Silicon Laboratories Inc (SLAB) 2017 Q3 法說會逐字稿

Good morning. My name is Marcella, and I will be your conference operator today. At this time, I'd like to welcome everyone to the Silicon Labs Third Quarter Fiscal 2017 Earnings Conference Call. (Operator Instructions) I will now turn the call over to Jalene Hoover, Director of Investor Relations and International Finance. Jalene, please go ahead.

Thank you, Marcella, and good morning, everyone. Tyson Tuttle, Chief Executive Officer; and John Hollister, Chief Financial Officer, are on today's call. We will discuss our financial performance and review our business activities for the third quarter. After our prepared comments, we will take questions. Our earnings press release and the accompanying financial tables are available in the Investor Relations section of our website at www.silabs.com. This call is also being webcast, and a replay will be available for 4 weeks.

Our comments today will include forward-looking statements subject to risks and uncertainties. We base these forward-looking statements on information available to us as of the date of this conference call and assume no obligation to update these statements in the future. We encourage you to review our SEC filings, which identify important risk factors that could cause actual results to differ materially from those contained in any forward-looking statements.

Additionally, during our call today, we will make reference to certain non-GAAP financial information. A reconciliation of our GAAP to non-GAAP results is included in the company's earnings press release and also in the Investor Relations section of Silicon Labs' website.

I would now like to turn the call over to Silicon Labs' Chief Financial Officer, John Hollister.

Thanks, Jalene. We delivered strong financial performance in Q3, with revenue ending at the top end of our guidance range at $199 million. This represents robust total company revenue growth of 12% year-on-year and our fifth consecutive quarter achieving target operating model product revenue growth.

The IoT products category achieved a record $100 million in revenue in Q3, representing year-on-year growth of 23%. Within IoT, we continue to see very strong results from our wireless portfolio, primarily due to good performance by our mesh networking and proprietary products and applications for the connected home, security, smart metering and the broad industrial market.

The Infrastructure portfolio also achieved record revenue at $39 million, up 2% sequentially. Our isolation products achieved another record quarter in Q3 with traction in applications for battery management for electric vehicles, industrial automation, solar power and high-efficiency power supplies. Revenue for our timing products was flat in the quarter and continues to be impacted by softness in the optical networking market.

As expected, we saw seasonal strength in Broadcast, with revenue ending at $43 million, up 17% sequentially and 6% year-on-year. Broadcast growth was across the board with a record in automotive and increases in consumer audio and video.

As we anticipated, Access was down slightly in Q3, ending at $17 million.

We saw revenue growth in all geographies with sequential growth in APAC the most significant, driven by third quarter growth in Broadcast, IoT and isolation.

By end market, strong performance in Broadcast drove increases in consumer followed by continued growth in the Industrial end market propelled by IoT. Industrial now represents around 50% of our total revenue, and that mix has more than doubled since 2013, underscoring the growing diversity of our customer base and the quality of the end markets we serve.

Third quarter distribution revenue was 70.3%, and our top 10 customers accounted for 21% of Q3 year-to-date revenue.

Non-GAAP gross margin in Q3 ended slightly above guidance at 58.8% on product mix. Non-GAAP operating expenses were also favorable, ending at $73.5 million, which was down about $1 million sequentially. Non-GAAP R&D expenses were lower than we expected at $41 million due to delays in tape-out spending. Non-GAAP SG&A expenses were also down slightly at $32 million, primarily on seasonal trends in payroll taxes and other employee benefits.

Non-GAAP operating income ended strong at $43 million, with non-GAAP operating margin at 21.8%, which is a multiyear high and well within our target operating range.

Our non-GAAP effective tax rate was better than expected for the quarter at 9% due to some favorable discrete items realized in the quarter. Non-GAAP earnings per share was also at a multi-year high, ending at $0.90, above the high end of our guidance range and representing 17% year-on-year growth.

On a GAAP basis, third quarter gross margins were 58.7%. GAAP operating expenses were flat at $92 million, and we generated GAAP operating profit of $25 million, or 13% of revenue. Stock compensation expense was $11 million for the quarter, and amortization of intangible assets was $7 million, both in line with expectations. GAAP diluted earnings ended the quarter at $0.46 per share, which was above the guidance range.

