Synopsys, Inc. (SNPS)
September 28, 2011 9:00 am ET
John Chilton - Senior Vice President of Marketing Group & Business Development Group
Deirdre Hanford - Senior Vice President of Global Technical Services
Joseph W. Logan - Senior Vice President of Worldwide Sales
Aart J. de Geus - Co-Founder, Chairman and Chief Executive Officer
Lisa Ewbank - VP, IR
Brian Beattie - Chief Financial Officer
Joachim Kunkel - Senior Vice President and General Manager of Systems Group
Paul B. Thomas - BofA Merrill Lynch, Research Division
Evan Bloomberg - Sigma Capital Management, LLC
Saket Kalia - JP Morgan Chase & Co, Research Division
Raj Seth - Cowen and Company, LLC, Research Division
Unknown Analyst -
Thomas Diffely - D.A. Davidson & Co., Research Division
Good morning, everyone. I wanted to welcome, all of you here to New York at the Waldorf Astoria and also welcome all of you on the webcast. We're very happy that you came all this way. We've actually had -- we have a couple of people here from San Francisco, one person who flew in from London. So we're very pleased about that. And before we get started with our presentation with 6 members of our management team, I wanted to just go over a couple of logistical things.
First, many of you have probably found the wireless login information on your table. There's also a power strip, either on top or underneath each table if you need it. The facilities are quite a hike. If you turn right out of the room, go past the elevator bank, turn left, go all the way down the ramp and the men's is all the way at the end and turn left, and the ladies' is a little bit closer.
We do have some evaluation forms at your site, so please let us know what you think and let us know maybe what we can do better, what you liked, what you didn't like. We'd love to hear from you. If everybody could please silence their phones. That would be great. I can see many people scrambling. Good. And we do have a mix of familiarity levels here. So we hope that we're going to provide a little bit for everyone, those who have been following the industry and the company for a long time and those who are fairly new.
So our agenda today. We'll start off with our Aart de Geus, Chairman and CEO, and he'll talk about the markets we serve, the market dynamics and how we're strategically managing the company. John Chilton will follow and will discuss how the market dynamics translate into demand for our products and will also highlight core EDA and manufacturing. Joachim Kunkel is Head of our IP and Systems business so he'll do a little bit of a deep dive on IP systems, which have grown to a meaningful scale in our company. We'll have a short 10-minute break and we'll come back with a view from the field, which I know everybody is looking forward to hearing. We'll start off with Joe Logan, who's head of Worldwide Sales and we will follow that with Deirdre Hanford, who heads up our Global Technical Services area. And we will end with CFO, Brian Beattie, who will give a financial update, including some preliminary commentary on 2012. And then we will have a lunch with management.
So this is our Safe Harbor statement. We will be making forward-looking statements and we encourage you to look at our SEC filings for risks and uncertainties that could cause our results to differ materially. We also have some statements at the end of the presentation. And we will be talking primarily about non-GAAP numbers. The reconciliation of GAAP to non-GAAP numbers are on our website and also included at the back of this presentation. So with that, Aart de Geus.
Aart J. de Geus
Well, thank you, and good morning. Welcome to our session here. It's a pleasure to be able to talk about Synopsys and why we should be considered for investment, because we have, all in all, a very vibrant markets that we're playing in, notwithstanding the fact there's a lot of changes in the bigger market around us. We actually touch a part of the economy that almost, no matter what the rest of the circumstances are, will continue to do quite well. Within that, we provide a number of key technologies and we are market leader in that. And last but not the least, we think that we have a picture of a very strong financial presence status and ability to invest going forward.
For those of you that may not be completely familiar with Synopsys, let me give you a very quick snapshot. And many of you are quite familiar with the numbers but in round numbers, we are about $1.5 billion in revenue. And you see the non-GAAP earnings numbers that we've given you guidance for this year. And by the way, our year finishes in October, so we have about 5 or 6 weeks to go. What is unique about Synopsys? We have pioneered a business model and software that essentially rents the software in 3-year increments and thus has allowed us to build a large backlog, has allowed us to have a very high degree of stability. And by the way, we have build a very international company as a bit over around 50% of our revenue is coming from abroad.
Last but not the least, we are a very technology-deep company. We enable the development of chips. So we're in the electronics design automation market. We also have a strong position in the IP, intellectual property market, meaning pieces of a pre-designed chips, and we work also on the higher end of the systems area. That market overall was about $6 billion or so and serving a quite substantial semiconductor market, which is over $300 billion. Which in terms, of course, is at the heart of the still growing electronics industry.
Why Synopsys? One should ask why not Synopsys, given that we are in a very good position, and starting with the markets that's driving things, has been actually quite remarkable. And it's useful to look at it I think a little bit from a historical point of view, because many of you, of course, have seen the coming about of computation at that time Moore's Law was coined, and with computation of course, the software world that's developed around that, we're talking really late 70s, 80s. And then in the 90s, mobile communications came about. And you may recall that mid-90s, there was this notion that convergence would be very important. Today, that's completely old hat, but what it did bring about was an enormous need for networking and connectivity in the world, and of course, the software equivalent of the worldwide web.
Now you take all of those things and you put them together, there's no question that it has changed the very way in which most of us communicate. And actually, it has completely changed the way in which the next generations communicate. And this is particularly relevant, I think, in markets that's go up and down in general, because our kids will give up on clothing and foods long before they give up on their electronics. And so I think there's a high degree of stability in terms of consumption happening there. And part of that consumption is driven by the fact that the world has gone video. I think about 20% of the bandwidth is already used by just Netflix in New York between 8:00 and 10:00 in the evening. But the very fact that our kids communicate increasingly via Skype, is just an enormous consumer of data and data bandwidth, all things that are excellent for the semiconductor market.
Now if you add to that, that we're going to see a next wave where sensors and actuators essentially will add elements to the Internet that are not necessarily human-driven, it is good news that most of the chips that support this are now multicore, meaning that they are really substantial systems on a piece of silicon. And we will, therefore, be able to enter the age of what I would call the Internet of things, where literally your toaster, your dog, your car and so on will communicate among themselves or in a bigger infrastructure. If we look at that, one can simplify this as saying well, yes, really cloud and mobile are 2 sides of the same coin, the more mobility, the more you need the clouds to connect it and do computation in the background. And then this next wave of really smart everything as products make it possible to absorb electronics in them.
If we briefly zoom in starting with mobile, as this is really the most visible driver, you're very, very familiar with that, but there are 2 products, of course, that stand out in the last 4 or 5 years, one is the handset market and the other is the tablet market. They both have the characteristics of just enormous growth and specifically enormous utilization of logic silicon capacity. And even if you take these numbers with a grain of salt and all this prediction, we're talking literally 3-Intel size capacities just to deliver on the silicon of this type of product.
Now the more there is mobility, the more there is creation of data, the more there is also manipulation of data. And with the manipulation comes this notion of cloud. And different people think about different things with cloud. There's those that think about the connectivity between the different product. Others are thinking about the large server site that are being built right next to rivers for cooling and power plant to get the energy. And so Amazon and Google will be great examples of that. These are mega plants of literally tens of thousands or hundreds of thousands of computers. And so the expectation is that as mobile grows, so will the computational cloud.
In addition, with that comes an enormous amount of the transportation of data and where there was an enormous amount of capacity built-in in the early 2000, that is being pretty much used up and now is being amended to. And if you look at the projections for traffic, this is literally measured in exabytes, and of course all of you are familiar with exabytes, it has a lot of zeros. I think 18 zeros on this. While if you look at the storage that's sitting behind that, and that's really the fourth component around data, that itself is actually measured in zettabytes, and of course you're already familiar with that, but for those that are not, there you go. In technical terms, that's called a boatload of storage. So this is an area that will continue to grow at a rapid pace.
Moving on to the third one, which is the smart everything, we can see literally how we're talking smart grids, smart building, smart cars, smart toasters and smart kids and animals and so on, meaning even connection to the body in various forms. But it will has the same characteristics, which is that on top of the traditional chips with storage, communication and a processor, we will see sensors and we will see, of course, the software that ties all of these together. And one of the key drivers is going to be how to bring the price down so that literally in not very expensive products, let's say your key chain, there should be a vessel that can tell you where they are. And these are just simple examples that are technically imminently possible. And I think over the next decade, we'll see more applications there.
If you take one where we've seen this already for a long time, the car, most cars now have a substantial amount of electronics within them. And if you just look at the number of lines of code, you may shiver to know that your car is now powered by 100 million lines of software code designed by software guys. And so I hope you're safe. But the fact is, they have incredibly stringent rules on how to do that.
To make a long story short, if you look at the semiconductor content in electronic products, surprisingly so it has still grown even above the 20%. I was thinking that late 90s it would sort of stabilize at around 20% and move rapidly towards the 25%. And the reality is, what you carry in your pocket instead of chips with some plastic around it and a little keyboard, but the fact is most of the value is actually the chips and the software that brings this together. And so in many way, I think this reflects this electrification of the most of the things that we do.
If we look at these markets from a numbers point of view, we mentioned EDA and IP at about $6 billion serving the semiconductor industry, serving the electronics industry and now increasingly the applications industry, as all of these electronic things are really platforms to build additional value on. Well, in order to do that it follows a very simple rule, and this may sound simplistic but is turning out to be very effective, that you have to innovate and differentiate. And differentiation comes really in 3 dimensions. Do something better, do something sooner, do it cheaper. Sounds very simple but in order to understand electronics, we need to understand what better means. And better invariably today means a tradeoff between performance and power, that's the key theme.
