ON Semiconductor Corporation (ON) 51st Annual J.P. Morgan Global Technology, Media and Communications Conference Call Transcript

ON Semiconductor Corporation (NASDAQ:ON) 51st Annual J.P. Morgan Global Technology, Media and Communications Conference Call May 23, 2023 3:55 PM ET

Company Participants

Hassane El-Khoury - President and Chief Executive Officer

Thad Trent - Executive Vice President and Chief Financial Officer

Conference Call Participants

Harlan Sur - JPMorgan Chase & Co.

Harlan Sur

All right. Good afternoon, and welcome to the second day of JPMorgan's 51st Annual Technology, Media and Communications Conference. My name is Harlan Sur. I'm the semiconductor and semiconductor capital equipment analyst for the firm. Very pleased to have the team from ON Semi here with us. We have Hassane El-Khoury, Chief Executive Officer; Thad Trent, Chief Financial Officer. The team held their Analyst Day a week ago. It was a great event. It's really been incredible to watch the team execute to its strategy of growth, margin, earnings, free cash flow power expansion over the past few years. So what I've asked Hassane to do is to sort of just kick it off with just a brief few minutes summary of last week's Analyst Day.

Hassane El-Khoury

Yes. I mean, like you said, it was a great day for a few reasons. I would say primarily, it was a testament of a solid strategy that would continue and expand beyond what we've been doing in the last couple of years. Like you said, we've embarked on a transformation. That has been going exactly according to plan. Our strategy is solid. Our strategy has been stressed with the macro environment, and it's sustainable. Our results are proof of that. Our investments in the right areas, what I call the sustainable ecosystem of electrification of the infrastructure, electrification of mobility have shown the results already.

And the way I judge was at a good Analyst Day is we didn't do a reset. We actually did a continuation about what's next, what's the journey ahead. That's what made it a great day for – not just for our employees, but our shareholders and really more importantly, our customers who have been on that journey with us the last couple of years. And based on the long-term agreements we have with them, they'll be on the journey with us for the next 10 years.

Question-and-Answer Session

Q - Harlan Sur

Perfect. I'm going to start off with – I'm going to go – I typically start off with near- to mid-term. I'm going to go backwards now because I do want to focus on some of the aspects of the Analyst Day and sort of the mid- to longer-term strategy. First question is going to be on your silicon carbide strategy, right? You guys have done a phenomenal job on ramping internal silicon carbide, sort of end-to-end manufacturing infrastructure, right? In fact, you are ahead of your ramp plan and looking at your internal capabilities to help drive your 70% per year sort of silicon carbide revenue growth profile with the gross margin profile is accretive, right, to your current gross margins.

You still have some cost and efficiency tailwinds over the next few years. I do think that the move from 150-millimeter 6-inch to 200-millimeter 8-inch diameter wafers is part of that further optimization and efficiency unlock. What's the status of the development efforts here? And what's the timing of the team's transition to 200-millimeter?

Hassane El-Khoury

Like you said, we've been exactly where we are as far as what we've outlined on the silicon carbide ramp. We've 5x the capacity on the substrate. Our yield since the time of the acquisition are only embarked on the silicon carbide journey have been up 1.7x. We're basically doing exactly what we said we were going to do. And actually, we're ahead of our own internal plans, which showed the results of we doubled our revenue quarter-on-quarter, and we're on track to deliver the $1 billion. That's what you need as a baseline in order to start talking about a transition. So we have a very solid baseline. We have a solid technology. And more importantly, we have a solid LTSA or revenue committed by customer that we talked last week in Analyst Day is at a $9 billion lifetime revenue for silicon carbide.

So all of these put us in a very good position to build upon. And building upon, one of the aspects we build upon is on the going – the migration from 150 to 200-millimeter. Where we are – and our plan has always been, we will enter the market in 2024, and we will generate revenue in 2025. We're still executing to that plan. We already have it today. Now people ask, well, why don't you accelerate it? Because we don't have a reason to accelerate it. And I'll tell you where the balance needs to be. Our 150-millimeter is scaling. It's yielding. It's highly effective. And we're on track to absorb the start-up costs for it, which puts us on that margin trajectory in order for us to establish a very solid baseline upon which to build the 200-millimeter.

Some of our peers have had struggles when they need to ramp very quickly. We've been able to 5x because we had a very solid baseline on the 150-millimeter. We are establishing that same baseline right now on the 200-millimeter. So when it's time to ramp, we are able to ramp very quickly and effectively.