Turning now to the balance sheet. We ended the quarter with $724 million in cash and investments, up $57 million sequentially. For the 9-month year-to-date period, we generated outstanding operating cash flow of just over $140 million, which represents a year-over-year increase of 33%.

Accounts receivable ended at $76 million or 35 days sales outstanding. Our inventory balance grew $5 million sequentially to $73 million and represents inventory turns of 4.5x, which is in line with our operating targets.

I will now cover guidance for the fourth quarter.

We expect revenue in the fourth quarter to be between $195 million and $201 million with growth in IoT and Infrastructure. We expect Broadcast and Access to decline.

We expect our non-GAAP gross margin to be approximately 58.5% and our non-GAAP operating expenses to be around $73.5 million. We expect our non-GAAP effective tax rate to be around 11% and non-GAAP earnings per share to be in the range of $0.83 to $0.89.

On a GAAP basis, we expect gross margin to be approximately 58.5%. We expect GAAP operating expenses to be approximately $91 million for the quarter, with stock compensation at $11 million and amortization of intangible assets at $7 million. We expect our GAAP effective tax rate to be 11% and GAAP EPS to be in the range of $0.40 to $0.46.

Before I turn the call over to Tyson, I want to briefly comment on the upcoming change in revenue recognition due to the implementation of ASC 606. We have been working diligently on this change over the past year, and we are well-positioned to proceed with the implementation of this new standard, which will be effective in January of 2018. Based on the new guidance, we will change our revenue recognition criteria for sales into distribution customers from sell-through to a modified version of sell-in accounting. We will enact this change on a prospective basis with a one-time adjustment to retained earnings effective in January.

I will now turn the call over to Tyson.

Thank you, John. We're very pleased with our third quarter results, achieving record revenue in IoT and Infrastructure and in total. We've delivered target operating model performance on product revenue growth, non-GAAP gross margin and non-GAAP operating margin for four of the past five quarters, with growth in our IoT products a key driver of these results.

According to Gartner, by 2020, IoT technology will be in 95% of new electronic product designs. We are seeing tremendous demand for our IoT products, which have grown to 50% of our total revenue. Silicon Labs' biggest opportunity and challenge is to scale our IoT business. We design our solutions using a platform-based strategy to drive portfolio efficiency and to provide a flexible cost-effective means to support end-market needs. We also create solutions which are easy to use and support, to more efficiently scale in the sales channel and across the broad market, serving tens of thousands of customers and thousands of applications. We're pleased with our progress in these areas, and we believe our approach will enable us to continue to meet or exceed our 20% growth target for our IoT products.

In Q3, IoT delivered a seventh consecutive record revenue quarter, with revenue at $100 million, and up 23% year-on-year. Growth in our wireless products continues to outpace the overall market. Our connectivity portfolio is gaining traction as we target low-power wireless end nodes and offer a broad range of protocols in optimal combinations for key market segments, including home automation, security, metering, lighting and other industrial applications.

Our Zigbee products continue to lead wireless revenue growth. We've established ourselves as the technology and market leader in mesh networking, shipping well over 100 million mesh-enabled devices worldwide over the past 15 years.

Zigbee's scalable, self-configuring and self-healing mesh networking capabilities are ideal for applications demanding robust performance. We see an increasing number of trends driving growing demand for technologies supporting low-power wireless end node devices.

First, government agencies are mandating utility companies to upgrade to advanced metering infrastructures as part of larger, smart grid initiatives. The smart metering rollout in Great Britain is one of the most concentrated opportunities for our Zigbee and Sub-GHz solutions.

Zigbee and Thread mesh-enabled Wi-Fi hubs, routers and gateways are a recent trend driving broader adoption of low-power mesh networks and the proliferation of IoT end nodes, where we have a significant share of the market.

Additionally, voice control is becoming ubiquitous, driving end node deployment in smart home ecosystems, including Amazon Echo, Google Home, Nest and Samsung SmartThings Cloud. We are intercepting this trend with our broad wireless connectivity portfolio and strong positioning in IoT ecosystems.