And if you look at the last downturn, it's interesting that in a downturn, immediately a number of people say, "Well, what can I do really better?" And downturn tends to spur a lot of innovation. And you may remember that at that point in time, the iPad was introduced and I always like to highlight that the evening before and the morning of the introduction of the iPad, about 95% of the professional pundits said, "Hey, this thing is not all that great. It's too big to make a phone call with, it's sort of a big phone." And within a matter of weeks, there were thousands of applications on it. And so it illustrates how difficult it is actually to predict which innovation will be successful. But what was clear is, within a matter of weeks, there are whole bunch of people racing to say, well, yes, I'm building one, too. And there are obviously tens if not hundreds of iPads going to the market today. And then today, we're already assuming the Far East in a race to see if one can build things cheaper. And so this innovation wheel is very much characteristic for the field that we're in and we, Synopsys, are absolutely at the center of gravity of helping our customers being able to -- in both innovation and differentiation.
Now if we apply this to the 3 market segments that I illustrated before, mobile is very clear where the push is today. It is a performance power trade-off, but power is mostly fixed. Here's the battery, now how many windows can I open on my thing and how many apps can I have run at the same time. So the push for performance is absolutely one of the key things that Dierdre and her team help our customers work on. If you look at the cloud it's almost the opposite, which is the performance is as high as we possibly can, the fastest possible chips and now the question is, how can you bring down the power because if you have tens of thousands of these processors in a building, you bet they consume a lot of power actually. Most of the cost is now the power consumption.
If you look at the smart, it's a little bit of a combination of all of the above, including the how do you integrate extremely sophisticated systemic problems while making them say, for example, in the car area, or while making them work across different industries such as the electrical grids, smart grids, for example. Now the good news for us is that in all of those markets, this drive for advancement is almost unstoppable. And this is illustrated by looking at some of the economic downturns. So you look at the -- what is it, I think the '89 downturn here, and you can see that R&D expenses kept going in the semiconductor industry. You look at the 2001 downturn, essentially, the same picture. And the most recent one the '95, virtually the same, which is R&D does not go down. It keeps continuing.
And we can absolutely see that, because if you look at the number of chip designs -- and for those of you that may never have seen a chip up close, this is actually what one looks like in terms of size. And this can easily contain close to 1 billion transistors and on-off switch. While if you look at those, the number of chip starts in the last decade have really been roughly stable. And I'll go a little bit later into the color scheme, those reflect the sizes of the transistor used, but the size of the transistors keep coming down, meaning you can have more and more on a chip. But the rough size of a chip is sort of the same. What is not the same is what's on it and the complexity has continued. It's an exponential. It's been an exponential for 50 years, amazingly so. And this exponential is the very reason why, when you have a good idea, you have to race forward because if you're late only 6 months or 12 months, you're way back on the exponential versus your competition. And this is our profession. For the last 25, we can honestly say that for the majority of the Synopsys products, we have been at the state-of-the-art. And I'll say it's not all the time, but in general, we are absolutely the technology leader in our field.
Now here's a different perspective for that and I don't want to go too deep, but for the different transistors, they sort of come in waves. Every 6 months, people refine things and roughly every 2 to 3 years, they go to the next size of transistors. So here's for example, the active designs at 90 nanometer, nanometer is 10 to the minus 9 meters, meters roughly a yard, and so very, very small. And the tapeout is the light gray of the chips that are finished. And the reason we can show this is because Synopsys virtually knows every advance tapeout in the world. We're engaged in all of those. So these were the 90 nanometer, and we stopped counting at 500. We're not a market research company, but these are the ones that we know intimately. Here are the 65 nanometers, so that was the next generation, it's roughly 2 years apart. Here is the 45, 40 wave and then you see the next waves in the work and there's actually quite a bit of activity at very, very advanced nodes. And I would actually argue that right now the race is very, very much on among a number of players to see how quickly they can go to the most advanced devices, including moving from transistors that used to be fairly flat to now turning them sideways and go vertical in order to be able to cram even more together. And you may have heard about the Intel announcement about FinFETs. FinFETs are these vertical transistors. Think of them as a little bit as high-rises, complex to do. But if we can cram more together, we have New York on a chip, so to speak.
Now if you look at the effort that goes into that, and again I won't go deep into all the tabs, but here are -- this is sort of the cost equation of doing a communication chips and the different steps that one can highlight of the task in terms of dollars per month. But really there's a portion that's very much dedicated to building the hardware of the chip, physical reality and a portion that's dedicated to the software portion of this. Well, Synopsys is very much in the middle of this story and there's a reason that the term system-on-a-chip today make sense. A term coined in the 90s because on this chip there can be one or multiple processors, there will be dozens of memories, there will be I/O interaction connectors on the outside, all on a one chip today. In order to realize that, one needs to be firmly anchored in the reality of silicon. Now you make things smaller and smaller, you can imagine that the physics become more and more relevant. We need to also understand the system aspect, which is very much the connection between the hardware, the chip and the software that makes this all work.
And so our business is really predicated on the set of sub-business, so the manufacturing tools connects to the silicon. The implementation, take an idea and design a chip with it. The verification, that's exactly what it's supposed to do, which verify that we didn't make any errors. The system arena connects very much to the software. And in the last decade, the IP, intellectual property pieces are essentially a shortcut of pieces of design that we provide that have already been realized. So that is the fields that we are in and we literally work with all of the most advanced chip designs in the world. I should mention one thing, and you'll see a separate presentation on that, which is the support and services around. As much as we pride ourselves of being absolutely the leader from a technology point of view, I think we also have very strong leadership from a global, technical support and services perspective. Because for people to finish their chips with our tools is actually a tour de force. It's really sophisticated work. And so we have a number of consultants that helps them make the fastest possible chip, and Deirdre will probably give you a little bit more color on that.
Which brings me to the position that Synopsys has starting from a financial point of view. You can see that of the EDA and IP companies in aggregate, we're clearly the largest one. And the good news is that we also are the ones that have been able to gain quite a bit of share during the downturn. And so just like everybody in this room it's very hard to predict where the global markets are going to go, but if there is a significant challenge in the market, I think we will do very, very well given the solidity. If 2012, '13 turn out to be like '11, I think we will do particularly well because we're in a very good growth position.
If you look at the profitability, we're also the most profitable in this industry, and in a minute you'll see why. That is important because the investment profile has to reflect this very rapid ongoing change ,and we have absolutely been able to do that in the last downturn, not just in technology but also in support. And for many customers they have reported to us that, that has actually turned out to be essential for their ability to survive the downturn on their own. We can do this because for many years we have been oriented to be very global. We are probably among the most global technical companies in the world with, not only customers and support in all of the main geographies, but also substantial R&D centers be they in India, Ireland, Armenia, Germany, China, Taiwan, you name it, we have about 10 outside of the United States. We are serving today all of the top semiconductor companies. In many of those, we are the largest provider. And it's interesting to see how many of those top companies are very, very much focused on either processors, mobile or cloud in general. And last but not the least, we can see that in Asia-Pacific, which is sort of a new region, that is already the second-largest today. We see also very high degree of consumption of IP block.
That brings me to the financial model and the way we manage the company, and I'd like to sort of do a little of flowchart on how we think about that. Because there’s always a lot of questions when you have so many changing or moving parts be it in technology, be it in the market. How do you manage the company that has our characteristics? Well, the objectives we set for the core of the company are very clear. The objective is clearly to outgrow the market. And the market we mean both our own EDA IP market and we mean the semiconductor market. Secondly, there's always a question of what are your targets for the ops margin? We should be very clear that for the core of the business our targets are in the mid-20s or even above. And actually for most of those business, we are there and maintain a strong position there. The reason being those are, of course, the funding machines to be able to invest in other parts. Last but not the least, we have decided to have a very stable business model by virtue of these ratable access to revenue and therefore, a very substantial backlog. While this machine creates a high degree of cash flow. And so on average, it's probably about in the mid-$300 million. It can vary greatly from one year to another depending on when the cash comes in, but it is actually a remarkably stable engine. And our cash position today is roughly $1 billion and in round numbers, about 1/3 in the U.S., 1/3 is in other parts of the world. So what do we use the cash flow for? Well, the first thing has to be, how do you make sure that you maintain this engine, this very profitable engine to be at the leading edge of technology because that assures us a bright future, and we spend about 30% of our revenue on R&D. Secondly, we do a number of acquisitions. We probably have done about 55 of 60 in our history. And this is absolutely part of either strategy to broaden our technology, footprint or in general to broaden the diversification of our portfolio. And we definitely invest in making sure that we grow our TAM.
At the same time, it's clear that in an earnings per share count, the shares itself is the other denominator. And so we are also consistently active on the buyback side, and have been able to maintain a flat share count or actually even bring it down a bit recently, and we will continue to do that. If we look inside of the businesses, you'll find the following profile, which is at the core EDA business, which is about 65% of what we do. Is typically growing actually in the low- to mid-single digits but for us has a strong profitability profile and, of course, it's essential to keep it there. Secondly, the IP and systems area is a strong grower, I would say somewhere in the mid-double digits. And initially, we invested from scratch, so it started unprofitable. It is now profitable. It's improving rapidly in profitability. So this is a business that we're actually very proud of because it was created as an adjacency to where we are in an area that is growing, and we have become very strong in this area. This is a very promising business.
Manufacturing, similar characteristics really as the core EDA. We segregated because it's increasingly a different set of buyers, different characteristics and you can see the percentage of the business. Last but not least, services, important but always difficult to drive to a high profit profitability for a simple reason, which is we have support and we have services. Support tends to be free with the tools, services is for specific task and we charge for. And the question is always a little bit, where does one end, where does the next one start. And so one could argue, one should put it together with the core EDA. It is an important part of our business and we'll continue to hone -- own it for either competitive advantage or profitability.
