davidson Technology Conference in New York on September 4, and the Dougherty & Company 2019 Institutional Investor Conference in Minneapolis on September 5. We hope to have the opportunity to see many of you there. Additionally, I'd like to note the date of our third quarter 2019 earnings call is currently scheduled for Wednesday, November 6, 2019. Now I'd like to turn the call over to Dr. Thompson Lin, Applied Optoelectronics Founder, Chairman, and CEO. Thompson?
Thank you, Maria, and thank you everyone for joining us today. We are pleased with our execution during the quarter. We delivered revenue in line with our guidance and achieved better-than-expected results on the bottom line. AOI delivered revenue of $43.4 million, non-GAAP gross margin of 27.2%, and a non-GAAP net loss of $0.26 per share. In looking at the dynamics in the quarter, the data center demand environment remained consistent with our expectations. We are starting to see early signs of recovery among two of our hyperscale data center customers. While one customer has yet to begin a recovery, we are encouraged by this early sign of recovery and believe the fundamental need for higher bandwidth within hyperscale data centers will drive long-term growth. However, in the short term, we remain cautiously optimistic on the market dynamics as the demand environment continue to stabilize among our hyperscale customers.
In CATV, we remain encouraged by customer activity, especially interest in our Remote PHY products. However, the overall CATV market demand continued to be soft, resulting in tepid demand for some of our legacy products. CATV demand in China is weaker than we had expected as a result of trade tensions. Diversifying our customer base remain a top priority for AOI. In the quarter, we secured five new design wins, including four with an equipment OEM for data centers and one with a data center operator. In summary, we are pleased with our execution this quarter, which contributed to our better-than-expected bottom line results. We remain focused on fostering relationships with both existing and new customers and expanding our technology leadership.
We believe our platform proprietary manufacturing process and vertical integration are keys to our success in the market, and we remain confident in our ability to monetize our innovation as the market improve and move to the next generation technologies. With that, I will turn the call over to Stefan to review the details of our Q2 performance and outlook for Q3. Stefan?
Thank you, Thompson. Overall, the demand environment in the quarter was consistent with our expectations. Total revenue for the second quarter was $43.4 million, which was above the midpoint of our guidance range of $40 million to $45 million. Our data center revenue came in at $31.8 million, compared with $69 million in Q2 of last year. In the quarter, 72% of our data center revenue was from our 40G transceiver products, and 23% was from our 100G products. The data center market dynamics played out in Q2 as expected. We are starting to see early signs of recovery among two of our hyperscale data center customers, while one customer continues to purchase product from us, but with reduced demand. As Thompson mentioned, while we are encouraged by these early signs of a recovery, we remain cautiously optimistic on the near term market dynamics.
We continue to believe that we have good relationships with all of our hyperscale data center customers, and that their need for high-speed optical connectivity remains fundamental to their business. We are focusing our efforts on continuing to foster relationships with both existing and new customers and expanding our technology leadership, which we believe will best position AOI for growth when market conditions improve. We are also encouraged by the pace and quality of the design wins we are seeing with new customers, many of whom are data center operators or equipment OEMs that supply the data center vertical. Building upon our strong foundation as a leader in advanced optical technology, we recently showcased the ability of AOI's 400G QSFP transceivers to break out into four individual 100G FR transceivers and interoperate with a leading 12.8 terabit per second switch fabric ASIC.
As data center operators continue to demand greater bandwidth, the migration from 100G to 400G will be the next major step in data center architecture. As data center customers add 400G connectivity to their 100G infrastructure, they are looking for validated and interoperable solutions to gain confidence and reduce deployment timelines. We are very pleased to have a solution with the demonstrated interoperability that our customers demand. Turning to our cable television market. Revenue from CATV products decreased 31% year-over-year to $9.8 million, compared with $14.2 million in Q2 of last year, as demand has weakened somewhat with North American MSOs and the China CATV market continues to lag expectations due to trade tensions and concerns about domestic economic growth in China. Despite these near-term challenges, MSOs, particularly those in North America, continue to forge plans for distributed access architectures.
We believe that our Remote PHY product is a key enabling technologies for these new distributed access networks, and we are excited about the customer interest in Remote PHY. We expect to receive our first significant orders for our Remote PHY product soon. Our telecom products delivered revenue of $1.6 million, compared with $4.2 million in Q2 of last year, reflecting lower sales in China given geopolitical trade tensions. In telecom, we continue to see 5G network deployments poised to become a large driving factor for the optical industry as a whole. We believe AOI is well-positioned to grow our share as the 5G optics market develops, given our deep optical expertise in harsh outdoor environments and our highly automated module production process. We remain in qualification with a number of vendors for both front and mid-haul applications.