Harlan Sur

And I think you did mention a number of years out, right? You will be – your next brownfield expansion for your silicon carbide manufacturing capabilities, you'll be committing $2 billion of investments, right? Do you expect that, that brownfield will be 200-millimeter?

Hassane El-Khoury

Yes. So look, as you look out in time, all of our investment are going to 200 millimeters because of, again, the transition is inevitable. Now one thing I do want to make sure everybody understands is our 150-millimeter capacity today is 200-millimeter capable. So when I say we've been running 200-millimeter already to establish the baseline, we're actually running it on the equipment that we have already in place. It's not some new facility.

And that adds that baseline because you can expect when we convert to 200-millimeter, if you have been running that 200-millimeter on the existing line, then you have the yields, you have the outlook, you have all of that already established to be able to run. If we are adding – as we add the new capacity or the end-to-end manufacturing site, that will more likely be on the 200-millimeter just because it will be in that time when we start the 200-millimeter conversion.

Harlan Sur

And you did bring this up in terms of some of the challenges that some of your competitors have had. The ON team, you've been doing power for 30, 35 years. You've ramps. You've built a learning curve in all of the different flavors of power transistors, modules, packaging, right? I mean – and then, obviously, doing it in a very systematic fashion. In the same way that you just described, doing it in a systematic fashion and moving from 150-millimeter to 200-millimeter, right? How much of this 30-year learning curve across power, module capabilities, a lot of the special intellectual property and in-house expertise is responsible for where the team is today, the $9 billion of committed silicon carbide revenues, how much of that is this learning curve and systematic approach to process manufacturing improvement and cost reductions?

Hassane El-Khoury

I would say pretty much all of it. You have to have solid IP. You have to have competitive design. So all of that are what I call the baseline upon which, if you don't have the competitive advantage on a technology, and I categorize technology as device and packaging in one. If you don't have that, then it doesn't matter how quickly or how effectively you build capacity. So we have that competitive moat that we have across all of our IP and capabilities. But to scale that very effectively and not just effectively, but at the speed by which we scaled it, then it's all about the knowledge, the history and really, call it, scars on our back for the last 30 years of ramping fabs, ramping power semiconductors and ramping customer designs in the automotive industry that's very demanding.

Harlan Sur

Yes. And I feel like that's a pretty big competitive advantage, right? You can talk about figure of merit or transistor performance, all that kind of stuff, but I think people fail to realize one of the biggest barriers in this industry is learning curve, manufacturability, yield, cost improvement, intellectual property.

Hassane El-Khoury

That's right. And the faster you can resolve these issues, the faster you can stay ahead of your learning curve and delivering and over-executing to your own plans then you'll get the results. And if you look at our first quarter results, we came ahead because of exactly all of these improvements and all of that firepower we have within the company, where we have individuals running those initiatives that have done it before and know what to look for kind of around the corner.

Harlan Sur

And as some of your competitors have struggled because they're still coming up their learning curve and cutting their teeth, as it relates to the $9 billion in committed revenues for silicon carbide, has there been an acceleration of customer engagements, new commitments just given what appears to be some issues with some of your competitors, start-up issues with your competitors?

Hassane El-Khoury

Yes. Look, it's a natural state where we have customers that have come to us because they have plans to ramp, and any concerns that will jeopardize their ramps, they have to find a solution. So we've seen an uptick in interest. Obviously, in 2023, we've said we've sold out. So it's not something that's going to impact the short-term. But definitely, this is a long-term view. That is not yet reflected in our LTSA's because that's the engagement we're doing now. So you're going to start seeing more and more engagement that we make public.

But again, at the end of the day, what you have to always remember is it's the technology and the superior technology that's going to make the win or loss. I tell my team, we're not going to win if somebody else doesn't win. We need to win when everybody else is able to win because that's the competitive advantage that is sustainable, and that's what we focus on. That's why I always start an anchor on technology, technology, technology, because without that, any win is short-lived.

Harlan Sur

Yes. You did a great job of giving us last week at the Analyst Day a deep dive into your automotive clearly, your industrial efforts, but often overlooked, you guys didn't overlook it, but I don't think focused a whole lot on it. But as you're focused on 5G and cloud power, right? Because it's a great way to leverage your entire power portfolio, high-margin, fast-growing market, right? And last week, you articulated a 20% growth outlook for your 5G and cloud power business. That's 2x faster growth than you had anticipated the last time you guys updated us two years ago, right? So what's driving the better growth outlook? Is it higher dollar content? Is it the data center move from 12-volt to 48-volt infrastructure power? Is it higher computing performance? What's driving that?