If you look at our results, we have already given you a guidance for this year and we'll deliver double-digit growth and that we are very confident of that number for 2011. He is the very predictability of the business model. And by the way you can notice that even in the downturn, which for us is more mitigated because it's a bit longer time, we were able to continue to grow the business and as a matter of fact, we grew substantial market share during that period of time. The non-cancelable backlog is more than our yearly revenue and because of that, we enter a year typically it was 80% of revenue in hand, we enter a quarter with 90% of revenue in hand. And I don't know that there are many companies in the world that can do that. So from a stability point of view in tricky economic times, we're quite unique. And as mentioned, we have been able to grow our revenue for now quite a number of years successfully.
If you look at the profitability side of the company, we will deliver double-digit growth on EPS this year and are quite confident of that. A number of you have commented about the fact that we should pay more attention to the ops margin. We are paying a high degree of attention to the ops margin. We expect it to go up a bit in the coming years. The reason it was down is twofold. We did invest in a number of acquisitions that typically come at some cost and of course, we went through a recession in which we decided to continue to support our customers. If one looks at M&A, M&A is great to increase positions in area one is not in or strengthen the technology, but it has an economic challenge. And the economic challenge is typically threefold. One is that on the first 12 to 18 months, there is a – counting the cash flow is just fine, it's just the impact on the revenue is not recognized. Secondly, most of the things we acquire are not anywhere close to our ops margin. We are actually doing well in the context of the industry that we're in. And third most of the companies we acquire do not have a multiyear business model, and so transitioning them puts pressure on the ops margin. And so this is why I think it's important to realize that on a consistent basis for the businesses that we have had for a period of time, we do manage in the mids or even above 20.
Cash flow. I've mentioned this already. I think you can see that it can vary greatly from quarter-to-quarter but it's overall has been remarkably stable. And of course, is part of the picture that allows us to be solid virtually with any economic weather around us, and also jump on opportunities when they present themselves. We have no debt, which does not mean that we are adverse to debt. Obviously, today the price of money is particularly low and there maybe opportunities, but it is also true that we have managed the company conservatively, and that we will take risk if there are great opportunities. Otherwise, we will tend to just focus on the traditional utilization of cash, which is the operations, M&A, which we have done quite a bit of and a systematic buybacks.
From a current operating model, we see right now the growth looking forward to be in the high-single digits. This year we will do better than that. If the economy looks good, we may be able to do better next year. But right now, this is the plan we use. We can say that in the IP and systems, we will certainly exceed on that. You can see that we will probably continue on M&A, but this is very much a function of what is available and when. We will continue to drive the efficiency of the company and we're quite sensitized to that as there are a lot of movements in the world economy that would indicate that one has to look at. Which regions to be in? Now the Far East regions, their pricing schemes and their compensation schemes are changing quite rapidly. I think we're quite agile in this experience. And last but not the least, we may be able to actually do a little bit better than maintaining a flat share count and is only on a bit of a track to reduce it somewhat.
How does this add up? Well, really, the only big question mark is the economy. And our sense is that sort of no matter what, within reasonable parameters, we will be able to continue to push on growth. We will solely push on operating margin. And last but not least, reduce the share count a little bit. If there is a downturn that's substantial, we will absolutely look at the opportunities to gain market share. Bottom line of all of that is without wanting to give guidance now, given that in December, we'll do that after we finish -- or late November, we'll do that after we finish the year, but fundamentally we're comfortable with the consensus call that you have.
So summarizing all of those, we are actually in a market that notwithstanding the up and downs, we'll continue I think in fairly heavy consumption of electronics. Within that market, there are right now quite substantial food fights for leadership. That is great news for us. The race for technology is on at many, many places. And this is through all the way to the depths of -- physics of the transistors all the way to the system development. In that, we are clearly the industry leader, both technically and economically. And the 2 cannot be viewed disjointly, they support each other. Our financial strength does show that we will grow this coming year, again, quite well. We're looking at good profitability. And last but not the least, we are very committed and maybe increasingly committed to see how do we make this reflect in shareholder return. We're well aware of the fact that as a company, I think we've done quite well. As a stock, not sufficiently so. And so our own attention is increasing for that.
With that, let me pass it on to John Chilton, who is our Senior Vice President of Business Development and Marketing, and who can give you a little bit more color on the market.
I should also mention that we'll be having a group Q&A session after Brian's presentation.
Thank you, Aart. Okay. I'm going to talk a little bit about core EDA and manufacturing. These are our traditional businesses, Aart called this traditional EDA. I want to first start with a little bit of perspective on the vibrant markets from a -- and give you another view on that from a sort of a supply-and-demand perspective. To clarify that and I want to show you how that is a driver for this core market of ours. If you look at demand, Aart talked about 3 areas, he talked about smart everything, he talked about cloud and he talked about mobile. I want to focus just on the mobile area and the demand, the upcoming demand for these smartphones and tablet devices really is quite explosive. We're looking at demand from the devices themselves in terms of units growing 8x to 1.4 billion units by 2015. That represents end-market revenue approaching a good portion of $1 trillion. It's a really kind of unprecedented market there. And non-memory semiconductor revenues as Aart pointed out going into that of $100 billion. So really the end-market demand is there, it's growing, how about the supply? In order to design, to develop, the manufacture of semiconductors, it's quite an expensive thing developing these fabrication facilities, and you have to do it well in advance. So if you look at the leading for fabless companies, you got the leading 4 companies investing in 2011 over $24 billion in CapEx. And this is not a onetime investment that happens only in 1 year. They invested a huge amount last year, they even invested in the downturn and they'll invest a huge amount again next year. If all that capacity comes online as planned, that represents a wafer volume increase of about 64%. If all that turns into revenue, that represents a revenue increase for these foundry companies of about 82%. So the supply is definitely there.
So what you have is, you have a supply industry investing tens of billions of dollars to be ready for this end-market opportunity, which is looking at good portions of $1 trillion opportunity. The trouble is that these wafers are not useful in things like phones. They're just wafers. They don't have anything on them. So turning those wafers into chips is actually the process of design. And that's what our semiconductor companies do. When you've had that much money at stake and you're going after that large, you want to make sure that your partner in design with a leader. These design companies -- these semiconductor companies that are designing these chips are doing it in concert with their design partners. We're the lead design partner in the industry. And let me talk about that just a little bit, talk about that leadership, talk about in terms of segment share and technology.
First thing to realize is that these designs are very expensive propositions. A leading-edge design in 32 nanometers is now about $100 million proposition. In other words, when a company looks at designing a chip for those very vibrant end markets, you're looking at $100 million project. So that's important to the company for a couple of reasons. One is, it's a huge allocation of their resources. It's probably a good part of what they're going to be doing that year. Secondly, it's going to capture a socket, which is very, very valuable. So it's really kind of a bet the company move. You cannot afford to lose. So you want to go into that kind of a project partnered up with the right people. Obviously, you're not looking for -- you're not looking to save a little bit of money by going with the second tier design partner. By the way of that $100 million it's important to note that the EDA tools, the portion that we get is actually relatively small. It's about 10% or less. So most of those dollars are the engineers in the companies doing those designs themselves.
So let's look at who the design partners are that they have to choose from, what does our industry look like in just a little bit of detail. Well this is us and the next 3 public EDA companies in terms of the last 4 quarter revenue as reported in the categories that they report them in. You can see that when it comes to being a design partner, partnering with the company to help them get that chip done, we have about $1.5 billion, $1.481 billion of revenue directed in that area. That means massive investment in R&D. That means support engineers around the world where you need them. The number 2 company is about 60% of our size, so quite a bit smaller in that. That means fewer R&D engineers. That means fewer support engineers. You have one more and you're down at a company that's about 1/3 of our size when it comes to helping you get those chips done and then from there, it's very, very small. So if I'm going to those $100 million investments going after those near-trillion dollar markets, I certainly want to do it with a vendor that's almost 2 or 3x the size of the next candidate.
So Aart introduced this diagram. These are the way we look at our platforms. We have system-level designs, then what we call the core or RTL design, which is made up of our core tools plus IP, and then the design is done, you send it off to manufacturing. I want to focus here on the core EDA segment that's verification and implementation and in manufacturing, and give you a little bit of data on that, a little bit of perspective under that. As Aart said, core EDA is 65% of our current revenue. It's a $3.6 billion market. We lead it by far with about 30% of the total and it's a highly profitable market. That business is made up of several product lines, most of which we are the clear leader. So you start that design process and synthesis, Synopsys actually started 25 years ago now as the Synthesis company, and is a very clear #1 there. You then do the detailed design in place and route, we're the clear #1 there. That's actually the largest revenue producer in core EDA. You make sure that design is correct in timing or signoff. We have a very high share there, we're #1. And you'll check out the detailed timing with very detailed, digital and analog verification. We're the clear leader in both of those. The only 2 of these segments that we don't lead are in physical verification and in custom-design. In physical verification we're the clear #2 and we have, we essentially redesigned the category and are picking up tremendous share there right now. In custom design, there's a long -- one of our competitors is a longtime -- practically the whole market. And that's a little bit of a challenge for us because as people want to consolidate over to us as the leader they want to make sure that we can satisfy all their needs. And so they're not, let's buy one thing from one vendor. So we've had a large organic effort for several years in custom. We're quite happy with how that is going out. We're getting good adoption there. So we're the leader in most of the categories and we have very good momentum in the 2 categories where we're second.
In order to remain the leader, what our customers are really looking for is they're looking for us to lead with the leading edge. So in other words, if you're going into a 28-nanometer design, you want to know you're doing that with a toolset and design partner that's done that many, many times before. And if you're looking at 22-nanometer design, if you’re on the real bleeding edge, you want to make sure that you're doing that with someone who's gotten there already, or at least is used to working at the leading edge, has the resources to work at the leading edge.
