All right. Good afternoon, everyone. I am Sidney Ho. I cover semiconductors, semi-cap equipment, and IT hardware at Deutsche Bank. The next company we have here is KLA. KLA is a market-leading provider of process control and yield management solutions for the semiconductor industry. Today, we're thrilled to be joined by KLA CFO, Bren Higgins. Welcome, Bren.
Thank you.
Before we start, for those investors who are listening to the webcast through our portal, if you want to ask a question, there is a box on your screen where you can type in your questions. I will try to ask those questions as we go through our discussion. Before we jump into Q&A, Bren, I know you want to make a few comments to set the stage, so why don't I hand it over to you?
Thanks, Sidney, for having us. Glad to really be here. Would be great if we could have done this in person. Hopefully, we'll get a chance to do it soon. I'll just start with the safe harbor statement. I'll make some forward-looking statements today. Those statements are subject to risk. Those risk factors you can find in our SEC filings, and you can access those via our website. Well, 2021 is just a very strong, robust business environment. It seems that every quarter, going back to the last quarter of 2020, we've seen updates in terms of just revisions of growth estimates for our industry. At the last earnings call at the end of July, we talked about an industry environment that was up somewhere between 30%-35% off a base of about $61 billion for WFE, so it puts it in excess of $80 billion.
With strength for KLA across our semiconductor process control business of growing year-over-year, somewhere in the high 30% to low 40%. Well-positioned business overall. We're very pleased with how we're positioned, both in terms of the products that are driving the business, but also with new product offerings that will support the business in the future. Our service and EPC businesses, EPC stands for electronics, components, and packaging business, which is our More than Moore business, are also doing well this year with mid-teens growth. We're certainly getting most of our questions about the semi process control business, and when you've got businesses that are growing in the mid-teens, and their diluting your growth rate gives you an indication of the strength of the overall business and the Semi PC part of the business.
Overall for the company, based on consensus, 2021 estimates look like their revenue growth rate in excess of 30%, EPS growth rate, given the leverage that we drive out of our business, greater than 50%. Very strong performance. I talked about the second half of this year for the company being somewhere in the mid-teens of growth versus the first half of 2021. I also said in the call, in response to a question, that I have a hard time seeing how the first half of 2022 wouldn't be at least as strong as what we're seeing here in the second half of 2021. Some sustainability in our business, really the strength across all the markets. We're excited about that. Competitively, I think we're well-positioned, and certainly reflected in the gross margins of the company. We're excited about what's in front of us.
As we were talking before the presentation here started, some of these problems are high-class problems, trying to keep up with demand, but they're problems or challenges, and I'm really appreciative of our suppliers and our employees with how we've executed through this year. Sidney, I'm sure you have some more questions, so fire away.
Sure. Thanks for the overview. Well, maybe we'll just start off with some near-term industry questions. You talk about WFE growing 30%-35% in that range, so call it low $80 billion. What are the main variables for the rest of this year, and that determine where WFE ends up, and what are some of the key trends that you are monitoring going into next year?
Yeah, no, it's a great question, and as I said earlier, it seems like each quarter it's strengthened. Even if you look at today, given what we're seeing from customers, I think we're probably biased towards the higher end of those ranges, more so than the lower. I would say just across the board, confidence is pretty strong right now as we head into the second or the last second half of the year, finishing this quarter and moving into Q4. Capacity constraints are probably the biggest challenge, certainly for us. We've been adding capacity through the year, and I guided that we would see sequential growth as that capacity comes online. We're seeing a step up in capacity virtually every quarter as we go forward. These are investments we think that are prudent to make given the through-cycle growth expectations over time for our business.
I think that the biggest drivers for us is our ability to continue to execute against that business, that our suppliers continue to perform, and that capacity and that supply comes online. Certainly, we have customers also that are ramping facilities. In some cases, they have new facilities, there's always customer readiness challenges whenever they're adding a lot of equipment and adding it into a new facility. Sometimes infrastructure requirements can slow things down a bit. In general, the environment's pretty healthy, and I think that the trend that we've seen through the year is that it's gotten better, I don't see anything that signals a different trend than that.