Hassane El-Khoury

So it's actually all of the above, which is a great market. So like you said, two years ago, when we talked about the cloud, we talked about 11% growth rate, now it's 22%. So Sudhir in Analyst Day talked about the power tree that also applies to the cloud, the same as it applies in automotive and industrial. That is where our technology competitive advantage comes in. You mentioned 48 volts. We're already in production with 48 volts. So we're at the cusp of this 48-volt conversion, and we're there to play. More content in per, call it, per server, therefore, per cloud infrastructure. That content has gone up, which is also driving our growth with it.

And really, more important – and the compute power is getting more and more. And that's very important because it's not just adding more content because it requires more power, but it requires more efficient power, which means when the power requirements increase, they can be proportional an increase of consumption to increase of power, therefore, efficiency starts to matter because that's all CapEx. That's all expenses that our customers are going to have to absorb.

So by us making the power conversion more efficient, they will monetize it on the back end in consumption reduction on electricity and cooling and so on. That's why we're winning. And we're – but again, it goes back to the technology to anchor on technology. And when you have the right technology at the right time in a growth market, that's how we end up outgrowing the market.

Harlan Sur

I think a metric that reflects all of the things that you've talked about, right, technology leadership, share gains, increasing complexity, portfolio breadth is cross-selling, right? And I think it's the dynamics that have made the team so successful is the breadth of the portfolio and the ability to cross-sell. And I think last week, you guys mentioned that your top 20 largest customers purchase on average per customer 800 products, 800 ON Semi products. How does that compare to two to three years ago? And do you guys have a particular marketing and sales programs in place that incentivizes the team to as you're going after this opportunity, let's make sure that we maximize ON Semi content per board, per car, per whatever the case maybe.

Hassane El-Khoury

Yes. So it grew from the last two years. And a lot of it is we've gone through a very constrained market dynamic. And we have made a very strategic approach to prioritize some of our strategic customers in the automotive and industrial at the expense of the broader consumer compute market that is not strategic for us. And what that allowed us to do is be able to service a lot of our auto and industrial customers when some of our peers could not. And that created a much longer-term share gain that we're benefiting from today.

So that's kind of the catalyst that started this whole thing. Do we focus on cross-selling as an active sales and marketing or a go-to-market? Absolutely. So when – in Analyst Day, I talked about how the structural change, the go-to-market change that we do, that is definitely a pillar in it of sure, well, it's always easier to sell to an existing customer a new product than to go and capture a new customer. And we have to be able to do both in order to have the best-in-class cost of sales. You do it by getting big wins, strategic wins, cross-selling, so lower customer capture and adopting new customers. That's our go-broad strategy with our tools and the software.

All of those are part of that efficiency that Thad talked about in the new model of getting to that 13% OpEx. We're going to keep investing in R&D, but SG&A is going to get, call it, subsidized by the growth on top and the efficiencies we do in the go-to-market.

Thad Trent

And Harlan, I would also add that the – our strategy with LTSA's have elevated the engagement with our customers. And now we're selling much more products to that same customer base because they have assurance of supply, right? So they're co-investing with us. So the relationships are much deeper, and that's allowing us to get more of their wallet at the same time.

Harlan Sur

I'd like to – on the financial targets that you put out last week, 10% to 12% revenue CAGR, 53% gross margins, 40% operating margins and 25% to 30% free cash flow margins. The team has been defocusing away from low gross margin businesses, right, over the past number of years.

You're clearly richening the mix of your portfolio with new products. And so looking at that 53% gross margin target and looking at your current business today, what percentage of your current shipment mix is driving gross margins higher than that 53% range? And more importantly, of the design win funnel, what percentage of the design win funnel that you're securing for the next few years carries gross margins that are higher than the midpoint of the target?

Thad Trent

Yes. So let me take that. So if you look at our design win funnel and you look at anything coming out of R&D, it's at or accretive margins, right? So everything is accretive. So as we think about new investments, it's at the target or above. So anything that we're working on currently should drive incremental gross margin improvement. If you think about the improvement steps of where we are to get to that 53%, a big component of that is mix.

