FormFactor Inc (FORM) 2017 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the FormFactor Inc. Fourth Quarter Financial Results Conference Call. (Operator Instructions) As a reminder, this call is being recorded. I would now like to turn the conference over to Jason Cohen, General Counsel. Sir, you may begin.

Thank you. Today, the company will be discussing GAAP P&L results and some important non-GAAP results intended to supplement your understanding of the company's financials. Reconciliations of GAAP to non-GAAP measures and other financial information are available in the press releases issued today by the company and on the Investor Relations section of our website.

Today's discussion contains forward-looking statements within the meaning of the federal securities laws. Examples of such forward-looking statements include those with respect to the anticipated effects and benefits of the completed merger between FormFactor and Cascade Microtech; projections of financial and business performance; future macroeconomic conditions; business momentum; business seasonality; the anticipated demand for products; customer requirements; our future ability to produce and sell products; the development of future products and technologies; and the assumptions upon which such statements are based. These statements are subject to known and unknown risks and uncertainties that could cause actual results to differ materially from those expressed during this call.

Information on risk factors and uncertainties is contained in our most recent filing on Form 10-K with the SEC for the fiscal year ended 2016 and our other SEC filings, which are available on the SEC's website at www.sec.gov and in our press release issued today. Forward-looking statements are made as of today, February 7, 2018, and we assume no obligation to update them.

With that, we will now turn the call over to FormFactor's CEO, Mike Slessor.

Thanks, Jason, and thank you, everyone, for joining us today. I'd like to start by spending a few moments reviewing the significant progress FormFactor made in 2017. Our worldwide team produced great results throughout the income statement, growing revenue by over 40% to an all-time record, demonstrating the leverage inherent in our operating model by more than doubling non-GAAP earnings per share and generating more than $0.50 of GAAP earnings per share.

From a business perspective, we further diversified and increased our served customer base, highlighted by the adoption of our differentiated MEMS-probe technology for testing of a major 10-nanometer integrated fan-out packaged mobile processor. This outstanding performance is the result for the scale of diversification we have attained by growing our businesses and successfully integrating FormFactor and Cascade Microtech. And with this performance, we demonstrated the first steps towards achievement of our targeted financial model.

For the fourth quarter of 2017, revenue of approximately $132 million exceeded the midpoint of our guidance. The gross margins were below expectations. While we are disappointed to have missed the midpoint of our fourth quarter gross margin guidance by over 2 percentage points, we are pleased the continuing operating expense discipline enabled us to meet the lower end of our non-GAAP earnings per share guidance range.

Several factors contributed to the gross margin miss, with the most significant being gross margins in our Systems segment that were nearly 10 percentage points below recent levels.

In the Systems business, we determined in Q4 that we could see future growth and gain competitive advantage in the emerging silicon photonics area, by placing subsidized penetration systems at 2 industry research consortia and at leading manufacturers.

As an example, one of these placements is in the new tool that enables the next phase of silicon photonics test development at Imec, advancing some initial work that was recently published jointly with Imec in Photonics Online.

These placements are part of the strategy we have described before to diversify our footprint beyond pure-play semiconductor applications into new areas of electrical-tested measurement such as silicon photonics and micro-LEDs and represent important long-term diversification and growth opportunities.

Turning to the present. Our view of the first half of 2018 remains unchanged from the commentary we provided in January at an investor conference, where we commented that first quarter revenues will decrease significantly from the fourth quarter of 2017.

As we described at that time, this reduction in revenue is due to a one-quarter delay in 10-nanometer production probe card demand from our largest customer.

I'd now like to provide further insight into the situation with our largest customer, a situation that we are very actively monitoring and managing, given that our lead times are approximately half a quarter.

At present, the available data continue to indicate that probe card shipments will begin to ramp in the second quarter to support their 10-nanometer node, with that same data indicating record shipments to this customer in the second half of the year.

Given the scale and importance of this customer to FormFactor and the recent dynamic demand changes, we are working closely with them to ensure clear visibility into the details of demand and capacity plans using multiple sources, such as customer forecast and discussions as well as independent indications such as process equipment installations.

Based on this, we have taken steps to reduce our short-term manufacturing capacity for this customer to minimize the cost impact of underutilized assets. We are balancing the need to match our capacities with the reduced short-term demand from this customer, while retaining the flexibility to ramp back up quickly.

There were multiple lessons learned from our last major ramp with this customer in the first half of 2016, and we've incorporated those learnings into our approach to this situation.

With the exception of the pushout and deliveries from our largest customer, our Foundry and Logic probe card business is performing at solid levels of demand to begin the year, with strong new design activity serving a variety of end market applications. The intersection of mobile processors and advanced packaging continues to be a key driver for FormFactor, as we execute on the initial ramp of a major 7-nanometer device with integrated fan-out packaging.