That's good news. In terms of the current environment of a strong demand, supply constraints, and potential for some disruption from the pandemic, I would assume customers are looking to put in orders earlier, and you should have better than typical visibility into the future demand. Can you talk about any changes in customer ordering patterns and your visibility into demand, say, three, six, nine months out? What are your lead times right now versus what is considered normal?
Yeah, I guess that's one factor I left out, too, in the last question was, does the pandemic and effects of the pandemic around the world cause any issues with supply disruption? I think we've managed around it pretty well, but there's always the unknowns. Yes, backlogs are high, and customers are giving us orders to get into the queue. It's important for them to secure their slots, we need orders for that. That certainly does provide more visibility overall, I think that supports the statement I made about our views into the first half of 2022. I think in an environment like this, I think given the pressure that their customers are putting on them and some of the shortages that are out there that are well chronicled, customers are trying to secure those positions. Our lead times are not coming in.
Earlier in the year, I thought we would be back down to a normal lead time for KLA. Normal lead time for KLA tends to be around five to six months or so, even though the intrinsic lead time on a number of our products is much longer than that, but we do a lot in terms of inventory and supply chain management to pull that in. We're running somewhere around eight months today, and it hasn't come in partly because, A, we've added capacity, but we've also seen demand strengthen. We're maintaining where we are. I expect to see that continue as capacity comes online, but demand continues to strengthen as well. I think it's been a combination of those effects. We'd like to pull that in and get something closer to our historical model.
We're making investments in our suppliers to be able to do that. I think in the long run, when you're investing in a through-cycle growth environment that we expect of sort of mid-high single digits, then I'm willing to make those investments in the long run to make sure the supply chain is continuing to prepare to support a growth outlook that's much different than it was historically, where you had a lot of cyclicality in the industry, but not a lot of through-cycle growth. I think that the environment's different here, and we're managing our suppliers in a way to make sure that we have the supply, and we can react to inflections in demand quicker and faster from a lead time point of view.
When thinking about supply constraints and the higher logistical costs in the near term that you are facing, do you think these issues will mostly resolve themselves as we move past the pandemic, or is it something structural that maybe demand is stronger that they won't resolve just because the pandemic is over?
Well, it depends. Something like freight, it's a little hard to tell. I'm not planning on it to be transitory, of course, how the market responds and how the freight market changes in terms of supply relative to this demand is a question mark. Typically, what I've seen in my experience is usually when price increases go up, it's a little hard to get them to go back down again. Whether it's in freight, and while I'm planning on it to continue, or whether it's in some of the components. At some level, some of the more commodity components can have more variability, for KLA, those are a smaller percentage of our cost structure.
For the things that drive our cost, we have very transparent and deep relationships, comprehensive supply agreements with our key suppliers, and so it's pretty clear about what the drivers of cost are, how we will pay for them. There are also volume incentives, and some of that has been balancing against each other, some of those forces. Labor is clearly, there is pressure on labor. It seems like everything costs more, so why not labor as well? We're feeling those pressures in different pockets around the world, in different functions, and so we're reacting to that as well. I'm planning on it being mostly permanent. We'll see what happens as we go forward. As I'm modeling, based on our expectations for top line in 2022, I'm modeling somewhere around 50 basis points in our gross margin line of some of those headwinds as they play through.
It's embedded in the guidance that I've provided. I think based on our expectations for top line, we're kind of operating somewhere in the 63% gross margin range, ±50 basis points or so, and I think we're going to continue to be there, and variability quarter to quarter will be more mix-driven than anything else. There's some effect there. There's certainly some headwinds there, and I think we're doing a pretty good job of managing through it, but I don't think it's going to change anytime soon.
Okay. That's fair. Bren, I understand you run the operations at KLA as well. Can you talk about how KLA manages the supply chain maybe different than when compared to some of the peers? Would that help you better handle the component shortages that we're seeing right now?