So to look -- just look at that historically a bit, back at the 45 and -- 45, 40 and the 30 to 28 nodes, the current -- kind of the node, we announced leadership positions with most of the important companies. Here are just a few of them, most of them got their first with us. We already see that now at 22 and 20, Samsung and FT have already announced that they've done their first chips in that generation with Synopsys. You also want to make -- in addition to the process, you also have the key IT, and you want to make sure that you are -- that you're also working with someone who's very experienced in the IT that you care about. Because, essentially, as Joachim will talk about, your designs today are made up largely of blocks of predesigned IP.
One important IT block there is the ARM A15. We signed an important agreement over the summer with ARM, sort of solidifying our leadership in being able to help our customers get ARM A15, their latest core designs done predictably, and with the highest possible performance. And then as you'll see, we have quite a bit of IPLs -- I'm sorry, quite a bit of IP ourselves. Here's one example: PCI Express 3.0, a brand-new, very high-speed interface specification, which we were the first to actually certify. So clear leadership there.
When you look at the manufacturing segment, that's about 12% of our revenue. It's a $430 million market. We are far and away the leader at 40%. It's a very highly profitable business for us. That business is made up of 4 areas: Mask synthesis, where we're the leader; mass data prep, where we're far and away the leader; yield management, which I'll talk a little bit about it later, which we've essentially pioneered; and lead in and TCAD, which is the business of simulating the actual atomic properties of those horizontal or vertical transistors, where we're again the leader and have a very high share. So these are the fundamental technology that you need to build those multi-billion dollar fabrication facilities and get those wafers up and running, and that scenario where we'll be essentially in every segment.
So let's talk about growth here. So going a little bit more in-depth than Aart did on the growth. This is a -- we think this is a very solid area. We see this growing for us in low- to mid-single digits for the next many years, and we'll get that growth at about segment growth and share gain.
Segment growth comes because the designs that are being done now are much more complicated, so the amount of software, the amount of automation you need to get those done does go up. There is a heavy demand for new designs and new devices. And there is quite a bit more complexity coming. Share growth comes essentially as people move from the second-tier providers to the first-tier providers. It comes from a -- and it comes because of our experience, because of the position and power of our tools, because of the new products that we're assuming.
So let me now going into a little bit more detail on those technologies, talk first about verification. Aart introduced this diagram of better, sooner, cheaper. It's certainly true that people buy our products in order to get their products sooner, get their products out cheaper or make their products better, so is verification. Verification is, really, primarily about lowering the cost of design. You might have noticed on that chart, where I talked about the $100 million design cost, that the absolute highest cost -- the highest portion of the design cost is software. Joachim will talk about that a little bit.
The second highest design cost is verification, and that's quite a bit higher than any other portion of the hardware design paradigm. So what customers want to do is they want to reduce that cost. They want to get their verification done sooner. They want to deploy fewer engineers on verification. They want to deploy fewer computers on verification. It's a very expensive part of the design process. So it's mostly about cheaper, but not cheaper chips, cheaper designs.
It’s also about getting it done sooner. Since verification is the long haul of the design, if you can speed it up, you can get your design sooner, have a better chance of capturing that high-value socket out there. So when you look at the verification area, we have a very comprehensive solution, 90% of the 32-nanometer designs were verified with our main functional simulation -- simulator VCS. 60% of the previous nodes designs had been verified with our functional simulator VCS. Every one of the 10 largest chips designed to date have been verified with our functional simulator VCS. 12 of the top 20 semiconductor companies use VCS. And then when you get down to the detailed simulation, CustomSim, that's the sort of the analog simulation, they were the clear market leader. 19 of the 20 top semiconductor companies use CustomSim. And then you get to the real detail, sort of "one transistor at a time" simulation, that's HSPICE, and naturally the golden reference out there, and has been for actually over 25 years. That's actually a very early product that really pioneered the whole EDA revolution that we acquired and continue to invest in.
So this leadership is driven by a number of innovations and differentiations, which I won't go into in the interest of time, but we can talk about it at lunch. But there -- you can see they're really all focused on speed. They're focused on making those tools faster, getting a de-bug faster, getting better models, getting your verification done faster.
Next that I'll talk about is implementation. Implementation, while not a very expensive part of the design cycle itself, is very important, because it's the area that will actually lead to a better chip. Better tools will give you a better chip. A better place and route tool will give you a chip that's smaller, faster and uses lower power. Because as Aart said, these chips can have 1 billion transistors in them, so you're not doing them by hand or essentially being designed by computer, guided by engineers. So the better algorithm in the computer will give you a better chip. That will give you a better chance of that socket. That will give you a better chance in those end markets. That's why this is the largest dollar segment in core EDA. So here you can make better chips, but you can also get those chips done sooner, because you can hit your design goals sooner with better tools, easier and, of course, when you're done, those tools will be cheaper -- or those chips will be cheaper.
So we have a very comprehensive implementation solution, the most comprehensive in the industry. 90% of the 32-nanometer designs were taped out using Synopsys. 70% of the 45-nanometer designs were taped out using Synopsys. Eight out of the top 10 semiconductor companies use us, as their primary implementation tool vendor. And there are number of innovations that we brought out in the last 12 months or so in this area, focused on enabling the designer to make a better product faster.
Finally, I'll talk about manufacturing. Manufacturing, now you have the design done, now it's primarily about manufacturing -- and remember manufacturing is -- even though design is expensive, manufacturing is much more expensive. Here it's about making cheaper chips, and to some extent, getting them out sooner.
In manufacturing, we have a clear lead. Our mask tools are used by 8 of the top 10 semiconductor companies. TCAD tools are used by every major semiconductor company in the world. And yield management, which is an area that we just pioneered is already reducing production yield and essentially changing the manufacturing process at the leading foundries already, in order to allow our customers to get these chips out sooner at higher yields.
To give you a quick illustration of what that looks like, here is just a small chip, not in the most advanced process. You can see the blue line here is the experience that customers see if they're using traditional methods. No yield enhancement capabilities. And you can see that they might get up to a 94% yield, that's means 6% of the devices on those wafers are bad and have to be thrown away, and it takes some number of months to actually get there.
Utilizing these yield management tools, which can -- which use the information developed during design to precisely target which portion of the chip is faulty, you can accelerate that time yield, make your customer much happier, because you can deliver your chips and volumes sooner, but you can also get to a higher overall yield. So in this example, which is a chip with a total COGS during its lifetime of about $50 million, you can save about $3 million. So you can imagine taking 5 points or so off the COGS is important to any manufacturing operations. So this is a tool we're seeing rapid adoption of and we're quite happy with.
Finally, we've invested in a new area, and that's software around optical. We're doing this because we're very interested in the LED market. We see the LED market as a market that's going to grow quite a bit over the next few years for general lighting and other applications. It's another semiconductor market, and it's a market that's going to need tools. So we acquired the lead company here. We now have a product called LightTools, which is the premier application for illumination as essentially viewed by every LED company in the world.
So that's a little bit about core EDA. You can see that we are going to some very vibrant markets here, markets that our customers do not want to take any risk in attacking. They don't -- they want to partner with the leader. That's a dynamic that we see out there. We can see the industry leadership that we have essentially across the product line. And we are quite well prepared for the more advanced technologies that are coming along with the new customers already.
Okay, that's it for me. We can certainly talk at lunch and in the Q&A, and I'll turn it on -- turn it over now to Joachim Kunkel, who's our Senior Vice President for IP and Systems business unit.
Okay. Thank you, John. What I'm going to be doing here in the next 20 minutes is to talk about this IP and systems, which is a business that we have been developing over the last years and has become a significant portion of the Synopsys revenue. Just to position this again, John has been talking about the verification, the implementation, the manufacturing piece. I'm going to be talking about the rectangle in the middle called IP, which are predesigned building blocks that go into the chips and about the systems portion.
Quick profile of this. It's -- represent about 20% of the current revenue of the company. We estimate the growth to be a double digit and continued double digits. What are the drivers? The drivers are number one, outsourcing of some activities by our customers, but in particular, also, the tremendous increase of software content on the systems-on-a-chip devices. We talked a little bit about profitability already. The profitability has been improving in the past and will continue to improve. And the business is profitable overall.
Let's talk about IP. The market segment of IP is about $1.7 billion. That's what [lay in. That's what we're addressing today. We're the #2 at about 14% share. And I'll give you some more detail about this. So what's driving this? It's really outsourcing. And what's happening is that on many of these chips, there are a number of common functions that you can find pretty much on every single piece of silicon.
Just look throughout one word like USB. Everything today that -- or almost everything has some sort of USB input and output. While companies don't want to design this or companies don't need to design this, their product is not going to be different, USB just has to work. So we take advantage of these opportunities and provide them their USB predesigned. They put it on their chip, and then they take their engineers and start developing something unique, this will allow them to differentiate their chip from their competitor's chip. And so the whole thing comes down to basically a trade-off, to make versus buy. And it's way less expensive for them to buy it from us and take the engineering and resources and put them onto something more productive and more differentiating.
We estimate that about 25% of the overall IP TAM has been outsourced so far. Those are conclusions that we read when we talk to some of the industry analysts. And it's pretty consistent with the picture that we also have when talking to our customers.
If you look at the systems area. The systems area is about $400 million market with the #3 at 20% share and growing there. We look at this as being an emerging market. It's still very fractured and fragmented, but some themes are starting to show up there. And it's basically the theme of complexity. These systems on a chip are very complex systems on a chip that consist of hardware and software that has to interact with the environment. And again, the exploding software content that makes it very, very difficult to bring these systems from a chip to market on time in order to hit the market demand wave.