And it's two components of mix. One is auto and industrial, as we flex more there, which we've been doing. Last year, we were roughly 73% auto and industrial. In Q1, we're 79%. If you fast forward, we're going to be 85% auto and industrial. But the other component is the mix to this higher-value product, proprietary products that drive higher gross margin. And that's these products and these markets that we've talked about. We're also exiting the low-margin business, kind of historically low margin, competitive. We'll exit that. We'd get another $400 million that we're going to exit this year. That helps with the gross margin over time as well.

Today, that's good gross margin. But as supply comes online, we expect our competitors to actually take some of that product in that business. And that's fine. We're planning for that. So we'll allocate that capacity into these high-margin products.

Hassane El-Khoury

Now to give you – we talked about a digital-first strategy as part of the Analyst Day. That, of course, is a big portion is the go-to-market with the tools and the efficiencies and so on. But internally, you mentioned on the funnel and the visibility we have over a longer term.

Both Thad and I review the funnel – before opportunities go into the funnels, they're [indiscernible], and nothing goes into the funnel if it's dilutive. We're not into the fab filler. We don't care about that. We're not going to be going after that. We're going to walk away from all that business and that culture in the company, now it's less and less. So we're done. That's kind of Phase I of our transformation. Phase II now is ramping and growing those new products that are all accretive to margin.

Harlan Sur

You guys highlighted some interesting differentiators that – last week that we certainly don't hear very often from power and mixed signal companies, right, and that is software analytics and simulation platforms. In the case of your power business, right, your software development kit for your image sensor customers and the move to digital control architecture for your power customers to optimize their – the power flow and their systems, right? Sounds like the playbook for a leading-edge semiconductor – leading-edge digital semiconductor company, right? So is the role of software, digital-based techniques, a strong differentiator for the team? And is the team spending higher proportions of its R&D investments in this area?

Hassane El-Khoury

Yes. At a high level, I agree with you. We are the, call it, the benchmark for what power should be. So I don't aspire to be like other power companies. We aspire to be like the best companies out there in the world, which is not typically thought of for power, but our financial target is the best-in-class if you look at it, hands down and not just in the power domain, but in semiconductors.

Having said that, that requires a very different go-to-market and a different mindset of what you invest in. But we've always said the combination and the competitive advantage is not on a piece part. It is how can you make all pieces work together to solve the customer problem. One of the customer problem is it is expensive, and it takes time to iterate on power semiconductor because you have to build the unit and run it in order to get – you get the efficiency. So that was an anchor point on, okay, if that's the biggest problem, how can we solve it?

We have the domain knowledge. We have all of the stuff because we're vertically integrated. So we're able to allow the customer with our tool sets to iterate in a few hours what they otherwise would have done in a year. And it doesn't cost them anything because there's no capital or they don't have to build the samples or any of that or wait for equipment to get delivered.

And that started off as a few customers, lead customers. And it turn into a very broad approach that we have, and that have allowed us to get a very big competitive advantage because what the customer is able to do is simulate a lot of these architectures. Do I go with quad SiC? So I go with 2 SiC? Do I go with IGBT and SiC? Imagine those architectural decisions at an OEM level, how valuable they are with our tool set, number one.

Number two is when they actually get to the silicon and get the package and the performance is exactly the same as what they modeled, they now can proliferate it within the account. That is, again, a competitive advantage. The same thing applies with image sensing. How can we get the image sensor with software to be tailor-made for an application? That is a competitive advantage, not just from a go-to-market, but from differentiation of the customer. If they're able to make things that are their own versus an off-the-shelf, they have a competitive edge against their peers. And we enable that, and that has value.

Harlan Sur

Let me shift to some of the near to midterm trends in the business. So demand trends remain strong in auto and industrial. As you mentioned, it's 79% of your revenue mix. While it seems like your non-strategic end markets, right, which are primarily consumer-focused, are stabilizing. Some of your peers have been cautiously optimistic on growth for the second half of the year. What's the team's view on growth for the second half of the year, let's say, versus the June quarter levels?

Thad Trent

Yes. I think look, a couple of things are happening here, right? When you think about consumer and compute that got soft mid last year, I think that's remained soft, and it's continued that way. Industrial got soft and then stabilized for us last quarter. And auto, obviously, has been robust for us and continues to be robust.

I think when we look into the second half of the year, we're very optimistic, but we're cautiously optimistic just given the uncertainty in the market. We have great visibility in what our customers' demand is because of the LTSA's. We're building to the LTSA's.

But when we look at where The Street is for 2023, it feels good. I think we're in a good spot there. If you exclude the growth from silicon carbide and you exclude the exits, the $400 million of exit year-on-year, the second half should outgrow the first half.