And finally, IHS predicts the market for smart lighting and connected lighting controls will more than double from $6 billion in 2015 to more than $12 billion in 2020.

Our IoT products also contribute -- our MCU products also contribute to IoT growth, led by our EFM8 MCUs. Silicon Labs offers 8- and 32-bit USB connectivity solutions, which have been shipping for more than a decade, enabling developers to add USB to embedded designs without the added cost and complexity of firmware development. In Q3, we also introduced a comprehensive reference design to simplify the development of rechargeable battery packs used to power portable devices through USB Type-C cables.

As we look at our IoT portfolio, sensors are a key element of end node solutions in markets we address. During the quarter, we introduced a breakthrough Hall-effect magnetic sensor, offering industry-leading power efficiency, programmability, best-in-class sensitivity and built-in tamper detection. Our new sensors combine the low-power consumption of reed switches with the reliability of Hall-effect sensors to create an optimal solution for white goods, flow meters, motor controls, security and other applications. Our new portfolio allows developers to choose the right magnetic sensor for their application needs, enabling them to design, test and deploy a wide range of sophisticated position-sensing products better, faster, cheaper and with longer battery life.

Now, I'll move on to Infrastructure, where we delivered record revenue in the third quarter, driven by our isolation products. Our isolation devices are among our fastest-growing product lines, underscored by record Q3 design wins and achieving an 11th consecutive record revenue quarter. Renewable energy and increasing government initiatives to encourage energy efficiency are global trends driving broad-based adoption of our isolation technology. Systems requiring high voltage protection include electric and hybrid electric vehicles, solar inverters, motor controls and power supplies. Superior performance and robustness have allowed our isolation products to gain share from traditional optocouplers and outgrow the digital isolation market.

Also in Infrastructure, our timing products continue to suffer from weakness in the long-haul optical networking market. This has impacted timing revenue performance, which has fallen short of our expectations for the year. Despite this near-term softness, we remain optimistic about our longer-term outlook for timing, bolstered by a strong and growing opportunity pipeline and portfolio expansion into new markets.

During the quarter, we introduced 3 new timing products, including high-performance clocks for data center and wireless infrastructure applications, further expanding our SAM and revenue growth opportunities.

As carriers lay the groundwork for 5G and transition to Ethernet-based fronthaul networks to increase capacity, they are deploying equipment at the edge of the network where cost, power and size constraints present unique design challenges. To address this need, in Q3, we introduced a new wireless clock family, which is 55% lower power and 70% smaller than competing solutions. These products represent the first timing ICs to combine 4G/LTE and Ethernet clocking in the same device, dramatically simplifying clock generation for heterogeneous networking equipment.

To meet ramping demand for more bandwidth and faster data rates, Silicon Labs also introduced a new family of high-performance clock generators, offering the industry's most integrated, lowest-power timing solution for 10-,25- and 100-gig applications, using up to 60% less power than competing products. These new clocks provide unprecedented feature flexibility as well as all essential timing capabilities in a single-chip device, enabling developers to simplify their clock tree designs without compromising performance. Silicon Labs' new clocks are ideal for demanding applications, including hyperscale data center switches, servers, storage, networking and small cells.

And finally, we introduced a new PCI Express product family, enabling developers to maximize jitter margin and eliminate stand-alone clock buffers, reducing development risk and complexity for data center applications. Silicon Labs now serves the clocking needs of the entire universe of PCIe applications, from servers and storage to industrial and consumer applications.

Our Broadcast products delivered strong performance in the third quarter, posting a 10-quarter high in consumer product revenue and delivering record revenue in our automotive products.

According to IHS, the global automotive market will reach 94 million cars in 2017, with infotainment systems ranging from entry-level to high-end. To address this market opportunity, in Q3 we introduced the industry's most scalable, flexible and cost-effective car radio solutions. Silicon Labs Global and Double Eagle AM/FM receivers and tuners and Digital Falcon coprocessors enable automakers and OEMs to deliver radio designs ranging from low-cost, single-tuner AM/FM radios to high-performance systems with multiple tuners, digital radio reception and antenna diversity.