So going into the IP market segment. If I go clockwise around, you'll see the microprocessor segment is the largest segment, actually. This is where companies like ARM and MIPS are active in. It's followed by DSP. Well, it's just a company [indiscernible] that you may be familiar with. Fixed functions is -- mostly speak for GPU, graphical processing units, and clearly, a big player there was a company called Imagination Technologies that you're probably also familiar with. And then going further to the left, we go into the interfaces, memory blocks, analog IP, physical libraries, sub-standard cells, et cetera, et cetera, et cetera. The market that we address -- or the market segments that we address is pretty much the left side of this picture. So starting with the interfaces and then continuing with memory cells, et cetera, et cetera, et cetera.
So let's take a look at the chip. I was talking about the fact that a lot of the design is being outsourced these days. So this is what we consider to be a relatively representative chip for what's being designed today. And again, going clockwise around, we first have the CPUs or many times, the chips that are coming to market today have like dual A9s or dual A15s or quads, right? That's where we're going. Of course, every CPU, every central processing unit or microprocessor on a chip, needs some sort of memory to work off, right? These are the RAMS and the ROMs. And many times, there's going to be some graphics engine on it.
You will have many of these electronic devices from flash memory, so you need some flash interface to do that. The main memory in all of these chips or for all of these electronic devices is DDR these days and different alternations, DDR, DDR2, DDR3, LPD DDR, et cetera, et cetera, et cetera that becomes an alphabet soup, but it is basically the interface to the installed memory. And many times there's some analog functions on this.
And as you heard Aart talking about the fact that everything is sensing the environment, it's talking to you, it's listening to you. You need for this, some audio capabilities, some video capability, data converters for the in and out of the analog information like sensors and actuators and then the very well-known interfaces like USB. I already talked about this PCI Express, which is very widely spread in the computer industry and computer chips data for disk drives, HDMI, so pretty much everything that has to do with video entertainment, whether it's TV, your set-top box or your phone in the meantime. And then there's a new class of interfaces called MIPI interfaces, which is actually an interface that's starting to gain ground in the area of mobile applications.
So what's the opportunity for us? Well, if we met our portfolio onto this chip, everything that's purple are things that we developed, and that if you add up how much these things cost, you start seeing a potential of about $2 million in IP for each of these designs that look like this. Now not every design looks like this. Some of them have few opportunities for us, some of them have more, but this is a pretty good number to go by.
So where are we, when this has put us in the market? We already mentioned that we're the #2 in the IP market, behind ARM, who's the clear #1 though. Imagination Technologies is #3, doing DPUs, MIPS doing processors, CIBA, et cetera. But what you can also see is that we are clearly ahead of #2 -- of #3, and clearly, ahead of #4 and 5, et cetera.
So how have we've gotten in this position in the market. What we've tried to depict here is starting in 2005, and we had no space, we kind of lumped 2005 and '06 together, all the way to 2011, everything in green I think we said we've been developing organically, in terms of IP building blocks. We call this the portfolio. And you can see that we have very few building blocks at the very beginning and kept them building more and more and more, which allows us to put more for our building blocks onto chips with our customers too. We've been growing our market position not just only organically, but also for acquisitions, and these are the blue boxes that you can see on top, we've done a few of them. The last one being the position of Virage Logic that got us into the embedded memories and spent us a little library business. And you add it all up, we've been showing from 2005 to today a revenue growth of about 3.5 times, which is a 19% CAGR, which I think is pretty decent.
So we have the #2 overall in IP, but, of course, we also have leadership, which means #1 positions in terms of market share in some of the segments. And I would like to highlight here in particular the Interface segment, we were the clear #1 and also the Embedded Memory segment that we're also the clear #1. These are 2 very important market segments for us within the IP business. We're also #1 in the analog IP smart segment, although this start becoming smaller segments here.
So how do customers buy IP from us? And let's always remember when we talk about IP, we're not really talking about Intellectual Property Rights, but what we're talking about is predesigned semiconductor building blocks. Well, the easiest situation is what we call the single-use model, which means a customer wants to design a chip, he finds out that means, he needs some USB interface and he needs a PCI Express interface and the SATA interface, this could be maybe a fraction of a set-top box, and we give them the right or some money to put our USB, our Saturn, our PCI upon this chip. Wonderful.
And how do we recognize the revenue? Well, it gives us a PO, we regularly we ship the deliverables to the customer and we recognize the upfront. Okay?
Now many of our customers, of course, don't do one chip a year, or one chip every couple of years but they do many, many chips. What they do is they look around then start serving their companies as to what are you going to be needing, the various design teams, in the course of the next year, 1.5 years or 2 years, because they can have some pretty good visibility into what they're going to be doing. And they come up with a whole set of control, why we need this, and we need the other need, we need that.
And you should also have most of this and to stop negotiating what they call a multi-use agreement. The multi-use agreement is basically a pool of IP that they started drawing down over time.
Now these agreements are then also multi-year. They draw down the IP over a fixed amount of time, so it's a use it or lose it, if you want. And now nobody likes use it or lose it, and particularly, nobody likes lose it, so what they will do is if they need x over the next 3 years, or if they believe they need x over the next 3 years, they will buy less than x for the next 3 years. Now we find ourselves, many times, customers actually having to re-up these type of agreements way before they have expired. And this is fantastic. It's a great situation to be in. We're not in this renewal business in IP, but we're in this -- it is a business that's driven by the actual design needs that customers have for the chips that they're doing.
Now on top of this, which is the licensing part of the business, we also have an engineering services business and this happens in cases where, for instance, a customer who has been doing a chip so far in the 130-nanometer node, just to take something really big, and they want to do a cost reduction, that is to go to the 110-nanometer node. 110, for most of you who are familiar with the semiconductor industries, this is not necessarily a main node. Our IP may potentially not exist for that node. But the customer was a customer for IP at the 130-node. They said, "Look, I mean, with some options, we would like to use this again for 110, but we need to find some sort of arrangement so that they could port this to the 110 node that's available, because he's willing partner for the next chip. And that's what we call an engineering service. We will port for that customer our IP for the 110-nanometer node. In the same way, we also do this for some IDMs. We have proprietary processes, are doing special process variance and just would like our IP to become available for something, but there's no real market, but the market is just one particular opportunity at the customer. This is not a major portion of our business, but obviously, as our customers depends on our IP and cannot get the chips out of that IP, we have to be able to offer this. And we have to be able to say yes to customers, so they can get their chips out.
So to summarize where we are on IP, we believe that we have a very comprehensive portfolio. In particular, what we offer customers are complete solutions consisting of the digital portion of their design, the analog signal portion of the design, the verification IP, any driver, software that's needed that make it very easy to integrate that functionality, that predesign IP onto their chip, and in particular, what it also does is it lowers the risk of the project.
If you're putting USB on your chip, you're putting it because you need it. For USB customers, there's no upside, there's only downside, if the USB does not work, so your chip does not sell. We protect you from that downside. Our IP has been used hundreds of times across many, many different customers across many, many, many different processes, so we lower the risk of designing in that function. We're dedicated to the developing high-quality IP. And this is really one of the drivers behind the adoption of IP. In the past, there has been IP available, but sometimes their quality was doubtful, which means that customers would have a chip that will have marks that would not work for all application cases, they might have to re-spin, re-tape out. Obviously, if that's the case, the large corporations will not enter that risk and include that risk and by third for the IP they will develop it themselves.
The barrier to adoption of IP has been lowered because company like us develop some very, very high-quality IP that I would dare to say is actually higher quality than what our customers can do themselves. Not because of our design skills, but because we can afford to invest much more in developing USB than any of our customers can actually do. We sell the same thing to everybody.
The business is growing. It has been growing for quite a while, and we see no reason why it should stop growing. Our customers are counting on us for the long haul, and we have about 1,300 engineers developing IP. This allows us to develop a very broad spectrum of IP, so a broad portfolio, but at the same time also to make sure that our IP is available in the semiconductor processes that our customers need it in, which -- or gives them the freedom to pick who is the best semiconductor partner for them, who is the best foundry partner for them.
Moving on to the system design arena. We are the #2 provider of system design solutions, in particular, in the area for accelerating software development, hardware, software integration and system valuation. Overall, I said we're #3, but in this particular area, that's the area that we're focusing, that's probably the #1. And we focus on this area by the means of 2 tools. On the right-hand side of the gray shaded area, one's called Virtualizer, which we announced recently. And Virtualizer is really the result of combining technology from some acquisitions that we have done lately in the area of system level design. Virtual, which we did a couple of years ago, VAST and CoWare. And then also the other product that we have. There was HAPS, which is FPGA prototyping product.
So why is this whole thing interesting? Well, imagine you're system-on-a-chip, designer and you're going to be committing $50 million, $100 million to developing a system-on-a-chip. You want to know if this thing is working soon as you can. And one way to find out if this is working is, of course, you do a lot of system simulation or verification using our tools like VCS. But also these things have to interact with the environment, and you don't really know if this is working until you've been able to try it out somehow within the environment of the chip. And one way to do that is to take your design and before you commit it to manufacturing a chip, you do a prototype based on FPGAs and the FPGAs are coming from companies like or XILINX or Altera that you're probably familiar with.
So this is a very popular technique and customers use that in order to validate the system and to get the hardware and software and also do some software development. And the technique is not limited to using our products, in this case, HAPS, but this is a pretty popular way of doing things among semiconductor companies and many of them have been building their own FPGA prototyping boards, et cetera, et cetera. And so far, we believe that only about 30% of that market has been outsourced. So we have some big opportunity to continue growing into that, because obviously, we're able to provide a much more complete, much more effective, cost-effective and also development efficiency -- efficient solution to customers than what they can do themselves.