Well, look, the businesses are different, right? The complexity of the products are different. As I mentioned, the lead time on some of our components can be 12 months, in some ways, your lead times are governed by those realities. Also, the nature of those subsystems requires us to have deeper relationships with those customers, those relationships governed by pretty clear supply agreements that cover not only performance and warranty and technical specifications, but also capacity requirements and so on. A big chunk, I would say, of our cost structure generally is in these kinds of components. Those agreements are pretty clear. As I said, they're pretty transparent. We can plan pretty consistently around what we expect to happen, those suppliers have a history of delivering to our requirements.
That being said, we continue to push them, and we're looking for ways to try to add more capacity to support the demand environment. In general, around those products, I think that that's working okay. I say okay, but we're managing it every day. Some of the other components are a smaller piece of our cost structure, the changes in cost don't have a huge impact on the overall margin profile. We're having to be very reactive. I have people every day, we're trying to resource and source in different ways to [source] components that we need. We do carry a lot of inventory supporting our service business. That inventory is about $1.6 billion. We do carry a lot of inventory of parts, and that's certainly one way that we're able to mitigate some of these risks.
We're also putting out demand a little further out than we might otherwise do to ensure that our suppliers are investing in this growth, and we're making our commitments to support them. I think in some ways, some of the things we buy are a little unique to us and our requirements, and that helps a little bit. Then I think, we do run with longer lead times in general, and that probably helps us sort of manage through some of the short-term variability that happens in supply. It takes us anywhere from 60- 100 days or so to build our systems. Usually when we start a quarter, we have a pretty good idea in terms of the parts that are required. We usually have those in-house to be able to meet our requirements. Of course, we have to execute, and everything has to work right.
In general, I think that gives us a little bit more predictability to the business overall.
Great. Maybe just one more question on the near term. Currently, we talk about demand is so much stronger than supply can handle. Just curious, from a supply point of view, what are you doing to ensure you have enough capacity to address the upside, not just near term, but also longer term? I know you talk about increasing capacity every quarter. Is there a way to think about maybe a CapEx as a percentage of revenue? What is the right amount of investment to ensure you have enough capacity?
Yeah. The right way to think about capacity is people, parts, and space, I think is the simplest way to think about it. What tends to be the long pole, obviously from a space point of view, you've got to have clean room space, and you've got to make sure that you're making those decisions further out. In some cases, we're expanding existing facilities or we're taking office space away and converting it into clean room space. In other cases, we're finding new space. You also get efficiency on space by running more shifts, right? That's one way that you can get space efficiency or facility efficiency. On the supply chain side, that's probably the longest pole, particularly around certain components, as I talked about earlier. What we're doing there is we're making longer term bets.
As I said, I believe on a go-forward basis, we're looking at a growth rate in this business that's at least in line with semiconductor revenue growth rates and with rising capital intensity, perhaps a little bit faster. I'm willing to make those investments to ensure that we have the right capacity. It does add a little bit more risk into the system if all of a sudden you end up with some sort of revenue correction out there that I have to carry that excess capacity. I think in the long run, and I think history has shown is that, in the long run, the business responds, the excess capacity gets consumed. If we have excess parts, we consume them, or excess systems, we always sell through them. We're making decisions like that.
What it turns out is not so much CapEx. Sometimes it's we're buying CapEx for our suppliers for dedicated capacity. Those are, in some ways, you end up with like a big prepaid or something like that gets decremented over time as you start to receive parts. The economic effects outside of, I guess, an opportunity cost are pretty consistent with our current relationships, but we get more responsiveness to the environments that we're in. We're looking at our relationships in that way in a lot of cases. I think so that ends up sort of helping us manage through it all, but also make sure that we're positioned in the long run to be able to respond.
On the people side, we're adding people across all of our factories, but I can usually add people and train them inside of the window of some of these optical component lead time challenges that we're facing. That's how we think about it overall.