Looking across Silicon Labs' 21-year history, we've established a strong track record of tackling difficult design challenges and disrupting markets while leveraging our strength and integration to deliver breakthrough solutions. We've achieved market leadership by driving simplicity throughout our entire hardware and software portfolio and delivering solutions to help our customers reduce costs, complexity, power and size in their end products.

Our portfolio is well-positioned in high-quality growth markets, including IoT, Infrastructure, green energy and data communications, where we generate more than 70% of our total revenue, offering a long runway for growth and share gains.

We are on a mission to develop groundbreaking technologies and products that transform industries, grow businesses and improve lives. Our life's work is to create the silicon, software and solutions that enable the connection of things, information, networks and people everywhere. The opportunities before us for growth are exciting and enormous, and we're up to the challenge.

Thank you for your time and attention. Before we take your questions, I'd like to turn the call back to Jalene. Jalene?

Thank you, Tyson. Before we open the call for our question-and-answer session, I would like to announce conferences that we will participate in during the fourth quarter, including the Crédit Suisse 21st Annual Technology Media and Telecom Conference in Scottsdale on November 29, and Barclays Global Technology, Media and Telecommunications Conference in San Francisco on December 7.

We would now like to open the call up for your questions. (Operator Instructions) Marcella?

(Operator Instructions) Your first question comes from the line of Tore Svanberg.

This is Jeremy Kwan calling for Tore. Tyson, I was wondering if you could give us a little bit more detail in terms of the wireless breakout. Last quarter, you mentioned -- you gave additional color on mesh versus proprietary versus Bluetooth. Could you just give us an update, if there are any changes you expect in the near term and potentially which of the three you might see having the largest revenue from a potential revenue standpoint?

Yes, Jeremy. This is Tyson. In terms of the wireless products, we're seeing growth across the board in proprietary, in 15.4, the mesh networking and also in Bluetooth as well as a lot of applications that combine multiple technologies into a single device, where we're running multiple protocols; for instance, where you can pair up a device with a Bluetooth phone, get it connected to a network and then it switches over onto a mesh network, for instance, or onto a proprietary network. And we're also working in areas where we can actually run those protocols at the same time where you can access an end-node device with your phone as well as having it connected to a larger network. In terms of the market opportunity, we see tremendous traction with our mesh networking devices. You see the deployment of these in gateways and routers and into a number of ecosystems. And we're seeing strong lighting traction, metering, home automation, home security, industrial applications in that area. And I would say that in terms of our near-term revenue potential, the mesh products are seeing very, very good traction, followed by the proprietary applications. And then Bluetooth, both in stand-alone as well as in multi-protocol applications, is also seeing strong growth. So it's kind of across the board.

Great. And in terms of a follow-up, is there -- do you see any significant variation in terms of the gross margin profile between the 3?

Overall, our gross margins in the wireless area have some mix. We sell a number of solutions and modules, where we see an overall lower gross margin percentage content, although the gross margin dollars, in general, goes up when you move from an onboard design in -- onto a module. But overall -- so strong module sales would tend to pull it down slightly. And certainly in a -- in the mesh networking and in the proprietary areas, we enjoy strong margins. The Bluetooth area is a little bit more competitive on a stand-alone basis. But then when you combine those protocols in with the others, we are delivering a higher level of differentiation, and those generally go for a higher dollar amount as you include additional content.

Your next question comes from the line of Anil Doradla from William Blair.

So Tyson, clearly you're in a sweet spot, IoT and other -- many other multiple factors. But when you say scaling is a challenge, you referred to the sales channel. Is it limited to that? Are there some other challenges in terms of scaling? And can you build up a little bit more on that sales challenge and channel challenge?