Now some customers said that's very good, "But you know I need to have designed the chip before I can do an FPGA of prototype, and I would like to be able to start my software development even way before I finish the design of my chip." And that's where virtual prototyping comes in. Virtual prototyping is basically a big simulation of the chip, it’s just a big simulator of your chip that runs very, very fast because it runs into very high level of abstruction. And you can do things like booting the operating system in nano seconds, so it really allows you to do software development on this. And this is a capability that's been used now, in particular, in market for -- the time-to-market pressure is tremendous. And the mobile application process on market for instance, sort of -- everything related to mobile phones and tablets fits this profile. And we have some very, very good position -- market position in terms of customers who are using our virtual prototyping capability specific in those markets.
There's another market where the time-to-market pressure is not that high, but reliability and safety aspects are predominant, and that's car autos industry, the automotive industry, and we also have some very good position in the mean time with our virtual prototyping tools in the automotive industry.
Now this is -- while the FPGA's prototyping methodology is well entrenched, and this is a pretty good market that -- where the outsourcing continues to grow, the virtual prototyping, we consider to still be a nascent market. There's still many, many application area for this technique, it's not being used at all and this is a great opportunity for us to continue moving forward.
Now one thing that I would like to mention is that in order to build these virtual prototypes, you need to have models of the building blocks on the chip. And in the end, it comes down to having the models of the processor. In many cases, what you need is a model of ARM processors and we build those models. And you need to have models on the other IP. And the other IP happens to be, in many, many cases, our own IP. And obviously, we're in a fantastic position to develop models of our IP, to put the whole thing together and give them, the customer, a model or assimilation of the whole chip.
So does this whole thing work? Well, I would like to draw your attention to the example on the upper right, Atheros. Atheros is a company that's been using our high performance prototyping capability for a long time. And as you know, they're a part of Qualcomm now, and that is a company in the wireless space. And for them, the whole thing was to be able to integrate the hardware and the software and validate this combination of hardware and software before they got the chips back, so that when they get the chips back, they just take the software, put it on the chip and magic happens, everything works as expected, because they were able to develop the software and integrate the hardware and software before they have the chip ready. And they do this on the FPGA prototypes.
On the lower right as an example from Mazda, an automotive company. What they have been doing is they've been using our virtual prototyping capability in order to develop the electronic control units in the cars. That's what controls basically the engine in the car. And there's 2 reasons. Number one is they want to do this way ahead of time that helps them. But in particular, what they want to do is want to make sure that they have enough time to make those easy-use safe and reliable.
One the left -- In the case of Renesas, that's very interesting, because what they did is they have adopted already FPGA -based prototyping, and then also adopted virtual prototyping, and they say, hey, you know what, actually things are getting even better: If we combine these 2 things, because we will be using FPGA-based prototyping for this portion of the design that we already have designed and we will be using virtual prototyping for those pieces of design that we haven't designed yet. We'll put it all together, and we can still run and prototype everything ahead of time.
And with this, I would like to finalize my section and let you have a 10-minute break until we come back here at…
We're giving them a few more minutes
Food is still across the hall, and we'll have beverages here in the back, and we'll see you in about 10 minutes.
Thanks, everyone. We're getting to get started with Joe Logan, who's Senior Vice President of Worldwide Sales.
Joseph W. Logan
Welcome back from the break, everyone. As Lisa just said, my name is Joe Logan, and I head up Sales for Synopsys.
One thing we're not confused about at Synopsys is what's important to us -- and that's most important to us, and that's the customer, so why don't we start there. John and Aart gave a view of how much opportunity's out there is out there for our customers, and there's a huge amount riding on every big win. So the right big win and the great execution can drive enormous revenue for those big wins, but those big wins don't come without a tremendous amount of pressure and risk. John mentioned $100 million for the largest most complex chips out there, so huge dollars at stake, schedule pressure and people pressure. It's a very dynamic world out there for top engineering talent.
So it's that combination of pressure and opportunity that really force our customers to focus on what's most important to their winning businesses, which is their own differentiation that provides opportunity for the EDA industry and IP.
So whether customers -- what's the expectation for both our industry and Synopsys in particular. They are willing to pay for the value, and particularly, the value of differentiation. And unfortunately, differentiation changes from customer to customer. Every customer is trying to optimize one of the 3 -- or waiting these 3 characteristics: better, sooner, cheaper, in some different configuration. So for instance, some customers are after the absolute smallest die size, some customers are after the lowest power, some customers are after the highest performance device, so our job in the field is to try to figure out what differentiation makes the most value to our customers and try to represent that. One thing that all our customers appreciate and that is support, particularly, the ability to optimize the design and get the design out on time with the highest performance.
To -- quality is important across our entire product portfolio, but in particular, IP, where the margin for error is extremely small, so close support and high qualities are very important.
In terms of business interaction what our customers expect from us is really just a very direct business models, clear business model, clear positions and negotiation. We view our customers in 2 categories. The first one is name customers of which there are about 150 of and that represents about 80% of our business. The balance are territory customers, and that represents the remaining 20%, we sell primarily direct.
In terms of sales organization, we're really a combination of account ownership that manage the overall business in relationship and sales specialists. The specialist brings deep technical knowledge to each of the categories that they represent, namely IP, core EDA, manufacturing, systems prototyping and optical, where optical is the only actual separate channel that we have at Synopsys.
Teamwork has become increasingly important, more important over the years. And the main reason is how our customers distribute their workers, their decision-makers. Over 75% of those 150 named accounts have multiple sites in multiple regions, so that means decision-makers, users are in various parts of the world, and it's not uncommon for our largest customers to have 9, 10 sites of users, decision-makers, influencers. So what we've done from a teamwork perspective is per named account -- is to have a lead relationship owner, which allows then a the structure of cooperation across the sales force to sell in the multiple geographies.
We're a very broad sales organization because we are very broad company in terms of our reach. We have 87 different sites around the world in 27 different countries, so we clearly, in all cases, are very close to our customers.
And then the last thing I want to highlight on this slide is to talk briefly about our Asia-Pac team, which very important team as they're selling into the highest organic growth part of our business. Our Asia-Pac team of the 6 regional country leaders in Asia-Pac, 5 of them have been with Synopsys for over 9 years and the regional VP for over 15 years. So it's that the combination of stability, experience, combined with organic growth that I think is really allowed us to achieve good success in Asia-Pac.
The breadth of solution is something that any sales organization want. So to represent that, let's take a look first at -- from product idea to end product.
We first start with what are really the roots of Synopsys, which is the verification and implementation of a hardware-description language going through a series of steps and terms of things like synthesis, place and route, a lot of analysis, verification and out comes a description, which can then be fed into a series of manufacturing products that John talked about. So that's -- these are 2 large core businesses of ours, and Joachim talked about how we've been able to grow a very large commercial IP business that complements very nicely our customers' development of their own differentiated IP.
Completing or moving forward with the breadth of the solution is we've invested a lot of effort and acquisitions and focus on both the hardware and software prototyping system. And what these do is link a -- the hardware representation to the software-development effort to be able to start software earlier, the benefit of getting better validated hardware and also getting a jump on software development. And then finally, we have now an optical speeds, and clearly, what's in an adjacency, and we hope to expand those adjacencies in the future.
So what are the benefits of this breadth. To Synopsys, the benefits are more contact with more decision makers, being able to provide, sell more products to more organizations within the customer. The benefit to the customer, it allows them to focus more accountability on fewer key suppliers, and clearly, we strive to be one of those key suppliers. And from a channel or sales efficiency, it's obviously -- we're not looking to greatly expand the sales organization, in fact haven’t -- we've been able to have more product through the same sales channel.
Although having a broad solution is a great thing, we also understand that we need to realize the individual value of the technologies across this solution. So how have we done this? We've done this by focusing on logical boundaries based on customer usage and customer decision-making. So categories such as systems and IP, core EDA manufacturing and optical are aligned with the customer and that they accept that these are boundaries where separate negotiations, separate value discussion and separate our contracts are developed.
So let me look briefly at the composition of our business. On the bar graph, you see the layers of our business which add up to our guidance for the year and 80% of this bar graph we believe is renewal business, a renewal business being software where we expect the use to continue after the expiration of the contract, and the balance 20% are now over $300 million is represented in what we call spot business, which is not renewal business. And as Joachim talked about, we actually view this growing spot business as actually very healthy for Synopsys. So we're more than a renewal company with a spot business or non-renewal business allows us to do is again make more connection with the customer, sell to more projects, get to know more people and be important to our customers.
And then a point that Aart made or John made about our time based or readable license model, it really allows for the right behaviors to take place in the field. That it's good for us a sales organization to have a sense of urgency to close business but you want that closing of the business to be for the right reasons, the customer reasons or efficiency reasons and not for the wrong reasons having to do with what Synopsys is going to do with that revenue. And then finally, I guess it's a business school credo in that you are what measure, we have a slight variant of that and that we achieved what we measure. Synapses, as you probably have picked up by now, is a metrics-driven company, or a metrics-oriented company and the sales organization is no different. We carefully measure run rate by customer, by territory, by product, by technology. So we have a lot of analytics, a lot of metrics to really understand. Which -- so what this does then is drive the communication within the company. It drives things like reviewing proposals and contracts for customers to understand what the business level change is for each contract. It also allows us to set up strategic plans to succeed at our customers using these metrics. And then finally, we do use run rates or business growth as one element in our sales compensation portfolio. So with that, I am going to turn it over to my partner in the field, Deirdre Hanford.