In addition, we successfully penetrated an important fabless customer for a 10-nanometer application processor in the China semiconductor ecosystem. This customer transitioned to a node and packaging density that required FormFactor's MEMS-probe technology, a theme that continues to play out across the industry.

In the RF space, initial engineering and prototype RF probe card activity for 5G applications is robust, paving the way for future growth in our Foundry and Logic probe card business. Even with these strong indicators and our long-term excitement for our mobile opportunity, we remain alert for any short-term changes, especially with the recent news of slower-than-expected handset demand and increasing supply chain inventory.

Our DRAM probe card demand business continues to perform well during what is typically the seasonally weak part of the year, with each of the major DRAM manufacturers on a slightly different cadence of node shrinks and new design introductions on existing nodes.

As you've recently heard from various other players in the DRAM supply chain, planned capacity additions, either through shrinks or new facilities, appear to be moderate enough that they will not offset the strong demand for mobile and server DRAM. We expect our market share leadership in the DRAM probe card space to allow us to continue to benefit from this healthy end market condition.

While we are disappointed about the sequential decrease in quarterly revenues to start 2018, we remain confident that this will be another growth year for FormFactor as we continue to realize our line-of-sight opportunities in advanced packaging, mobile data and automotive applications. In addition, we expect to continue to make progress towards our target financial model to deliver $650 million of revenue and $1.50 of non-GAAP earnings per share.

Finally, as you saw in our press release from earlier today, Mike Ludwig has announced his retirement as our CFO effective March 2. On behalf of all our employees and shareholders, I want to take this opportunity to thank Mike for his leadership and contribution to FormFactor's evolution and growth over the last 7 years. On a personal note, I'd also like to thank Mike for his partnership during my 5-plus years at FormFactor, and I wish him the best in the future.

A search for a replacement CFO is underway, and I'm pleased that Mike will be assisting us after March 2 to help us execute an orderly transition to a new CFO.

With that, I'll turn the call over to Mike for further details on our fourth quarter results and to provide further insight into FormFactor's guidance and financial outlook.

Thank you, Mike, and good afternoon. As Mike mentioned, I will be retiring from FormFactor on March 2. I've been privileged to serve as CFO the past 7 years and work as part of a team with so many outstanding, dedicated employees to transform the company into a profitable entity, taking advantage of several demand drivers shaping the test and measurement ecosystem.

I would like to thank the Board of Directors, Tom St. Dennis and Mike Slessor for allowing me to serve in this capacity, and to also say thanks to those of you in the financial community that I have worked with over the years.

2017 results were a proof point of the benefits of successfully executing the inorganic growth and diversification elements of our strategy. We grew revenue 43%; increased non-GAAP operating profit margin by 8 percentage points to 17.7%; and increased non-GAAP diluted EPS by 147% to $1.21 per share.

In addition, we achieved GAAP EPS of $0.55 per fully diluted share and generated over $72 million of free cash flow.

As you saw from our press release and heard from Mike's comments, we generated mixed results in the fourth quarter, with revenues above the midpoint of our guidance, a gross margin below our guidance range. Good operating cost control allowed us to deliver non-GAAP EPS at the low end of our guidance.

FormFactor's revenues for the fourth quarter were $131.9 million, down 8% from the third quarter. Probe card segment revenues of $107.2 million decreased 10% compared to the third quarter, while Systems segment revenues of $24.7 million increased slightly compared to Q3.

Within the probe card segment, Foundry and Logic revenues of $68.7 million decreased 16% compared to our third quarter. The decline was attributable to an anticipated revenue reduction from our largest customer, as they delayed the ramp of their next technology node while digesting prior quarter record probe card shipments. On a more positive note, during the quarter, we began shipments of a new design to a foundry who utilizes our probe cards in advanced packaging applications and leading-edge nodes and probe card for RF applications increased in the fourth quarter.

Foundry and Logic revenues comprised 52% of total company revenues in the fourth quarter, down from 57% in the third quarter.

DRAM revenues were $31.9 million for the fourth quarter, a decrease of $0.5 million sequentially.

While the results were down slightly, overall, we continued to see a robust demand environment in the quarter as technology node transitions and a strong data center demand environment continued to positively influence probe card demand.

DRAM revenues comprised 24% of total company revenues in the quarter, a slight increase from the third quarter.

Flash probe card revenues of $6.6 million were $1.5 million higher than the third quarter, with over half of the Flash revenues in the fourth quarter for NAND Flash applications. The slight sequential increase in Systems segment revenues was driven by increased revenues from 150 and 200-millimeter platforms.