Okay. I have an inbound question. In calendar 2021, your revenue growth is probably 5 to 6 points above WFE growth. It seems like a lot of these growth drivers are sustainable or even stronger in calendar 2022, which should again allow you to outgrow the WFE. Is that a reasonable assumption to think about that delta in calendar 2021 can continue into calendar 2022?
Yeah, I think if you look at calendar 2020, we were mostly in line with the market. I think some of that had to do with the market starting to grow towards the end of the year, b ack to what we've spent so much time talking about our ability to actually ship those systems. It took us a little bit more time maybe to get that going. We're really pleased with the overall demand environment, particularly around logic and foundry. Logic and foundry as an overall mix of WFE is probably somewhere in the mid-50%. You're seeing a nice demand environment across multiple technology nodes. Of course, significant leading-edge investment with the introduction of EUV and a lot of design start activity at the leading edge, d evelopment on the next generation node.
Also a lot of trailing edge activity, which is supporting not just communications infrastructure for 5G, but obviously the automotive shortage that's well chronicled out there, industrial and IoT applications. There's a lot of investment in those areas as well. You've got demand that's stratified across multiple technology nodes also a breadth of investment across different customers across all those nodes. It is really a good environment. I think because of that, we're seeing the usual leading-edge demand, and EUV has been a big driver for that around certain products specifically, but really across the entire portfolio. Also in the trailing edge, you're seeing not only new facilities and some new customers, but customers coming back and investing that haven't invested a lot in the past.
There's upgrade opportunities, some of the nature of reliability and performance of certain products is changing, which is causing them to think differently about the process control requirements. I think even in the trailing edge areas, it's not just the usual stuff that we've sold or older versions, but they're buying newer capabilities or newer platforms that give them an upgrade path over time, and even thinking differently about the process control strategies because of changing demands from their customer. Of course, that's good for the service business as well. I think that environment's been really good. DRAM's been very solid this year. We expect market growth to be in line, maybe a little bit better than the overall market for us. Flash has also been strong this year, less overall than the overall market.
If you were to think about logic foundry a little faster, DRAM kind of in line with the market, flash a little less, and that's how you end up this 35% or so WFE outlook.
That makes sense. Maybe just to follow up, since we talked about calendar 2022, one of the pushback we've been getting from investors is that the foundry logic spend has been so great in calendar 2021 and possibly in 2022, that the spending in this area has to go down or has to be a peak. 2023 has to be a down year. What would you say to those skeptics other than just the answer to your last question?
No, it's a good question, and I guess we've been hearing that since 2019 when foundry logic started to invest again. Then you have to look back from 2013 to 2018 or so, there was really no growth in foundry logic investments with the timing of EUV and some of the other challenges that were out there. Certainly the last few years have been strong. As I said in the last answer, I think that if you look at the broad set of demand is compelling and certainly driving that investment. We're seeing our customers in an environment where capital intensity is increasing. We're seeing very healthy profitability levels, and I think that's coming from pricing and ASP discipline, but I think it's also coming from the mix of business and the business that comes from those trailing edge nodes.
The profitability model in the trailing edge nodes, where you've got depreciated equipment, is higher, and so you're seeing a greater percentage of revenue from those nodes that is fueling or at least contributing to our customers' profitability levels. As I look at 2022, as we started earlier today, I think most of the demand is continuing to be strong. It gives us certainly confidence as we think about the first half relative to the second half of 2021. We've had customers that have made multi-year commitments or statements in terms of overall CapEx plans. I think if you look at that and you look at the breadth of investment, and particularly with more competitive dynamics, potentially at the leading edge, we feel very good about that.
Some of these trailing edge opportunities from customers that haven't spent a lot in the past, a lot of that business, which is next year. We feel very good about that. There's been a lot of activity in logic in China. I'm not sure what the growth rate is, but I don't see that abating anytime soon. As far as that part of the market, I feel pretty good as we head into next year. Beyond that, look, I'm a believer that in the long run, that semiconductor equipment or WFE is going to grow a little faster than semiconductor revenue. Capital intensity is going to increase a bit. What do you think about semiconductor revenue over time?