Yes. So if you look at IoT, it's a very broad market. You're selling into thousands of different applications and across tens of thousands of customers. And that really presents two challenges: First, in the R&D area, you have to be able to deliver a differentiated portfolio that responds to those needs across those different applications. So you need to have multiple products that have functionality that is both differentiated for each one of those applications, while at the same time, cost-effective. They may have different amounts of memory, different amounts of peripherals, sensor interfaces and such. And so we use a platform approach there to drive efficiency across that. And as we've gotten out our next-generation devices, we continue to enhance our differentiation, and we continue to build on the R&D side to achieve that scale, both on the hardware and on the software side. So essentially, we can address that entire market opportunity with a common platform, where we're not creating specialized solutions for each individual application, which would really be not scalable. So we're really trying to achieve scale on the R&D side. Specifically around sales, we see -- again, reaching that tens of thousands of customers and being able to drive support, you have to make these solutions simple to use. You have to do a lot of training. You have to build the sales force. We have to have application engineers. And we try to make that, again, as scalable and self-serve as possible. We're introducing a number of web technologies and community support -- self-support so that customers can find the information that they need without us having to hold their hand. You can also look at modules as a way of simplifying the support of designs, where you have to have antennas and RF design that can create a lot of support challenge. And especially in the broader market, having a module makes it much easier for customers to take those. They're pre certified and streamlines our support needs as well. So just in general, it's both scaling on the R&D side, scaling on the SG&A side. We see a lot of customer demand and a lot -- the biggest challenge for us is to prioritize within that demand and also to be adding resources both within the channel, within our support infrastructure, to be able to address that opportunity.

Great. And as a quick follow-up, the optical softness, you talked about it coming back. Can you tell, when do you think it will come back?

Yes, Anil. This is John. This year has been impacted by what we believe is a general global slowdown in the optical networking market. We think that's most pronounced in China. And given order lead times, which really have not changed, we do not have direct visibility into the recovery there. That said, we do see timing as a long-term growth driver for the company. We're introducing new products and continuing to invest in this area and continue to see a long-term strategic growth target of 10% for the Infrastructure product category, of which timing is a key part of that.

Your next question comes from the line of Matt Ramsay from Canaccord Genuity.

Tyson, this might be a bit specific. But in the IoT market, I wonder if you might expand on some of the opportunities and the partners you have in the connected lighting space. It's a market that my team's been doing a little bit of work on and I think a market that's set for an uptick here. And just want to know how you're aligned in that market, what the product portfolio and ASPs, margins, those types of things might look like and who the partners are that you're working with. If you're able to talk about that in detail, that would be helpful.

Thank you for the question, Matt. We view lighting as one of the most exciting opportunities within IoT. There's a lot of benefits for connecting lighting up to applications and to the cloud to deliver additional functionality and features, and -- everything from security to adjusting colors, to automating timing schedules, to saving energy. And we're starting to see a lot of development in this area. You also have a number of ecosystems such as the Amazon ecosystem with Echo and a number of -- on the consumer side, you also see a lot of action on the industrial and commercial lighting side as well. This is an area that we've been taking seriously, I mean -- and Zigbee is a -- the mesh networking is a -- is really the appropriate networking technology to use in light bulbs, and it's been -- it's had a presence there. But having these in a mesh network as opposed to a star network is -- has a lot of advantages in terms of robustness and response time. And Zigbee is, in particular, well-suited for that. So we've been optimizing our software, been optimizing our hardware for this market. It is a very high-volume market and requires a more optimized solution so that you optimize the cost, in order to hold the margins there where we -- where they need to be. But we believe that we're well-positioned with a number of the leading suppliers in the timing market. We haven't made specific announcements around the -- a number of those opportunities. Singled is one that we did a press release on recently, but there are multiple others that are adopting these technologies and deploying across the ecosystem.

John, as a follow-up, on the operating expense levels, I think you guys had originally guided for 3% growth this year, and it looks like you're going to be around 2.5% or so. And with the momentum in the top line of the business, the op margin starts to get into the low mid-20s here pretty quickly. Maybe you could talk about how you're seeing expense growth. Too early to guide for next year, but just how you're seeing that trending going forward here and how you'd think about modeling that?