With that said, please keep in mind that given this is an emerging market, the timing of qualification and deployment schedules are difficult to predict. For the quarter, 73% of our revenue was from data center products, 23% from CATV products, with the remaining 4% from FTTH, telecom, and other. In the second quarter, we had three 10% or greater customers. Two in the data center business that contributed 30% and 29% of total revenue respectively, and one in the CATV business that contributed 14% of total revenue. We continued to build on our earlier success in diversifying our customer base and are pleased with the steady progress we have made. In the quarter, we secured a total of five new design wins among two U.S.-based data center customers, one of which is a data center operator.
I will also note that several of these design wins expand on a new customer relationship we secured last quarter with an OEM supplier to the hyperscale and enterprise markets. Moving beyond revenue, we generated a gross margin of 27.2%. A 170 basis point improvement from 25.5% reported last quarter, and slightly higher than our guidance. Total operating expenses in the quarter were $19.5 million, or 44.9% of revenue, compared with $20.3 million, or 38.4% of revenue in the prior quarter. We continue to be targeted with our investments, with an emphasis on developing and enhancing our next generation of optical products, while also tightly managing expenses. Operating loss in Q2 was $7.7 million, compared with an operating loss of $6.8 million in Q1. Non-GAAP net loss after tax for the second quarter was $5.2 million, or a loss of $0.26 per basic share, which was better than our guidance.
This compares to net income of $12.9 million or $0.64 per diluted share in Q2 of 2018. GAAP net loss for Q2 was $11.4 million, or a loss of $0.57 per basic share, compared with GAAP net income of $8 million or $0.40 per diluted share in Q2 of last year. The basic shares outstanding used for computing the net loss in Q2 were 19.9 million shares. Turning now to the balance sheet. We ended Q2 with $84 million in total cash equivalents, short-term investments and restricted cash, compared with $77.5 million at the end of the previous quarter. This reflects $7.2 million in cash generated from operations. As of June 30, we had $81.5 million in inventory, a decrease of $3 million from Q1. This inventory reduction is consistent with our long-term plan as we continue to rationalize inventory levels.
We made a total of $13.5 million in capital investments in the quarter, including $6.2 million in production equipment and machinery and $6.9 million on construction and building improvements. Looking ahead, we now expect capital expenditures in 2019 to be approximately $56 million, which factors in a continuation of the construction of our new factory in China. We continue to monitor end market conditions and may adjust our spending plans as necessary. Moving now to our Q3 outlook. We expect Q3 revenue to be between $46 million and $49 million, and non-GAAP gross margin to be in the range of 27%-29%. Non-GAAP net loss is expected to be in the range of $4.2 million-$5.7 million and non-GAAP loss per share between $0.21 per share and $0.28 per share, using a weighted average basic share count of approximately 20 million shares.
With that, I will turn it back over to the operator for the Q&A session. Operator?
Yes. Thank you. We will now begin the question and answer session. To ask a question, you may press star then one on your touchtone phone. If you are using a speakerphone, please pick up your handset before pressing the keys. To withdraw your question, please press star then two. At this time, we will pause momentarily to assemble the roster. The first question comes from Simon Leopold with Raymond James.
Great. Thank you for taking the question. Just a quick clarification, if I might. I think you mentioned your 10% customers. I didn't get down the color you offered on that. Could you just repeat that comment?
Yeah. We had three 10% customers during the quarter. Two were in the data center business. They contributed 30% and 29%, respectively, of total revenue. There was one customer in the CATV business that was 14% of total revenue.
Great. Thank you very much for that. Sorry. On the CATV business, clearly we've heard from the major operators spending less money, but it seems as if we're still very early in the fiber deep Remote PHY. I think you made a comment suggesting that you were only just beginning to ship your Remote PHY boxes. If you could give us a sense of how you see this playing out, and I guess what I'm really getting at is how should we think about the trending of this business, both near term, third quarter, and then really looking at kind of 2020?
Yeah. The comment that we made, I think, on the last few conference calls, we've mentioned that we have been selling Remote PHY products. The comment that we made is that we're expecting to start getting our first, what we would term sort of significant orders. That is something that would be the beginning of more of an ongoing business for those Remote PHY products and one that we hope would grow into a larger number over time. As far as the overall cadence on cable TV, I think if you look year-over-year, what we've seen is primarily related to China slowing down. We've seen some slowdown in North America in the last quarter or two. It depends if you're looking on a sequential basis or year-over-year in terms of what's causing the downturn.