That you're probably, with WFE a little faster than that, you're probably looking at somewhere between, I don't know, 6%-8% sort of through cycle growth over time. I think that's where ultimately the industry normalizes to understanding that we'll have years that are way above that, like this one, and you can have years that are a little bit below that. I think given the strength of demand that we're seeing across those markets, the discipline in spending, I think some of the other indicators that we've talked about, I feel like, absent some macroeconomic event, that the industry seems pretty healthy right now.
Great. Speaking of the longer term, for the WFE, capital intensity for I should say equipment capital intensity is roughly 15% this year. That's quite a bit higher than the past few years. Do you think that trend is sustainable, or does this continue to increase over time? More specific to KLA, do you see process control opportunity within WFE growing?
Yeah. To the last question, I look at capital intensity as a percent of revenue. I look at it as a percent of profits. I look at some of these other dynamics that we've talked about, and while we're seeing it increase, we're also seeing it increase at a time when our customers are also extremely profitable. They're spending more, but they're also making more, too. In terms of process control, I think the introduction of the EUV has allowed linear scaling to begin again after a period of time where it wasn't, and that's driving a lot of end market activity at the leading edge. We talked a little bit about what's driving the trailing edge, but that leading edge activity is there's a lot of demand that's driving a lot of design starts through fabs.
All these design starts have different process flows and test design rules in different ways. All that drives customers to have a deeper or a broader process control strategy and sampling strategy. We are seeing, again, our customers investing as a percentage of WFE at a much higher level than what we saw in the past. Customers also don't have the ability to move capacity given the strength of demand in some of the earlier nodes. The design starts, if you go back to the middle of the last decade or so, 2014, 2015, you had very limited new design start activity at those nodes, the 20 nm node, the 14 nm and 16 nm nodes. As the higher volume products moved through, there wasn't a lot of follow-on business that came in to consume that capacity.
EUV was delayed, the technical driver wasn't there as much for customers to have to buy new equipment as they moved to the next node. They were able to take a lot of that equipment and reuse it. If you look at where we are today, you've got linear scaling and technical drivers that are forcing a technical upgrade, but you also have this capacity that has to stay in place to support all this follow-on business. There's over 200 design starts, I think, at 7 nm, somewhere between 200 and 250 design starts. You've got a significant number at 5 nm, and you're starting to hear some of the activity that's coming at 3 nm. All this is forcing our customers to have to deploy new capacity, and so that's been a driver for process control intensity.
As a share of WFE, we would expect in our 2023 plan was to see from 2019 to 2023, to see KLA's share of WFE to grow somewhere around 75-100 basis points or so. I think we're on pace for that. I think the mix of the business and then some of these dynamics that I talked about has probably been the biggest driver of that. We're encouraged by it. Our share has always been strong. We're 4X our nearest competitor. I think there's opportunities for us with the right products to gain a little bit more share as we move forward. It'll be a combination of share and intensity, I think that as a share of WFE, there's a 100 basis point opportunity for us, over the next couple of years.
Great. Staying with process control here. When we look out to the next few years, we're going to have a number of inflections like Gate-All-Around nanosheet and the logic side and the 3D DRAM, more layers, and 3D NAND. How should we think about the opportunity for process control when we start going through these transitions?
Yeah. Change and complexity have always been good drivers for process control. If you look at not just what's happening on the lithography front, but also in chip architecture and some of these other design changes that we've talked about, those are all great opportunities for us. It will drive certain businesses. When you have a portfolio approach, you'll see inflections in certain businesses and less reliance on other product lines. When you have the full portfolio, you're positioned to be able to really effectively take advantage of those trends when they materialize. If you're a point product competitor, like some of our competitors, you don't necessarily have that ability to do that.