Yes, you bet, Matt. So if you kind of go back a few years, we undertook an investment cycle in the business in the kind of 2013 to 2015 time frame and saw the op margins dip down a little bit. We've spent the last couple of years exercising high discipline on spending and hiring and -- with the goal of getting the op margins back into our model range. We are achieving that goal today. And looking forward, as we kind of enter the next phase of growth and scaling in the business, we do see the need to pick up the investment level and anticipate increased hiring into Silicon Labs heading into next year. So we'll have more color to provide on the January call. But as we're looking at the annual operating plan cycle right now, we are looking at increasing the level of staffing to help us scale further. And so as you look at the operating margin results that we're seeing, it would be more towards the lower end of this overall range and kind of holding where we are now and then letting the revenue growth govern how much resources we can add to [this story]. We have a lot of opportunity, as Tyson indicated, and we want to be well-positioned to continue to lead in the growth markets that we're participating in.

Your next question comes from the line of Blayne Curtis from Barclays.

Tyson, I wanted to ask you about seasonality into March, broad-based. You talked about your business, 50% industrial, a lot of industrial businesses are up. Maybe things like home automation would be down. I think you've been up the last couple of years. I'm just kind of curious, your thoughts as you look into March, kind of what the kind of normal seasonality will be for broad-based.

So the main seasonality that we see going into Q1 is really the consumer products. So if you look at the Broadcast home, we had a very strong quarter in Q3. That's typically the seasonally highest quarter, although the seasonality is not as strong as it used to be. The television market, for instance, is not as seasonal as some of the other consumer markets. And then there's a -- there's some consumer seasonality concentration within IoT, although that is a fairly small component of the business. The majority of that IoT revenue is in industrial, where you have home automation. There is some Christmas cycle in there, but it is mostly Industrial where we see really not as much seasonality in the IoT. And then you layer on top of that the growth that we're seeing in the business. That business is growing 20-plus percent per year, and we -- our goal is to try to outgrow the seasonality within the IoT category.

And Blayne, this is John, I just want to quickly add on the seasonal topic. I will remind everyone that there is a seasonal aspect to the OpEx as we reset the payroll taxes and other employee benefit-related cost. So please anticipate an uptick in OpEx for first quarter.

Got you. And then actually, my second, just to follow up on that, John. You mentioned that OpEx is lower, some missed tape-outs. But then with the OpEx flat in December, do those come back in March as well?

That activity will be completed in Q4, so the flatness is more indicative of just the ongoing employee benefit and tax-related roll-off throughout the course of the year. But the first quarter effect would be more around hiring and increased seasonal payroll tax-related costs.

Your next question comes from the line of Cody Acree from Drexel Hamilton.

Can you just talk about the health, maybe Tyson, of what you're seeing in the inventory channels, just your order linearity through the quarter [bit]? Been a lot of investor concern about possibly seeing a bit of overheating in the distribution channel, and just curious to see what you're seeing.

Sure, Cody. This is John. Overall, we think the channel, both from a distribution perspective as well as our overall inventory levels, are well-positioned to continue to drive growth. We think the levels of inventory we have on hand and in distribution are healthy and normal. You did see a little bit of an uptick in third quarter in disti but that's appropriate given the kind of growth that we foresee coming ahead. So we're pretty comfortable with that as it relates to Silicon Labs.

And then a little longer term, with your midpoint guidance for the December quarter, it looks like if December works out, you're going to meet your double-digit growth target for the full year. When you look for it at 2018, do you need optical to come back? Do you need to start to see some material improvements in that Chinese optical infrastructure market for you to get back into a -- or to maintain the double-digit growth rate next year?

So if you look at the year that we've had here in 2017 with the softness in the optical market, if we had hit our targets for the year, we would have really -- it really would have had a nice impact on the results that we've been able to report. That being said, the largest component of our revenue is now IoT. And that now -- right at 50% and the fastest-growing business that we have and the stated confidence that we talked about in the script of having a 20% target on that over the mid, if not the long term, that really -- if we can execute on that and continue to execute within the Infrastructure category both in timing, but even in the absence of timing with all of the isolation products. And then really, the other factor there is being able to hold onto our business within Broadcast. We had a very strong year in Broadcast this year. We shipped 55 million TV tuners in the third quarter and have really become the de facto standard for TV tuners. And there's going to be ASP declines there. We're going to have continued move into the China market, where content is lower, but that -- there's a growing share that we have in China and in the over-the-top boxes. But we've got a 10% long-term decline within Broadcast and in Access in terms of our overall model. And so -- it really depends on how that comes out -- and also on timing as to where we are on the overall being within that model. That being said, if you look at the strategic growth areas, both Infrastructure and IoT, now you add those up and it's over 70% of our revenue, so it's a larger fraction of the revenue that we have that's growing. And so that gives us more confidence in the -- being able to hit that long-term strategic growth target in 2018 and going forward.