Thank you, John. Good morning. I am your 7th inning stretch. We're going to get a chance to relax a little bit but don't be fooled. I represent as says Joe, the pressure. The pressure of bridging our technology into customer's hands and getting them through to tape out. It's an exciting, thrilling but challenging job that we have in the field organization. You heard Aart talk about this asset that we have that is the field and I want to go a little bit deeper today and share with you some -- what makes that very rich in differentiating for Synopsys [indiscernible] for our customers. So I want to touch on 4 areas. We address our customer’s key challenges or we might as well go home. This is essential to our success, but we need to do that in a global and differentiated way. We need to operate in the ecosystem but at the end of the day, we must compete or we're again nowhere. This in turn helps our customer succeed in the market delivering great products. So what I'd like to do is traverse through this and share with you a number of different experiences. Starting by ways that we address our customers' very, very toughest challenges. Now we conducted a survey, actually John Chilton's organization had an outside firm conduct a blind survey, asking our customers a number of different questions. For instance, what are your toughest design implementation challenges? Seen here are posted answers, timing closure, power management, taping out on schedule in the implementation domain [indiscernible].
So let's dig into one of those and here's a lovely e-mail. We keep these in the field when we get nice e-mails, we keep them and share them in our all-hands meeting, and here's a customer who's thrilled that they got their chip taped out not only on time but early, and why is that important? Because that customer then can get to market that much faster and start to add value in their product areas. Now another area that is important is making sure that we grow our business. So we need to solve customer problems and sometimes that's at the expense of one of our competitors. So this is the great example of a customer that was in entrenched with one of our competitors and interestingly, they were not -- the customer was not able to get their chips taped out on time and the support was pretty inconsistent. Their chips were large and complex so this creates an opportunity and usually there's something that creates that opportunity, in this case, there were 2. One, there was a central group conducting essentially a bake-off looking at alternative because they were feeling some pain with their current income. And then the other was a group that had been recently acquired. They were satisfied Synopsys users and were looking over at this new group saying, how are you getting your chips out on time? How can this be? And so they were able to say, well, we're not using what you're using. And those 2 sparks then creates [indiscernible] to go in, compete for the business and we have since displaced the competition and as you can see, we have a number of [indiscernible] underway not just in one site but across the enterprise. So this is an example of the sequence we go through to deliver value and turn that into business results.
Now verification is a very complex domain. You heard earlier from several of my colleagues that this is a very time-consuming space. So the host of headaches for challenges are actually much deeper. And I want to share with you are a couple of examples. Now for those of you that are -- have a cell phone on, or are looking at your cell phone right now, I will tell you that there are -- you want us to be helping customers design very, very low power devices. And what that means is when you're looking up a contact, you're not stressing that phone as much as when you're streaming sports. So we want to help our customers be very nimble in the way they design low-power chips. Maybe there's a portion of the chip that you want to turn off when it's not being utilized. So that sounds straightforward, right? Well, as it turns out that's a very, very complex process and the thing with low-power is, it's not only helping them design it but verify it. And this is the case where our customers said, when I'm doing a [indiscernible] verification, I want to make sure that I'm getting that correctly, that I'll actually be able to build what I intend. Now I need to set up the next case for a minute and Brett is going to help me because Brett's from Boston and he drives rotaries every single day. Now when I approach a rotary, I go through a moment of terror because remember it's getting on rotary or [indiscernible] rotary and Brett will tell you that, that depends on what state you're in. In Massachusetts, which who has the right of way, when you're on it, okay. But in other places, that's not the case. So that's why -- hence, my moment of terror. Well, verification is somewhat like that because there's a protocol. Trying a design for the protocol and then verify the protocol. Well, in England, my colleague from England back here by the door will tell you that there's a rotary of rotaries in a town called Swindon. Okay, so this is amazing because there's an inner rotary and there's 5 little outer rotaries, and then there's a -- outside, you can go clockwise, inside, you can go counterclockwise. So imagine if you're the traffic engineer trying to describe that protocol and then stimulate it to make sure you don't create like an [indiscernible]. Apparently, young people like to hit this place [indiscernible] after they hit the pub. So you can imagine now that this is a pretty complex rotary and imagine how you would verify that, describe it and then simulate it.
Well, our customers in verification have much, much, much more complex jobs to do. They have to describe very complex protocols, adhere to the standards and make sure that the chip is going to work. Much, much more complex than this magical roundabout in Swindon. So we had a customer in Japan actually that was adopting a new protocol, and what they realized is that their verification environment, even though they were using DCS, their verification environment was getting fairly complex and they knew -- they kind of redesign the way they created scenarios and added protocols. So we went in and helped them. We helped them clean up their verification environment so they could easily add scenarios and we also -- actually, since they were installing a new protocol in their design, put in place our verification IP. So simulation remained constant. We put in a much more extensible design environment and actually our verification IP helped them find 6 bugs within a week. So this again underscores the value of what we bring to the table. Not only great technology but know how to help our customers be more nimble in the marketplace and navigate things like this magic roundabout. By the way, they have assigned us as the professionals go on the inner section and then the tourists take the outer sections. So check it out next time you're in Swindon.
All right, the last here that I touched on earlier in this section is low-power design. Now this sounds pretty straightforward as I mentioned but when you move into low-power design, it's also a brand-new domain. This is similar to going into 20-nanometer or anything new that incurs massive risks for the VP of Engineering and incurs risks for the design engineers as well. So I'll take you through a sequence with a customer that was upgrading their designs for the low-power and essentially, the first chip was pretty darn rocky. They went in aggressively, started to employ all kinds of low-power techniques, then they found that they were having a difficult time. So we supported this customer. I heard from this design manager about 3x a week for several months during the high point of the project. But, we -- together we got them through with this implementation but essentially what we did afterwards [indiscernible] said, we want to go mainstream with low-power, let's put in place a coherent design flow, let's adopt standard practices that we can take broadly through the organization and we worked together on this. The next chip was pretty smooth as were subsequent chips, and every 6 months or so, we put new releases of our software out. So we're constantly helping our customers upgrade [indiscernible]. And then finally today, we've had about 5 chips taped out in low-power across this 2.5 year period and there's 22 chips in flight today including these design flows and practices. So you take something fairly risky, a new design technique like low-power design and over the course into collaboration, we helped this customer be successful. So again, if we're not solving our customer’s toughest challenges, we're really nowhere but we do need to do that on a global basis. Our customers have global design [indiscernible] even in these small territory accounts that Joe talked about. Typically, we'll have [indiscernible] in the United States and a design center perhaps some place in India or China. So global design teams require that we have global scale. Our customers are solving complex design solutions so we need to bring deep design expertise to the table. They had pretty lean infrastructure groups themselves. So we need to bring flows and collaboration also to help make them whole with their knowledge and then it's pretty usual [indiscernible] either it's been acquired, has spun off a division, has reorganized. So it's also very important that we partner with Joe's group and are very nimble. So this is really a requirement for success in the field and one that we absolutely bring to the table for our customers.
Now interestingly in our survey, we asked our customers, what are the [indiscernible] that influence your vendor selection? And surprisingly, even above licensing and business terms, the quality of support and service. This is across the board, the #1 attribute when considering who you're going to work with in this very high-profile, high-risk, high-stake business of technology. So interestingly, among the management teams and the design teams that were surveyed, Synopsys came out on top in this domain. We do that with a breadth of things like training and hotlines and so on but we also do it with a high level of engagement. We bring the whole company to bear when we engage with our customers. This is the classic case of a broad scale customer that's operating on 3 continents and 12 sites, multiple divisions, complex design styles and so on. So we engage with them broadly across and in a technical and business level. We have broad deployment of our technology and we have active elaboration across the [indiscernible]. So again, I want to give you a feeling for the breadth of the technology that we have but the way that we can partner with our customers to help them to [indiscernible] great success.
This is a letter that -- a snippet from a letter that Aart actually received from one of our customers. And the customer was celebrating the success of their recent design that has been finished [indiscernible]. It's really the collaboration with us that helped him achieve that success. And then he was really reflecting on the fact that is it only through that level of collaboration that people can achieve these great heights in design today. And that's why I think it's something we feel most proud of for our quarter-century of serving our customers. But tough challenges, delivering differentiated support is important but we operate truly in an ecosystem and we have an amazing user community. For over 20 years Aart has been assembling our users regionally in user group meetings. In fact, Aart will be kicking off our new England user group meeting tomorrow. So be careful of the rotaries and remember the protocol. But if you look at this, we've actually 9,000 users across all of our customer bases assembled this year. Silicon Valley and Bangalore, India being are very largest user group meetings but here's a picture of our user group meeting actually in Israel, overlooking the beautiful Mediterranean Sea. And so you have an amazing collection of design engineers coming together to share their experiences with Synopsys. This year alone, we've published 250 papers, highlighting special uses of our tools in different application domains.
Here's some papers that were written, one in Silicon Valley and one in Taiwan, talking about how you can design an ARM-based system to get a really fast slick cellphone out the door. And it's a representation of designers sharing their knowledge, but partnering with important partners like ARM in the ecosystem to get the job done.
Finally, competition. We do this in 4 ways, and let me highlight that quickly. Defending. As you heard from John, we have a #1 share in many of our subsegments, so people are constantly looking to pick off our business. And it's critical to have our reach to be able to watch if somebody is sneaking into one of our customers' Singapore design centers and trying to get them interested in their technology versus ours. What we find often is that our customers may not be using the best and latest features in our tools. So this is a case where we're defending by just making sure they're upgraded with the latest release and using the latest features. So defending is a key component of what we do. But sometimes we're differentiating our technology and then displacing our competitor. And this is a great example where somebody said, I'm happy with my simulator but if you can give me certainly a faster simulator, I'll move. But it can't just be a little bit faster, it needs at least 2x and if I'm going to get rid of emulation, it needs to be at least 4x. And so we did just that. We demonstrated at a much faster speed and simulation, but the customer said, don't just show me some fancy benchmark where you may be much with the results. You need to show that kind of speed up in my design environment. And this par for the course for what we do every day. We realize [indiscernible] in our customer environment therefore, showing the value. So we demonstrated great support, we plugged in the simulator into their environment and now they're off and running with DCS for simulation and they're actually also not using emulation on this particular project. Now once you've displaced the competition by differentiating our solutions, then we go into broad deployment and this now often a case where we're doing a pilot design, maybe going into a division and then going broad scale. So once again, there's a huge amount of competitiveness required, doing change management for our customer, operating on a global basis but making sure ultimately that our customer can, in turn, compete in their market segment.