I think it's back to my earlier statement, that the opportunities we see over the next few years will be fueled by some of these technology changes that are coming in the industry and will drive an opportunity for us to grow our share of the overall market. Mix is the biggest factor, when I say mix, logic foundry mix versus memory mix in terms of process control intensity, because it's roughly 2x as high in logic foundry versus memory. If you look at the way the markets are forecasted to grow, I would expect that we'll see this sort of 55/45 logic foundry mix, going forward, and that's embedded in these assumptions. I think it creates opportunities. Certainly nano or nanosheets, Gate-All-Around, another term for it, is a huge opportunity for our film measurement business. You're seeing significantly more layers. There's an intensity opportunity there.
Because of the Gen 4 and Gen 5, we think that some of the materials and architecture issues in Gate-All-Around is more conducive to the wavelengths of the Gen 4 product line. We think that having both there creates opportunities for us. I think given change and complexity is the biggest driver for our business, I think it creates an opportunity for us to see the share of WFE grow.
Got it. Maybe one area that I feel like I don't think a lot of investors pay too much attention is the optical metrology. Can you talk about why that's an important segment and how is KLA positioned there?
Yeah, it really is. It's a business that in the last couple of years has almost doubled. We do have businesses in KLA. Sometimes I get questions from investors about, well, KLA isn't as capacity levered. It's much more about R&D and fab ramps, and that in a capacity environment, customers don't buy as much process control. In some businesses, that's true because they're doing their fab, their yield learning, they're ramping a fab to entitlement, they're maturing processes, and then they get to a level where they know what they're looking for, they know what they're monitoring for, and so they'll change their sampling strategies or use tools that are more volume-weighted versus sensitivity because they're always managing some of these trade-offs. We do have businesses in KLA that scale with capacity. Film measurement is clearly one of those.
We've seen that market grow in share over the last few years and also grow much faster than the market. I like to say it's the biggest market in KLA that nobody ever asked me about. It's been a great opportunity. I think, as I said, we've seen it double. Thin-film metrology is a $700 million revenue level for KLA this year. It's inflecting in these areas, and the Gate-All-Around, given the incremental layers that come with that, will be another opportunity for it. We also are introducing new X-ray technology into that space to create intensity opportunities in memory. It is one of those businesses. Unpatterned inspection is another one of those businesses that scales with capacity, and so we're seeing nice growth in that business as well across weather. That's inspecting bare wafers, but also as customers are running processes.
As they're ramping a fab, they do a lot of unpatterned inspection to do process tool monitoring to make sure the environments are clean and there's no particle introduction through some of the wafer transfer in the processing steps. There are some businesses that inflect with the overall market, behave a lot like capacity businesses, and we're seeing nice growth from those.
Excellent. I want to switch topic to talk about the EPC business, which we have a few minutes to talk about this. It's an area that I feel like investors continue to underappreciate. It has now been over two years since you closed the deal. Can you talk about any surprises, positive or negative, that you have seen from the deal? Secondarily, what are the areas within EPC that you are most excited about or most likely to see an inflection point in terms of growth trajectory over the next few years?
Yeah. One of the sort of the investment thesis behind the acquisition of Orbotech and the creation of EPC was that we wanted to make sure that we were ensuring and increasing our exposure to the areas our customers were going for capability, cost, performance. Certainly, we have lots of exposure to wafer front-end, and that's an area of that. They were also investing in things like advanced packaging and what we call the Moore opportunities. We already had a business in finished component inspection, our ICOS business. We paired that and put that with the Orbotech businesses in the creation of this group. It's, I think, worked very well if you look at the specialty semiconductor part of the business, which is making specialty semiconductors for RF, for power semiconductors, but also for advanced packaging.
These were products that are uniquely designed for those applications. Those were niche-y markets for a long time, and they've inflected because of 5G, but also automotive on the power side. That's a very strong business, a very differentiated business in terms of its performance. We do segment reporting. We run mid-50% gross margins. Most process tool gross margins are in the 40%, so it gives you an indication of the differentiation we have in that market. We're really pleased with the performance there. We have future plans for some new product offerings in that space for other segments within those markets that we think will create some opportunities for us. The PCB business within Orbotech has a great service attach rate, but also you're seeing the PCB board evolve into the advanced package through IC substrates, and so that's increasing our exposure there.