GAAP gross margin for the fourth quarter was $48.6 million or 36.9% of revenues compared to 40.1% for the third quarter.

The fourth quarter included $6.5 million of reconciling items, which you can find outlined in our GAAP to non-GAAP reconciliation table available on the Investor Relations section of our website.

On a non-GAAP basis, gross margin for the fourth quarter was $55.2 million or 41.8% of revenues, down from 44.5% in the third quarter, missing the midpoint of our guidance by over 2%.

The decrease was primarily due to a 9.2 percentage point decrease in the Systems segment gross margin to 42.5% in the fourth quarter.

There were a couple of principal causes for the sharp decline. As Mike mentioned earlier, one significant cause of the decrease were the strategic investments in the quarter to establish our technology for emerging silicon photonics applications.

In addition, we experienced a less favorable mix of thermal subsystems sold, and our margins were negatively impacted by a continuing stronger euro.

While several of these strategic opportunities were being discussed with the institutions and customers as we entered the quarter, the timing of the placements in subsequent revenue recognition was uncertain. During the fourth quarter, we made decisions to take the actions to allow us to recognize the revenues and related costs to ensure we did not lose the revenue in the transition to the new revenue accounting standard.

Given the onetime nature of the investments in some silicon photonics applications, we expect our Systems non-GAAP gross margin percentage to return to the high 40s to low 50s in the first half of 2018.

Our Probe segment non-GAAP gross margin was 41.7% in the fourth quarter, a decline of 1.4 percentage points compared to Q3, due primarily to less Foundry and Logic revenues and lower fixed cost absorption, resulting from lower volume in the fourth quarter.

Our GAAP operating expenses were $43.2 million for the fourth quarter, $1.2 million less than the third quarter. The fourth quarter included $7.2 million of GAAP to non-GAAP reconciling items.

Non-GAAP operating expenses for the fourth quarter were $36 million or 27.3% of revenues, a $0.9 million decline compared to the third quarter.

Our non-GAAP operating expenses declined in the fourth quarter due to the lower performance-based compensation.

The non-GAAP effective tax rate for the fourth quarter was 1.1% compared to 4% in the third quarter. The lower effective tax rate was due to the reporting of 2017 refundable AMT credits in the fourth quarter. Our non-GAAP effective tax rate for fiscal 2017 was 3.5%.

We ended 2017 with approximately $244 million versus remaining U.S.-based net operating losses. As such, we expect to have a non-GAAP effective tax rate of 4% to 6% while we utilize our net operating losses.

Currently, we do not expect the new tax legislation to have a material impact on our non-GAAP effective tax rate for fiscal 2018.

GAAP net income was $5.6 million or $0.07 per fully diluted share for the fourth quarter compared to net income of $0.17 per fully diluted share for the third quarter. We were GAAP profitable in all quarters of 2017.

Fourth quarter non-GAAP net income was $18 million or $0.24 per fully diluted share compared to $25 million or $0.34 per fully diluted share for Q3.

Company noncash expenses for the fourth quarter included $7.5 million for the amortization of intangible assets, $5.1 million for stock-based compensation and depreciation of $3.6 million.

Moving on to the balance sheet. The company generated $23.5 million of free cash flow in the fourth quarter, bringing our free cash flow for fiscal 2017 to $72.4 million.

In the quarter, the company spent $9.5 million on principal and interest payments and $3.8 million for capital expenditures.

Capital expenditures for fiscal 2017 were $17.8 million, within our expected annual capital spending of $16 million to $20 million.

With proceeds from employee stock plans, we bought back $8 million of our stock in the fourth quarter and $19 million for fiscal 2017 against an authorization of $25 million.

In total, cash comprised of cash, short-term investments and restricted cash ended the fourth quarter at $141.7 million, $6.1 million higher than Q3.

The balance of our cash, short-term investments and restricted cash exceeded the balance of our debt by $36 million at quarter end.

Turning to the first quarter non-GAAP guidance. We continued to experience broad-based demand across product lines and markets in both our probe card and Systems segments. As Mike discussed, however, we are currently experiencing a delay in 10-nanometer production card demand from our largest customer that will weigh on our first quarter revenues.

Therefore, we expect revenues in Q1 to be in the range of $112 million to $120 million.

Consistent with the reduced revenues, the first quarter volumes and factory utilization are expected to be lower than the fourth quarter. While we will continue to actively manage our fixed cost in a lower volume manufacturing environment, we need to balance our cost management with the expectation of increased volumes from our largest customer in the second half of 2018.

As such, we currently expect first quarter gross margins to be in the range of 40% to 43%.

We expect to realize fully diluted earnings in the range of $0.12 to $0.18 per share.

Our Q1 non-GAAP guidance assumes consistent foreign currency rates.