And if you take that base business, excluding silicon carbide and the exits, it's growth of like low single digit. I think that feels good right now kind of in this environment. We've been very proactive in managing wafer starts, managing inventory in the distribution channel, just being cautious in terms of the outlook here and supporting our customers. And I think we feel good about where consensus is and kind of the outlook for the rest of the year.

Harlan Sur

Does some of that confidence come from – I mean, I know it is the case during periods of macro uncertainty and end-market weakness, right? You go through these periods where customers are wanting to reschedule, move backlogs, very high levels of activity on that front. Then you go through a period of that activity sort of wanes and then things start to sort of stabilize. And then obviously, the next move would hopefully be the upward inflection.

The team has accommodated customers on shifting timing, helping customers move shipments around. But has that activity level in terms of push outs started to stabilize? And more importantly, is the team starting to see some improvements in like level of your global POS or sell-through?

Thad Trent

Yes. So let's back up and talk about our business. Our LTSA's or what we're managing, right? And when customers are coming in, they need help with LTSA's. We're getting the call quarters in advance of something is going to get solved.

If we can create a win-win with that customer, we'll try and help that customer. And there's got to be a win for both companies. We saw a lot of that happening late last year. We've actually seen in the first quarter and even so far in the second quarter is that's actually stabilized and actually coming down a little bit, which gives you some confidence in the [indiscernible] well.

Again, think about the visibility we have with the LTSA's and the fact that we're getting less of those requests. There's pockets of them. But late last year, we saw a lot more uncertainty, and I think people remain cautious. So that allows us to really have a dialogue with our customers versus historically, they place a PO, and that PO would disappear 30 days before it shipped. Now we're getting that call in advance, and we're able to help those customers. But it is kind of slowing in terms of the rescheduled request.

Harlan Sur

Yes. Touch upon the LTSA's. You've done a really good job on these long-term supply agreements. It gives you visibility, right? It helps your customers – well, helps you guys plan your production requirements, right? It gives your customers peace of mind and assurance of future supply. You guys have increased your LTSA's by $1 billion, right, last quarter to $17.6 billion with $5.8 billion committed revenue from those LTSA's being recognized over the next 12 months, right? So that's approximately 70% coverage on consensus next four quarter revenue outlook.

Obviously, that's great visibility. Some of the LTSA's obviously are running their course as well. But given how beneficial it is for these customers, I would assume that the renewal rate is quite high and maybe encompasses more products versus the last LTA agreement. I think we talked about that. But does the team have any metrics to share with us, renewal rates and increasing part numbers?

Thad Trent

Yes. I mean, your key metric right there, you said it is $1 billion in Q1, right? Our LTSA value went up $1 billion in one quarter in a market that's uncertain.

Harlan Sur

A lot of that was renewal? A lot of that $1 billion was renewal?

Thad Trent

It's a couple of things. It's customers extending their LTSA's. So maybe they're tacking on additional years. So on average, an LTSA is four to five years. So we're a couple of years into this and maybe tacking on to the end. The other is they may be expanding it. So we've got hundreds of part numbers on these LTSA's. As they [indiscernible], customers are wanting to put additional part numbers on LTSA's.

And then the third element is customers that didn't have an LTSA that have now come in and saying, that's the new mode of operation. It's working. We want to get on LTSA. But I would say that, that is the leading indicator of $1 billion in Q1 is a very favorable sign in a soft market.

Harlan Sur

I want to step back for a second on the image sensor business, strong market position, 68% market share in auto ADAS. Overall, strong number one position in auto and industrial markets. This business has been constrained over the past three years. I think it's still somewhat constrained even now, I believe.

And so can you just give us an update, Hassane, on the efforts to bring in some of that image sensor mix in-house to your 300-millimeter fab. I mean, I know – I assume that in addition to better supply management, some in-sourcing will help drive an even better gross margin profile, right, due to the lower cost of your 300-millimeter facility. Is there – also, is there any way to kind of quantify that benefit as you bring some of that in-house?

Hassane El-Khoury

Yes. So just at a high level, that business just delivered over 50% gross margin from historically where it's been.

Harlan Sur

The majority is outsourced, right?

Hassane El-Khoury

Majority is outsourced. So you can think about gross margin trajectory that is favorable. When you go even to 12-inch because some of it was still on 8, going to 12. Some of it is 12 to 12, but outside versus inside, which gives you a cost benefit. So where we are, we've already done a tape-out for that business. That tape-out is basically the first proof that we're able to do it in East Fishkill, which is historically not an imaging fab. So from a technology perspective, we have the technology readiness, product sample and then the proliferation. Obviously, these are – go into the automotive market. So you have time for design in before you start seeing the revenue and therefore, mix. But what you said is absolutely – the vector is more supply assurance but also a margin expansion.