I think we're looking for the North American MSOs to begin to invest in these distributed access architectures. As I've mentioned in our prepared remarks, we're a technology leader in Remote PHY, which is a key aspect of these distributed access architectures moving forward. It's hard to say exactly when they're going to do that. I think they're poised to. I think some of the slowdown that we're seeing now among the North American MSOs is probably related to the immediacy of their transition to this Remote PHY-based architecture. They're kind of minimizing their investments in legacy networks while they look to add Remote PHY. The precise timing behind that is difficult to predict.
I think it's probably not a Q3 or Q4 kind of thing before we start to see a real resurgence, but it's a little hard to predict at this point.
I guess I'm sort of reflecting back on the cable TV business in 2017, where it was very much transmission oriented. You did about $60 million. I'm just wondering if we should think about that as a reasonable expectation for 2020 or at least the timeframe where these initiatives really get going. Is that a reasonable way to think about that line of business?
Again, I don't want to give you a sort of precise number. I think there's every reason to believe we can get back to levels similar to or greater than what we've been at in the historical periods. It does require this transition to Remote PHY, I think, to happen in North America. Like I said, I would expect that that would happen in 2020, although the cable TV market is notoriously difficult to project, specifically the timing of when they start to get going. I think the overall trend we can be fairly certain of. Exactly when they get going and how fast they ramp up is still a little bit tough to forecast.
Thank you. Just one more, if I might. You mentioned the 400G products starting to come out. Just if you could help us think about how you could be competitive versus the silicon photonics variants that are coming out from some of the OEMs and some of the larger semiconductor companies. Just wondering how silicon photonics sort of plays into the competitive landscape when you're in a market at 400G inside the data center. Thank you.
Sure. Silicon photonics is not a new technology. As you know, we've had silicon photonics at 100G, and there's been silicon photonics solutions at lower data rates even before that. Our competitive advantage is built upon our vertical integration. That is our ability to manufacture a significant part of the cost-driving elements of the transceiver in-house, and also on our manufacturing expertise. I think we've talked extensively in the past about our automated manufacturing processes and our platform technology that has allowed us to automate those processes. It's not just the automation itself, but it's having a design for our 100G products and our 400G products and even future generations where we can manufacture those in an automated way in a very cost-effective manufacturing process. That's what gives us the ability to compete with those other technologies.
Simon, this is Thompson. I want to emphasize for 400G and 800G, making EML in-house will give AOI even stronger advantage, all right, compared to 100G using the DML.
Great. Thank you very much for taking my questions. Appreciate that.
Thanks, Simon.
Thank you. The next question comes from Samik Chatterjee with JP Morgan.
Hi. Thanks for taking my question. If I could just start off with a clarification as well. I know you mentioned the 100G and 40G mix in the data center revenues. Could you just repeat that? Sorry, I missed that.
Sure, no problem. The 72% of the data center revenue was from 40G and 23% was from 100G in the quarter.
Got it. I think that kind of implies a decline in the 100G mix overall, a strong decline in the revenues. Is that primarily driven by the lack of recovery that you saw with one of the hyperscale customers or data center customers, as you call them, or was that more driven by something else that I'm not really thinking about?
No, it's almost entirely driven by the one customer who has yet to recover. We are seeing strength in our 40G, which I think is actually a good thing for AOI. We've been a leader at 40G for some time. The fact that our customers continue to be interested in 40G and continue to find new use cases for 40G and are continuing to buy significant quantities of 40G, I think is very good for us. The 100G downturn is not related to other customers. It's pretty much isolated to the one customer.
Got it. Just a question on the tariffs, like with the proposed 10% tariff now on incremental goods imported from China, are you expecting any impact to your gross margins? Additionally, I believe you have a facility in Taiwan. Are you seeing any pickup in interest from customers in expanding their business in that facility? If you were to ask to expand capacity there, how much flexible capacity do you have there?
We do have a facility in Taiwan that can manufacture the data center transceivers. In fact, it is already manufacturing a portion of our data center transceivers. We have had significant interest from customers in our ability to manufacture in Taiwan. What we can do is move some of the manufacturing operations between our Taiwan and China factories such that we can add additional capacity if needed as these tariffs come on board. In other words, what I mean is we can take some of the manufacturing for other ancillary products that are maybe not data center related, move those to China and increase the capacity in Taiwan for the data center products. Where we stand right now, we think we're pretty well positioned to be able to manufacture what the customers are asking us in the Taiwan factory, at least for customers that are U.S.-based.