Your next question comes from the line of Suji Desilva from Roth Capital.

On the multi-protocol products, I know you're focusing on trying to do this simultaneously, running 2 protocols. Are those products already available? Or when will those be available? And what are the apps that are pulling for that demand, to be able to do that with a sufficient power and cost envelope?

So those products are out, both in terms of silicon and in terms of software. We have a number of customers running networks with Bluetooth and Zigbee, and a number of products that are using -- where you boot into Bluetooth to be able to connect to a phone and then position the -- put that device onto the network. That requires a little bit higher functionality and high -- really more memory on the part to be able to run multiple protocols at the same time. If you look at the markets, it's really a lot of different markets that -- really, anything that's using the mesh or using the proprietary protocols. Those like to have multiple protocol. And to be able to talk to the phones, lighting, in particular, is one where we see a particular interest there, balanced by the fact that that's also a very cost-sensitive area. So as we drive the road map forward both on software and on hardware, drive costs, where Moore's law takes the memory to be lower-cost and more cost-effective over time, we see this as a growing trend.

Great. And then on the balance sheet. In terms of the cash, you're doing a great job generating cash here. Can you talk about your thoughts on use of cash here generally and more specifically, the acquisition environment and whether you think your organic portfolio is what you want to go with the next several quarters, or whether any product holes are looking -- are there in terms of acquisitions that you're thinking of?

You bet, Suji. This is John. So we are pleased with the cash generation in the business thus far this year and continue to see the share repurchase program as an area that we will continue to be opportunistic about as a deployment of capital for shareholder return. In terms of inorganic growth drivers in the business, we continue to look for opportunities where we can find targets that will accelerate our road map or add some important adjacent capability. We don't have (technical difficulty) right now, but we do continue to work in that area as well.

Your next question from -- comes from the line of Craig Ellis from B. Riley.

I wanted to start with a longer-term question, just tying two things together. Tyson, in your discussion of new product activity and timing, it seems like there's a significant amount of new product flow in the third quarter. So -- I think we're all aware that end demand has its ebbs and flows, but how do you think that business is positioned for the next wave of demand strength and -- in broader communication infrastructure? And you talked about very good positioning with respect to 20% year-on-year IoT growth. What does that mean for the company's goal to get to $1 billion in sales in the 2020 time frame?

So let me first address the timing opportunity. We believe that we're quite well-positioned both within the core optical networking area where we've traditionally enjoyed strength as well as expanding into more data center and wireless infrastructure applications. So if you look at the product introductions that we announced in Q3, we're really optimizing a number of our solutions for those markets, for the 5G rollout, which we view as a substantial expansion of our SAM, and introducing products that are optimized for that market, as well as data center applications, including PCI Express and other clock and oscillator-type products. So that -- we feel confident about our technology there, our competitiveness and our ability to gain share in these adjacent -- or markets outside of our core market in optical, despite the slowdown that we're seeing right now in the overall market. And we're not the only ones. There's a number of companies out there that are talking about this within China and within additional suppliers here in the U.S. If you look at the goal of hitting a $1 billion run rate, and -- the thesis is that if we continue the IoT growth and continue the organic growth there at 20% or greater, and -- looks like we're going to hit that here this year, as well as the growth that we're seeing within the Infrastructure area. We had our 11th consecutive record revenue quarter in isolation, and those products continue to see broad adoption across a number of different markets and a continued growth in the timing area, and those products now 70% of the total -- you run that math out, and we're going to certainly be at that $1 billion level if the markets hold up here over the next couple of years. And then you layer on top of that, we have $724 million, John, of cash in the bank, and we continue to look for opportunities to expand on top of those businesses with some M&A as well. So that could accelerate the time frame, depending on what we can pull in there.