Harlan Sur

And as you think about your prior fab-lite strategy and now fab right strategy, right, what is the optimal mix of image sensors, in-sourced versus outsourced, you think?

Hassane El-Khoury

So we're more focusing not on the optimal mix. We're more focusing on – so fab right is, more importantly, it's getting the right product in the right spot whether it's inside or outside because there are products that would be more suited depending on the area that are outside versus inside. That's fab right, not just on image sensing but across our whole chain.

So we have fabs that potentially are running six variants. They will end up running four variants and moving two out, so we can get more efficiency within the fab and therefore, more outs, which drop the cost. So there's a lot of that benefit that's going to get us to that 53% margin, I would say, not a destination but a milestone. But that's all of the manufacturing efficiencies that we still have ahead of us. So there's a lot of margin expansion opportunities that are baked in that we need to execute to.

Harlan Sur

Absolutely. On the – maybe again on some of the more near term. On the supply side, lead times remain stable, 41 to 43 weeks. I think though, traditionally, your historical average lead times were more in that sort of eight to 12-week range, right? And this does obviously reflect. The current 41 to 43 reflects continued supply tightness but strong demand profile, right? The team has mentioned that there's still supply constrain. Can you just articulate some of the product categories where lead times are stretched beyond 41 to 43 weeks? And do you anticipate lead times coming back down to more normalized levels over the next few quarters?

Hassane El-Khoury

Yes. So some of the technologies, obviously, I think we've been consistent with that. Obviously, silicon carbide is fully constrained when we sold out. IGBT or other silicon power technologies are still constrained, mixed signal analog and image sensors. So a lot of our strategic is still constrained, and that's where we're adding capacity in order to support LTSA's.

But we are adding capacity for LTSA's, which makes really the lead time not really a relevant metric anymore for the bulk of our business. What I mean is I don't think lead time are going to go back to where they used to be because you have fab cycle time, our new technology is longer cycle time than that crank commodity, and we're not in a fab filler.

Therefore, we're not going to keep building just to put inventory and reduce lead because there's a cost associated with short lead time, which I don't think is necessary anymore given the LTSA's. So they may come down for some technologies, but we don't see it come down to where it used to be because, again, it doesn't need to. And there's a better cost structure associated with a tight balance based on the LTSA versus just driving lead time to drive lead times down.

Harlan Sur

I mean, I guess that's one of the benefits of the LTSA's to give you exact timing on delivery, right? So in that – from that perspective, lead times doesn't really sort of come into play.

Hassane El-Khoury

That's right. Because, look, lead time drop if you don't have an LTSA, so does cancellation. So it doesn't really matter. LTSA is what gives the visibility. Everything else we do based on the LTSA is operational excellence.

Harlan Sur

On the distribution channel business, that remains lean at seven weeks and below the historical 10 to 12-week range because you guys continue to proactively manage your channel inventories. When the supply and demand environment does normalize at some point, does the team expect to remain – to maintain channel inventories above current levels, but maybe slightly below historical levels? I mean, I would think a linear profile is quite optimal for you guys, right, because I think it drives better responsiveness from the channel. I just wanted to get your views. Do you ever go back to those – that 10 to 12-week range?

Thad Trent

Yes. I don't think so. I mean, historically, we've been 11 to 13 weeks. We've been running at seven weeks for several quarters now. That's been proactive management by us. Our distributors would take much more inventory if we would ship it to them. We are on a FIFO basis, and it's exactly what you said. It allows that channel to be more efficient for us to have better visibility in what's selling through and where there's inventory pockets as well. So I don't know that seven is the end goal, but it's probably somewhere in between.

But I don't think we go back to 11 to 13 because I really don't think we need that much inventory in there especially as we ship more of proprietary products, you just don't need it sitting on the shelf, right? And with the LTSA visibility, we don't need it. The distributors will support a long tail of customers. So we do have to have inventory there, but you just don't need that 11 to 13. It's just not efficient. And it doesn't give us the visibility we need.

Harlan Sur

Great. Well, we are just about out of time. Hassane, Thad, great discussion. Appreciate you guys participating.

Hassane El-Khoury

Thank you.

Thad Trent

Thank you, Harlan.