I want to remind everyone Even among the U.S. hyperscale customers, not all of their transceiver usage is actually in the U.S. We won't necessarily be manufacturing all of our data center transceivers in Taiwan. For the ones that need to be imported into the U.S., it's certainly a possibility for us to manufacture those in Taiwan, and that's our plan should the tariffs come in place.
Got it. Thank you for taking my questions.
You're welcome.
Thank you. Once again, please press star and then one if you would like to ask a question. The next question comes from Fahad Najam with Cowen and Company.
Hi, Stefan. Hi, Thompson. I apologize for the tough question, but did I hear you correctly that 100G was 23% of data center revenue?
Yes, you did.
If I look at the broader landscape going forward, why should investors believe that you would meaningfully have any success in 400G, especially when 40G rolls off, when you have had little to no meaningful success of late since the quality issue with your lasers in 100G? What would you tell investors to have hope in your story? Probably for the broad-ended question, but I'm just struggling to see how if you're not succeeding in 100G, how will you succeed in 400G?
Well, I think it's a mischaracterization to say that we're not succeeding in 100G. As you noted, we have sizable sales of 100G last year. In fact, it was our largest selling product line by far. I would not call that not having success. What I would say is that different customers purchase different applications and different data rates for different applications at different times. Not every customer, as we've noted in our prepared remarks, has yet begun a recovery cycle. We would expect that in that recovery cycle, if they're still purchasing large quantities of 100G, that is, if they haven't moved on to 400G, then we'd expect to be a part of that. Now, with respect to your sort of more blunt question about why would we be a player at 400G, we are actively involved in a number of qualifications right now.
I think if customers had decided they weren't going to use AOI or they weren't attracted by what AOI had to offer, they're not going to waste their effort and resource working with us on these qualification efforts. Now, those qualifications are ongoing. I can't tell you for sure what the results of all those are going to be, but so far, the results are good, and I would expect that some of them at least would be concluded successfully, maybe perhaps all of them. The other thing I'd like to say is, 40G and 100G weren't our first data rates. We've been involved in the data center market for a long time. We've been a leader in the data center market for a long time, and I don't see any reason why 400G would be materially different.
As Thompson mentioned earlier, technologies like our electro absorption modulated laser or EML are critical to not only to the performance, but to the cost structure of the 400G transceivers. By having that technology in-house, we think that gives us a really good position to be not only a technology leader, but a cost leader in 400G, as we were at 100G, as we were at 40G, as we were at 10G, and as we expect to be at 800G when that comes to fruition in the future. I think it's wrong to say we haven't been successful, and the technology that we've developed for 400G is very compelling, which is why we have ongoing qualifications going with customers.
Let me add for the two points. One, for EML, 100G EML, there are very few suppliers, okay, compared to 25G DML. Making EML in-house, the cost of it is very big, much, much bigger than 25G DML for 100G transceiver. That is number one. Number two, yes, we have quality issue, we have mentioned we have solved the problem. In the past few quarters, we have many design wins of 100G transceiver with many new customer, okay? Not only in U.S., including Asia, including many big equipment, OEM company, and many hyperscale operator, okay, worldwide. The slowdown related to this specific customer is not the quality issues. It's their demand really slowed down. Okay? We are confident, and we believe when the demand come back, we're still one of the major supplier, okay, in the future. Could be sometime next year. All right.
All right. If I may ask on the 200G. I know 400G is still a second half 2020 story with most of your customers. Two of the largest hyperscale cloud titans are moving with 200G in the interim. One of them happens to be a customer of yours. Are they doing any 200G? Are you shipping 200G? Do you have any share in 200G at the moment?
Yes. We've been shipping 200G since last year. It's not a huge quantity, obviously, if you look at the percentages for the 40G and 100G, but we do have design wins at 200G.
Thank you. The next question comes from Michelle Walker with Needham & Company.
Hi, guys. I'm on foot for Alex Henderson. Just a quick question on the gross margins. Can you guys give any color on 40G gross margins or 100G? We're just trying to wonder here if gross margins are positive for 100G or no.
They're definitely positive for 100G. We don't give specific guidance on individual product gross margins. Certainly they're positive.
Okay.
The gross margin for 100G is pretty good. As we said, we have very strong cost advantage compared to other suppliers because of the vertical integration, because of the automation of the transceiver manufacturer in Taiwan and China.
Okay.
I'm sorry, was there a follow-on question?
No. That's good for me. Thanks.
All right. Very good. Thank you.
Thank you. As there are no more questions, I would like to return the call back to Dr. Thompson Lin for any closing remarks.
Again, thank you for joining us today. As always, we thank our investors, customers, and employees for your continuous support, and we look forward to seeing you at our upcoming conference.
Thank you. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect your lines.