With that, let's open the call to questions. Operator?

(Operator Instructions) Our first question comes from Tom Diffely of D.A. Davidson.

First of all, I'd like to say farewell to Mike Ludwig. It's been a pleasure working with you for 7 years, and you will be missed.

Thanks, Tom.

Now some hard financial questions. When you look at the NOLs, how many -- how much of the NOLs were used up in 2017? And then once they are used up, where would your tax rate be?

So we used probably -- we used approximately about $50 million of NOLs in 2017. We started the year at about $295 million of NOLs. And I think we said that we have about $244 million as we enter into 2018. So about $50 million. And once we utilize -- and we expect probably -- the forecast is obviously hard to say, but we think in the early 2020s is probably when we will utilize the remaining NOLs. And at that point in time, we would expect our effective or non-GAAP effective tax rate probably to be in the 20 -- mid-20s, I think, is about the right answer. Obviously, we're still studying that, but somewhere in the mid-20s right now makes sense, Tom.

Okay. And then, when you look at those low-margin Systems business. So it was a little unclear. Was that turns business in the quarter itself? Or did the pricing change in the quarter? I'm just curious, how it wasn't officially captured in the guidance going into the quarter.

Yes, Tom, there were a few opportunities that were turned in the quarter. As Mike Ludwig described in the prepared remarks, there was also one significant one where we accelerated revenue recognition because of the change in the standard. But I want to return to the penetration of that application represents one of the really significant growth areas for us along with micro-LED in our Systems business. And these are opportunities where we get to work with leading consortia like an Imec to place a system and then do the fundamental test development for these emerging applications. So we view them as things -- when the opportunities emerge, we jump on them. There were a couple as, again, Mike Ludwig said, that we had already shipped but accelerated the RevRec -- the revenue recognition, so that we didn't have any issues crossing the year boundary with the change in revenue recognition standards.

Okay. So this is for end markets like VCSELs or something like that where you expect pretty high volumes to utilize your technology?

Yes. I think VCSEL probably is [sub] example. Some of the other ones are things like different modulators and switches that are going to end up in data centers to drive higher data rates. So a variety of electro-optical applications, these ones pretty focused on the next generations of switches and modulators.

Okay. And then, finally, you talked about a new opportunity of the 7-nanometer node. When do you think that actually goes to production where it would give you some more scale?

So good question. So we highlighted that we were beginning the ramp of a 7 -- major 7-nanometer mobile application processor that's packaged with integrated fan-out advanced packaging. If you recall, a little over a year ago, that represented a big uplift to the middle of our year as we penetrated one of the last major foundries that FormFactor had not been a supplier to. So we see the same kind of theme happening here in 2018. I think the good news for FormFactor is we're in earlier, having established a good track record in the initial engagement last year. And we're reasonably confident we're going to be able to grow that business this year.

From a timing perspective, we are shipping some of those initial units as we speak in the first quarter, which is a little bit earlier than we were last year and expect that to ramp through the second quarter and have some volume even in the third quarter. Obviously, the tail end of that volume is pretty significantly driven by overall handset forecast. And obviously, there has been a bit of noise around that recently. But the front half of the ramp, where all the preparations in the supply chain are going to drive the capacity, are going full speed ahead at this point.

Our next question comes from David Duley of Steelhead.

I guess, the first is on your largest customer's 10-nanometer ramp. I guess, they pushed it out again as you've indicated. And I'm wondering, you talked about a strong second half. But how should we think about the second quarter? Will we see an increase in that business in Q2 and then a much bigger increase in the second half? Or will it remain flat and just spike in the second half?

Yes. So David, good question. I guess, a couple of points. One is I want to make it clear that I'm not making any comments about that customer's 10-nanometer ramp, but instead talking about their demand for production probe cards for us. So that's the first point to get across. Clearly, those 2 things are linked. But given the lead times we have of approximately half a quarter, obviously, we need to be very responsive to this, both on the upside and the downside. Now as we move to timing of this, as I said in the prepared remarks, we are monitoring and managing the situation very closely, especially given the recent changes that we've had, that we first started to update you in early January about. There have been no changes to that information in the trajectory. But in sum total, what it means is we expect a very light Q1 from that customer as evidenced from our guidance. The currently available information, both direct forecast, narratives around the forecast and triangulating things like process equipment installations, all indicate that this ramp will start in the second quarter. So it's not going to be a step-function up in the second quarter, but we see it building in the second quarter. And then again, those pieces of information are consistent with us getting to new record levels in the second half. But obviously, I want to reiterate for everyone, this is a situation that bears significant management and monitoring given that we've, from an efficiency standpoint, driven our lead times down to approximately half a quarter. So we can be responsive and reduce our costs. So the flip side of that is you don't have a lot of direct forecast visibility and fundamentally, as Tom mentioned on the first part of the call, we're running a turns business here.