And then the follow-up is for John. Just two clarifications: One, very helpful to get the industrial sales mix at 50% of total. Can you break out the other end-market exposures? And then on the accounting change, thanks for the preview in terms of how you'll handle things. Just clarifying that, that means there won't be any revenue or gross margin perturbations as you cut over to the modified sell-in basis next year.

You bet, Craig. So looking at the end market, rough numbers there. The automotive is still running around 8% of our mix, and that is predominantly the automotive radio tuner. We do sell isolation products into automotive applications as well as some of our MCU products into automotive. But most of that is on the broadcast radio side, really not much in computing. And then the balance of the mix is roughly split between the communications and consumer end markets for the end market topic. As far as the accounting change, really the key there is that, going forward, our revenue results will be more sensitive to distributor inventory levels under the new guidance, that's effective next year. We don't anticipate margin changes or impacts to a material extent. And with stable levels of inventory in the distribution channel, we would not expect material effects on the top line, but it will be more sensitive to the inventory level.

Your next question comes from the line of Raji Gill from Needham & Company.

This is Robert Mertens on behalf of Raji. Just to go back to your Broadcast business. I know this is seasonally stronger this quarter. But just trying to get an idea of long term, if you're still thinking of around negative 10% year-over-year decline or if you think there could be any potential upside with the strength in the automotive business there.

So if you look at the performance of our Broadcast products over the last couple of years, we've been able to really grow our share within the consumer spaces in both audio and video, and really are the de facto standard for TV tuners across the industry. We enjoy about 2/3 market share in the flat-panel TVs with a number of the Tier 1 players as well as China. And there's also opportunity within the set-top box area. This is -- they're -- that business in that market is not necessarily growing, and we do see ASP declines over time as well as some shift in content from high-end type devices into lower-end devices. So that business will continue -- will be down next year, depending on the markets and depending on our market share. But overall, that trend continues in a downward trend. And that's balanced by share gains and growth in automotive, although that is modest given the product life cycles. These are high-quality, long-lived revenue, but it also takes time to ramp those into production as you ramp different models and different geographies within the automotive radio area. So net-net, the automotive and the consumer parts of Broadcast, we still stand behind that 10% revenue decline year-on-year. Could be more than that, depending on our ability to maintain market share within the TV market.

Great. And then as a follow-up for the Infrastructure business, could you give a sense of what the growth rate is between timing and isolators? I mean, perhaps more of a 2018 and beyond without the weakness in China, if you think about one growing more than the other, just based on your design pipeline of products.

So if you look at Infrastructure, our strategic growth target there is about 10% -- or is greater than 10%. We will underperform that this year. If you look year-on-year, the Infrastructure was only up 1% in Q3, and that was really driven by strength in the isolation area, offset by weakness in timing. So the Infrastructure continues to deliver across the board in multiple market segments -- in automotive, in a lot of industrial applications and green energy applications, and -- but -- and timing has offset that. So we have underperformed our strategic growth target there, although we are confident going forward that the -- that 10% growth target is a reasonable one to take for the Infrastructure, given our investment levels and continued product momentum there in that market.

Yes, Robert. This is John. I mean, we do not have unique goals for timing and isolation. The goal is really for the overall Infrastructure category. But looking at timing, I mean, we see areas to really enhance our SAM, moving into wireless infrastructure, as Tyson indicated on the call. So we do still see longer-term gross -- growth prospects for timing. In terms of the overall model for the company, you've got IoT at 20%. You've got Infrastructure at 10%, and then Broadcast and Access at a 10% decline. So with Broadcast and Access at essentially 30% of revenue or so, that's about a 3% headwind on our growth, offset by much stronger growth in the IoT and the Infrastructure area. So overall, our growth businesses are much larger than our more mature businesses, and we see that over time becoming even more prominent.

I would now like to hand the call back over to Jalene Hoover.

Thank you, Marcella, and thank you all for joining us this morning. This concludes today's call.

Okay.