We've been able to take some of the KLA know-how and technology and try to deploy those into these businesses in terms of new capability for future products. We've been able to engage with customers, I think, at a higher level as a bigger supplier. We carry more weight in those relationships, and of course, it carries more risk, too. We have to make sure we succeed, and we don't put our customers at risk. Sometimes customers don't want to deal with small suppliers that may not have the scale to support the broader effort or make big bets. I think with KLA behind these businesses, I think it's helping drive much stronger relationships there. I mentioned components. You've had mobility driving that, but you also have high-performance computing and data center applications with more complex packages, and so that's been a driver for that business.
We have a flat panel business, which we've done a lot of work on sizing that business to make sure we had a cost structure that matched the top line and tried to position that business to be able to grow when that part of the market recovers. Certainly, things like microLED, which have semiconductor-like processes, are opportunities we think to leverage some know-how within KLA in terms of how we capture or think about those market opportunities in that business going forward. We like it, I think one of the other things that we thought a lot about was these were market-leading positions, and we could get more out of them than they could on their own in the past.
We've been able to drive incremental operating margins, the operating leverage in the business, consistent with the overall leverage of KLA, and that was an important aspect of our thesis as well. I think overall, there's some very interesting opportunities. We're excited about them, and we're making a lot of progress on integration. We'll have systems completely integrated in the middle of 2022, and I think that will drive further efficiencies out of the operating model there.
Okay. Maybe we have time for one more question, so I'll focus on a target model. You're tracking well ahead of the 2023 financial model. Obviously, the WFE environment has improved since the 2019 Analyst Day. Can you talk about other company-specific factors that have allowed you to be tracking ahead of your model? Is there a certain framework that we should use to think about your earnings potential down the road if we have to come up with a whole estimate?
Yeah. I don't know if I'll give you a whole new model here today, but it's certainly something that.
I'm not expecting that.
If you go back and look at the assumptions underneath the model, part of the assumptions was this assumption on industry growth. At the time, we didn't spend a lot of time defining what the market growth would be because part of that story was much more about some new product offerings that we thought would drive intensity and share opportunities for the company going forward, and that that was going to drive some changes. The market forecast will be what it is. If you think we're conservative, then fine, go model it because we will get our fair share of that like we always do. Certainly, the strength of the market environment around WFE growth has been the one assumption that's been very different than what we were forecasting.
Because of that and because of the profitability profile of those products, that's driven the excess model performance. That, coupled with the fact that we've seen such significant revenue growth, mid-teens last year, talking about 30% or more this year, that it's very hard for your cost structure to keep up with that type of revenue growth. I think because of that, the mix has been more favorable, and that's driven us to where we are. I think that a lot of what we've been doing is sustainable. Certainly, as I mentioned earlier, there are cost pressures in the business, and so we'll have to invest in that. There'll be some catch-up spending that will happen. Certainly, there's volume-dependent activities that we need to invest in to support the business at these revenue levels as well.
Our model is, there's a few ways to think about it. There was the profitability level at certain revenue tranches, there's also how you think about the leverage in the model, and there's two ways to think about that. One is the growth rate of revenue relative to the growth rate of earnings. We've said that we're going to grow our earnings 1.5x the revenue growth rate. The other is we're going to deliver 40%-50% incremental operating margin leverage on our revenue growth. We should drop through $0.40-$0.50 on every revenue dollar. Certainly, over the last two years, we've outperformed that. We've been in excess of mid-50% in terms of the incremental leverage.
I think on a go-forward basis, I think there'll be some catch-up to that, given some of the pressures I talked about. At the end of the day, though, we believe it's the right model to think about leverage 40%-50%. In a normalized revenue environment as we move forward, I would expect us to perform within that range. Maybe, depending on what happens with revenue growth next year and beyond, the lower end of the range versus the higher. I do think it's still the right model for KLA, and we should be able to perform in line with that.
Okay. Well, I think we're out of time. Thanks for spending the time with us, Bren Higgins, and enjoy the rest of the day.
Thank you. I appreciate it.