I guess, one more question on this particular topic is, is there any other things that you can do as far as operationally to -- when customers -- large customers change their demand forecast on you, that you can mitigate the circumstances a bit more? Can you take deposits? Or is there anything else that you're contemplating to help you smooth this -- these events out?

Yes. I mean, I think the first action we continue to take and continue to drive towards is the diversification of our customer base and market drivers. We made good progress in that, obviously, with the acquisition of Cascade Microtech. With -- getting ourselves qualified at the world's largest foundry in 2017 and driving and growing that business adds another element of it. I think we want to be careful to understand the demand profiles associated with each of these customers and work with them on the different commitments to support their ramps. But we're in an industry that obviously has some very significant short-term flexibility required of it. And I think the solution for us is really to continue to grow and diversify our businesses, both organically and through M&A, so that we don't have any one customer that represents 30% of revenues that can cause such a material impact on a single quarter.

Okay. And then, on a different topic. I was wondering -- you've talked about this on previous calls. But maybe just kind of highlight for us how you feel your position in China for the upcoming memory ramp. And I guess, it's more important for you guys as obviously on the DRAM side. And then, are you -- could you share with us how many overall DRAM fabs you might be tracking the shelves are being built up or facilities that are being put in place that could be facilitized at a later date in DRAM, not just in China, but outside of China?

Okay. That -- so the latter one is probably a question I got to get back to you on. The major one that we're tracking that will impact capacity here in 2018 is one of the Korean facilities. But if I try and enumerate all of the China fabs, I'll get mixed up doing it live. So why don't we get back to you on that one?

Our position in China. So we have been a supplier with a significant China footprint for a little more than a decade now. It started off with some of the domestic U.S. manufacturers moving operations over there. And we were supplier to those people and had to offer local support and design and repair for probe cards. And so we've built on that footprint and taken advantage of the FormFactor worldwide brand, especially in DRAM. But we do see some opportunities in Flash as well. And the team in China working with our U.S. team has done a very nice job in capturing a lot of those initial designs. Our view is it's still very early innings in those production ramps. Obviously, the signal you'll see first is from some of the big equipment companies as they start to ship significant volumes into China. And in a situation like that, I'd expect probe card volume to follow 2 to 3 quarters later: one, because this is a bit of a new yield ramp for all of those China memory manufacturers, and so it's going to take them a while to get their yields fixed. But if you think about what a probe card is, it's a design-specific consumable tied to each design. And so customers are going to have to work through their yield learning, their yield ramps and then tie down the mass sets and the designs before we see significant probe card volume. All told, that probably puts it close to the end of this year at the inside, but we feel like we're in a solid position as the initial development work and initial pathfinding work starts.

Our next question comes from Craig Ellis of B. Riley.

And Mike Ludwig, good luck as you move on to other things. It's been a pleasure working with you and appreciate your help.

Thanks, Craig. Appreciate it.

Just want to start with gross margin and make sure I understand the different dynamics that play in the fourth quarter and the first quarter. So I think if I heard correctly and I got on the call a little late, is that there was a probe card-related issue, that was about 2/3 of the variance, which was 270 basis points in total. There was a Systems issue, that was about 35% of that. One, is that correct? And then two, did those issues resolve in the quarter? Or are they impacting the gross margin guidance that we see for the first quarter? And if so, to what degree relative to the issues that we're seeing with your largest customer?

Right. So first, let me address the Q4 gross margin. Yes, and I would actually say that the majority of that was from the Systems business, not really the probe cards business. And again, as we talked about on the call, it really was driven by probably a couple of things on the Systems side. One was that the placements of some -- and strategic investments that we made with both some institutions and some -- and customers regarding silicon photonics investments. So that was probably the major driver. And I think Mike spent some good time talking about that. That was probably the major driver from a Systems perspective. The other things impacting Systems would be the product mix with our thermal subsystem business and then again, to a lesser degree, but still some impact would be from the continuing strengthening of the euro. Now going forward, I would say that the issues in what we talked about is we do expect that the Systems business will get back to a high-40s to low-50s non-GAAP gross margin in the first half of 2018. So what I will tell you is that the strategic investments that we made in silicon photonics really were more onetime investments. Doesn't mean we wouldn't make similar investments somewhere down the road with other applications. But at this point in time, we don't see any of those opportunities impacting us at this point in time. So from a Systems standpoint, when we look at Q1, we see that getting back into the high 40s. And as we move throughout the second quarter, depending on the strengthening euro, could be even back into the low 50s. So nothing that we think really interrupts our financial target model as it relates to the Systems business.

On the probe card side, I would say it was a lesser degree in the fourth quarter. What really impacted the probe cards business was 2 things. A little less Foundry and Logic business. So as we've always talked, there are really kind of 3 areas that get impact or 3 items that impact the probes. One is mix, one is volume and the other one is manufacturing execution. In Q4, we had -- the mix was lower on Foundry and Logic, and it was low with respect to volume of transactions. Therefore -- or the volume going through the factory. Therefore, those were the 2 primary items impacting. And I think those 2 items will continue to impact us in Q1, primarily related to our largest customer.

Okay. So is it fair to say, as we look at the -- I think it's 5 items in total that we have there that those are significant. Your largest customer is going to be the predominant issue in 1Q. And then it's really the or 10-nanometer ramp that determines the degree to which that issue in 2Q and you mentioned the 46% target model. Can you just walk us through how we can think about some of the major milestones between where we will be in the first quarter and an intercept point with that 46% target level?

Yes, I think it really, again, gets back to probably 2 pieces. One is the mix. So I think as our largest customer, the volumes increase there, as we continued to gain more business, particularly in areas that Mike had talked about that have a favorable impact on Foundry and Logic business going forward. We think that mix will be -- will help margin. And then, secondly, again, the volumes -- the volumes coming up play a big impact on that. And those are probably the 2 biggest one. And of course, every quarter, right, it's about how we execute within our factory. So those are the 3 areas where to go from where we are in Q1 into our model. And again, I want to reiterate that as Mike said, I think we still feel very, very good about our financial model -- our target financial model and getting to 46% and $650 million revenue in the 2019 to 2020 period.

Okay. And then if I could just flip a couple over to Michael Slessor. 5G is very topical right now and RF probe card business within Foundry and Logic is an important business for the company. Can you just talk about some of the milestones we should looking at given that there are more and more reports that handsets that are 5G enabled are going to be coming in the first half of 2019? And then, it seemed to be a very solid and steady DRAM performance in the quarter. Any thoughts on DRAM as we look at the full year would be very helpful.

Thanks, Craig. Why don't we start with RF? I think as I mentioned in the prepared remarks, we're really excited about the initial activity we have in enabling these 5G handsets and the 5G infrastructure. Obviously, this is not really a material business right now, but it is a big set of R&D investments for us. And a big set of activity as we work with customers to essentially define the test protocols and standards that they're going to use, whether it's the RF front-end in the handsets themselves or some of the infrastructure in base stations, high-speed digital chips, things like that. So 5G for us, right now a lot of development activity, but working with some of the really key driver customers that are going to be enabling both the handset side and the infrastructure side. I think I agree with your timing that it's a 2019 event for us. But again, we're laying the groundwork today with those customers to make sure it's a business for us in 2019 or whenever 5G adoption really is.

On the DRAM side, I think the simplest characterization of it is it remains a overall healthy business for us and for our customers. As you've heard from virtually everyone who participates in the DRAM supply chain, it appears that capacity and demand are roughly in check. And the difficulty of node shrinks for our customers really doesn't leave you with too many ways to throw that balance [of that] check unless, obviously, the demand side of the equation reduces significantly for some reason. That doesn't look like it's going to happen in the short term. The other thing, our DRAM business, now getting more specific to FormFactor, is benefiting from is each of the major customers seem to be on slightly different design release and node-shrink cycles. And so where they all used to move as one and shrinking to the next node or releasing a new design, there is a little bit more of a smoothing of timing in that. And we're supporting each of the major 3 DRAM manufacturers and the initial projects in China I talked about, with a different emphasis on the activities. But as you can see from the fourth quarter results, DRAM continues to be a pretty strong probe card business for us even at 20%, 25% of revenue.

Our next question comes from Edgar Roesch of Sidoti & Company.

I add my congratulations to Mike Ludwig. It's been short, but it's been a pleasure.

Thanks, Ed.

So Mike Slessor, you mentioned still trying to grow revenues for the full year, I believe I understood that correctly?

Yes.

Okay. And so if I kind of plug in the Q1 midpoint and some revenue growth for the full year, I guess, I'm just trying to understand if Q2 is going to bounce back more strongly than I'd originally modeled it? Or you're going to be looking at some pretty big quarterly revenue run rates in the second half? You tapped out around $144 million or so last -- this past year, 2017. Could we be -- could you ship $150 million to $160 million in a quarter?

Yes. It's a great question, Ed. And obviously, the simple math with the guidance midpoint we've given for Q1 and then the implication of growing the year overall means that we're going to have some pretty significant quarters as we move through the back half of the year. I think that's predicated on a couple of things, quite obviously. The first is our largest customer essentially executing the demand that is currently reflected in the data we talked about, whether it's forecast data or other pieces of data. There are also some of the other growth opportunities we have in front of us. I just spent some time talking about DRAM. That looks like a pretty good opportunity in 2018 and beyond. And then, advanced packaging and things like automotive applications that we haven't talked about are really driving some secular growth in the business. And those things are unchanged despite the light Q1 guidance. So we view those businesses as growing throughout the year. But by far the biggest variable is the 10-nanometer ramp and demand from our largest customer. On the ability to produce those kind of levels that are implied by the math, we've gone through a planning process to figure out how we could do it. We don't have the capacity in place today. It's mostly direct labor related. And we believe we can bring that direct labor on in response to the volume, as we've done in the past in significantly growing different elements of the business.

Sounds good. And then, I guess, wondering about one question about the long-term model. I know it's a target, but the R&D at about 14% of revenue versus around 12.5% it seems for 2017. I was just wondering if you could speak to whether there are any areas of future investment already identified, where you'd be ramping up the R&D at a more aggressive rate or whether you happen to be getting maybe just some unexpected efficiency in your current R&D dollars. Maybe favorable revenue mix or higher ASPs are impacting that? If you have any color on that, that would be great.

Yes. You're right. So in the long-term model, we did call out that we felt like 14% -- R&D spend of 14% of revenue was where we wanted to guide the business long term. Obviously, that's going to have some fluctuations year-to-year and quarter-to-quarter. And I think 2017 clearly in growing revenues as we did, you can have a long R&D wish list, but you're not going to add costs as rapidly as we grew revenue to get to a 14% level.

I view R&D, especially from the market share leadership position that we have, as something that essentially gets guided in enabling customer road map and enabling industry road map. Things like the ability to keep pace with integrated fan-out packaging, with high-bandwidth memory, with some of the 5G RF opportunities we talked about earlier on the call, those are the places where we're going to spend our R&D. And we're confident we can fit within that 14% budget and are going to continue to make those investments, while balancing the operating cost discipline that we've shown in the past.

All right. You've done a great job there in many ways.

(Operator Instructions) Our next question comes from Edwin Mok of Needham & Company.

So first, my question really is on your largest customer. I just want to be really clear. You said that the second half you expect to have record level or a very high level. Are you talking on a quarterly basis or a full year basis? I'm just trying to understand what you're referencing to, just to be exact correct.

Yes. So good question, Edwin. So what we said and to clarify if we hadn't said it clearly enough. We obviously expect our largest customer to be at very low levels in Q1. All of the available information we have, whether it's direct forecast as we receive from them, whether it's different pieces of other planning data we triangulate, point to the quarters in the second half reaching record revenue levels. And if I point you back to Q3 of 2017, our largest customer was approximately 30% of $144 million quarter. So that ballparks you a little bit. I want to make sure though that everyone understands, and this is why we included it in the prepared remarks, that we're operating with lead times that are essentially half of the quarter. And we did that. We've done that in all of our businesses to improve our agility and flexibility, and it's been one of the big components of why we've gained market share over the past couple of years. We're able to respond to demand quickly. Obviously, also makes your internal operations much more efficient. So if I do the simple math, right, we are a long way from 6 weeks from the third quarter and 6 weeks from the start of that ramp. And so the specific visibility on POs being placed is not yet there. That's why we're making it clear to people that it's a situation we're actively managing and keeping in very close contact with the customer to make sure we can ramp up in time and reach those record levels when the demand comes in the third and fourth quarter.

Okay, that's helpful. That actually is kind of tied to my second question, right. So with that customer at a very low level, you mentioned that you have tried controlling your capacity, right. Obviously, not having that capacity, you're putting a big drag on your profitability, right? But given that it's almost like a turns business, how do you kind of flip that switch just like that? I mean, you mentioned that you have to hire some direct labor. So I want to understand how you can flip it in a few weeks from that pretty low level to above that 3Q level that you just saw? Can you kind of explain how do you go forward to that?

Yes. And if I take you back, I made a comment in the prepared remarks about the lessons learned from the first half of 2016, where we did have to flip that switch. And we didn't do it as fast as we needed to. It is, as you say, primarily labor driven. And we do have a pool of direct labor both locally and across some of our other businesses. And I realize it's a 0-sum gain in producing the overall revenue for the company. But we have different places where we can bring direct labor in relatively quickly and then transfer direct labor from maybe to some of the higher-skilled positions required to drive that business. There is a set of lessons we learned in the first half of 2016 that leave us pretty confident that we can ramp up quickly. Now you're not going to double the capacity in a 6-week time frame. That's one of the reasons why we've talked about hanging on to some of the underutilization based on the forecasts and the information that we have. But clearly, a key challenge in front of us and one that where we need to be flexible, responsive and act decisively and quickly on the way back up.

Okay, great. And then I noticed that Flash was pretty strong this quarter again. I'm sorry if you mentioned in your prepared remarks, because I may have missed it. But where does shrink come from? NOR, NAND and do you think that's a sustainable level?

Yes. It's interesting. So Flash was relatively strong compared to at least some recent quarters. And I think Mike Ludwig mentioned that it was approximately 50% NAND. So it's split between NOR and NAND. The NAND piece of it is really consistent with our Flash strategy that we talked about before. We're at the very high end of 3D NAND. These are high-layer count devices, probably not even in full-scale production yet, where customers really need FormFactor's high-density MEMS technology to test these devices. Whether that becomes an industry-wide trend or not, we're not really in a position to speculate on at this point. But it is encouraging that we've seen some moderate growth of our Flash business what was admittedly off some very low levels. It really is centered around high layer count, very dense kind of structures that require almost a DRAM-like probe card.

I see. Great. That's helpful. And then, my last question. I guess, on the operating expense level, right. Obviously, you guys account for a weaker quarter share, but you expect a strong ramp in the back half of the year. Do you think like going for that -- this kind of cycling, do you see your OpEx have to fluctuate around that? Or you think OpEx can stay relatively stable throughout the year?

Yes, we've -- I think from our perspective, Edwin, the one thing that we have worked hard at is to put an infrastructure in place that scales. So the one thing I would tell you is that when we talked in early 2017, we said that operating expenses should be somewhere between $35 million, $37 million. And I think we've pretty much stayed within that range the entire year. I would say it may tick up just a small amount, but the structure is highly leverageable. So I think as we talked about, the fourth quarter was around $36 million, probably it'd be slightly lower than that in Q1. But I still think somewhere in the $35 million to $37 million range for 2018 and maybe as we get into the back half of '18 with the volumes that we're talking about, maybe they'll be slightly higher than $37 million, but really very leverageable and it won't be much higher than that.

Our next question comes from Patrick Ho of Stifel.

First off, I want to also wish Mike Ludwig the best. It's been many years, and it's been a pleasure working with you.

Thanks, Patrick.

First for Mike Slessor. In terms of your 10-nanometer foundry opportunity that you capitalized upon in 2017. As that partner migrates to 7 nanometers, it's projected to be a much bigger node with a lot more opportunities than just the mobility market. How do you see that potentially translating for FormFactor, as that foundry partner expands its market opportunities into many more markets?

Yes, so an interesting question, Patrick. Because I think both that foundry and the industry in general view 7-nanometer as going to be one of the major nodes, maybe akin to the way 28-nanometer was. And we view the opportunity there as expanding far beyond the single integrated fan-out mobile processor that we're currently serving at 7-nanometer. Whether it's in the mobile space, clearly, there are going to be short-term opportunities for different designs around application processors that adopt integrated fan-out packaging, advanced packaging. But more broadly, our entry into that foundry and then our participation in enabling the 7-nanometer node and the larger wafer starts and design volumes on 7-nanometer node do represent a big growth opportunity for us. The timing of that and the fundamental details of it, I think as 7-nanometer ramps will become more clear, but generally we are very optimistic about what that opportunity holds for us.

Great. That's helpful. And my follow-up question, going to the DRAM probe card side for a second. You were seeing the industry, as you noted, transition to the next generation nodes and obviously, that in itself is very, very helpful to your business. But we're also seeing increased bit densities, higher speeds. Does that potentially positively impact your pricing tactics? What I mean by that more is given that DRAMs have always been a little bit of a lower-margin business, there is competition out there that you faced. But given some of the more challenging complexities involved, do you see any potential pricing benefits, particularly in a market environment that's very healthy right now?

Yes. I think there's a couple of different pieces to it. As you noted, generally, the requirements continue to get more and more challenging, whether they be speed-related or density related. And with bid counts for die going up and speeds going up, that does move things towards FormFactor's technical strengths. And it's one of the reasons why we've continued to gain share and improve the profitability of our DRAM business over the past couple of years. I think as far as pricing goes, in DRAM, for sure, pricing is intimately related to the value and the cost benefits you bring to your customers, but I think the historical strengths that FormFactor has had in enabling customers to test more chips that want that higher speed play right into that theme. And obviously, when we're able to do that and reduce their overall cost of tests, they compensate us for it.

I'm showing no further questions at this time. I would now like to turn the call back to Mike Slessor for closing remarks.

Thank you all for joining us today and especially, a thank you to the worldwide FormFactor team for delivering a great 2017, and for dedicating ourselves to continued growth in 2018. We'll talk to all of you soon. Thanks very much.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program, and you may all disconnect. Everyone, have a great day.