So hopefully in a few minutes, I will have been able to give you a really good feel for what it's like in the field. There's a lot of pressure but it's really the pressure of our customers that we feel. It's the late night and all night-ers that will put [indiscernible] our customers are doing the same. It's the engineers that are on beepers all weekend because they know that their customers are relying on them if they hit a hitch. We deliver differentiated support on a global basis. This is our bread and butter and it's something that we feel very proud of because often, we'll know more about a particular project that spanning Europe, North America and India, that perhaps even our customer knows. We partner actively in the ecosystem not just with partners like ARM and our users, but also with semiconductor vendors like UMC, TSMC, Samsung, Global Foundries, et cetera. And then finally, we have to be competitive. We need to make sure that we're helping our customers compete but we're also the leader constantly looking to expand our space and to pick off the space of our competition. Because ultimately, we want to make sure that we're helping our customers get better, sooner or cheaper designs out to market. So thank you very much and at this point, I will introduce our CFO, Brian Beattie.
Great. Well, thanks, Deirdre. It's great to see all of you here today. Again, looking at all of our shareholders, our sell side analysts team, our bankers, and it's great to have you all here. Thanks very much for coming. Our job right now is to give you the wrap-up here prior to our Q&A session and we'll do that letting you see our financial performance, highlighting a little bit more view on our perspectives for 2011, which is almost complete, as well as some preliminary discussion of 2012. So to start off, the business is focused on the strength, what we've got today and its focused on growth. And that's about growth in terms of the top line, growing the profitability of the company, growing the cash flow of the company over time and continue to deliver shareholder returns. So we're going to give you a perspective of how we do that, how we've done it so well over the past few years, focusing on the amount of stability and predictability, the model has and why we have that covered. So let's jump in.
Well, to start off with, looking at our total revenue, you can see that it shows [indiscernible] 6 years here, how we've grown from FY '05 to our outlook for '11 of $15.34 is the midpoint of the guidance that we've given. So you can see that we've both grown the company, we've grown it very predictably overtime. If you look back over the numbers, we have now generated 28 consecutive quarters of coming in line with what the company's guidance is, both on the top line and the bottom line and I'll show you how we get that level of predictability while we continue to drive for top line growth and bottom line growth over that same period. So let's jump into some of the details of that as we've got lined up.
Okay, so one of the things that we're doing as we look at FY '11 at this point in time, we're also reconfirming the guidance that we gave. We still have just a little bit less than 1/2 of a quarter to complete our year for FY '11 so at this point in time, midway through the quarter, we're reconfirming the guidance that we provided earlier for the year and that's at the $15.31 to $15.37, which as you see, represents an 11% top line growth. And then EPS, hitting a record of $1.80 for us, which is a rate in between a very tight range of $1.79 to $1.81, which is 12% to 13% growth. Also we've reconfirmed again our expectation to be over $400 million of operating cash inflow this year, which is also very strong. We're coming off a very strong Q3, where we again had great levels of business performance. The business activity was very strong from the sales team and gave us that confidence to bring out very, very strong results for the year.
So much anticipated is the discussion of FY '12. So I'm sure you're all paying attention on our slide that we put up earlier and what you can see is that revenues, in terms of our first-call estimates, are about just over $1.6 billion of revenue and also $1.96 of earnings per share growth. So our first point is that we're saying all of that is in the ballpark with where our expectations are at this point in time for FY '12. We still have, as we said, half a quarter to complete the final profile of the orders, how that comes through with FY '12 and, of course, do a lot of detailed work on cash flow estimates and we'll give you all of the details of that as we typically do at our year end. The earnings call is scheduled for the end of November at this point, and we'll give you a lot more details but we didn't want to let you know this morning that both the ranges that are built in the first call at this point are in the ballpark of where we anticipate giving you a final detailed guidance at the earnings call at the end of November, okay. And you can see where the ranges are coming in. So we think again that's good news that you can take back from this conference.
So then also about stability and growth, let's then look at why do you have this level of predictability in terms of your results. We start off with $2.4 billion of firm noncancelable backlog. All of the orders that we've taken in, that goes into the backlog number and then as revenues come out, you reduce it by that. So we had $2.4 billion at the end of FY '10 and then I'm also going to share with you, I'm not sure if you've seen this before too but it's really the split of how we do this. So Joe's team provides a level of detail so we can track revenues by customers as he said but also by quarter. We also can track cash flows, when is the cash supposed to come in from the various customers, and so we can do that over the 3-year period, which is roughly the length of the contracts that we sign each quarter. So 80% of year one, you've heard that. You've heard 90% of the first quarter. We also have about 50% of the next year's estimates. So as we set targets for year 2, we typically will have at this point, 50% of that revenue for the second year from today, and even year 3, that last year of the contracts we've got, of course, anticipating revenue growth, that's how much we would have at any point in time.
So again this is a very good level of predictability as we put together our own 3-year business plans of how we're going to drive that. So again, very solid, very strong, nicely higher than the annual revenue piece and it continues to go through. Over on cash, this is also an important part. You've seen how we've been able to grow our cash balance, over $500 million to the billion dollars that we've got. So it's also very safely invested. Unfortunately, in today's environment, not a whole lot of return on the cash balance but again a commitment we've made is a significant investment of putting the balance sheet to work for us, recognizing both low returns and opportunities we believe are in front of us for much higher returns. We're going to talk more about where the cash is and also how we're getting our cash flow lined up as we go through. So we're in good shape in that.
Now also, just show the level of diversity that we've got, both technologically as well as geographically. These are the trailing 12-month revenue growth projections that we've got. As you saw, we have about 65% of our business in the core EDA business, IP & Systems from the Joachim is 20% and growing very nicely. Manufacturing is about 12% and then it's about 3% of services and these are the effective growth rates we're seeing in the business over the trailing 12 months. Of course, IP & Systems is a function of the level of M&A activity that we did last year that we're seeing in the trailing 12-month performance and more of an ongoing basis. You heard Joachim describe double-digit revenue growth from this on an ongoing basis. Okay, over to the geographies. You saw that we're half and half. Half of the revenues come from the U.S., the other half were outside of the U.S. and you can see the relative growth rates from each of those key markets with Asia-Pacific being the strongest at 27.5% growth on a year-over-year basis. So again, a very vibrant market opportunity for us as well as each of the markets continuing to grow at that high-single digits, low double-digit levels. And you see the split of that over the last 2 years as well. So nice diversification of the portfolio.
Let's look at the cash flow itself. So you can see the profile going back again over the 6 years, very strong performance this year. Similar to -- back to '07 where we had a couple of payments from one of our largest customers as we noted right down here, gave us that ability and this here again taking the cash flow estimates up in Q3, up over $400 million. So doing very well. The other we highlighted and cash flow can be a little lumpy as you all know between the collections from your customers, when they come in, their schedule, very high quality customers. So payments are very solid and lined up as are the accounts and then we just looked at the average. So we don't have a cash flow estimate for '12 but giving you this perspective on -- look at the 3 years, the number is very strong. Number one, it's growing as we've seen, but again anticipating that it ranges about $300 million, $327 million for the past year. Over time, how do you forecast cash flow? Well, over time, it tracks EBITDA less our cash taxes paid. So that's another way that you can look at it if you're just looking at the traditional P&L perspectives. So identifying that it's a -- can be lumpy from year-to-year, very strong generation and also that we'd anticipate this growing up over the next few years as EBITDA continues to grow. So again, we'll give you more details of the cash flow at the yearend call.
All right, so let's shift over then to how we're continuing to drive shareholder value in this piece and it's really important. Now you've seen the 5 tenants of our goal of high-single-digit EPS growth. And Aart and John and the team here described both the growth in the traditional business of core EDA, as well as the IP & Systems, double-digit growth. So let me talk to the next 3 of how we're going to do that. And the first part we're going to mention is about the spending allocation. So what you do with spending is again you want to allocate those spending dollars to the highest return areas and certainly we've seen that shift, both through our M&A strategy, primarily focusing on systems and IP, while maintaining the operating margin across-the-board in the other segments as well. The other thing we note is a pretty significant growth in terms of the absolute dollars of operating margin that come through. That's up about 38% over that period and then you'll say, well, where are you allocating your dollars? Well, the company is focused on very strong R&D spending and the field expense, which is in the COGS, more of the development that we do, helping the customers get the products working.
Over that period though, you'll see a very significant intentional reduction here in terms of sales and marketing expense going from 27% down to 23%, again margins have gone up over that period, and also on the G&A side, 7% have dropped down to 6% of our revenue. So as the top line is growing, we're trying to not grow the other businesses at the same rate of growth. So that's giving us the opportunity for continued margin expansion. Okay, we've got COGS, which is cost of goods sold reflecting on a little bit of hardware business but mostly some of the work that's done in the field by the R&D teams, good growth and again, we're putting the dollars in the right categories.
So let's then move into -- we got a lot of questions about capital allocation strategy. So I'll give you the considerations of how we look at how we use our cash, how we structure the balance sheet and how we continue to put that to work for growth. So the first part is about the resource allocation to the right segments, which also relates, of course, to not just your expense base but also your capital base, how we can then focus on M&A. I'll give you some details about that, and also, of course, to keep the share count flat and work on reducing it.
So this is a view over that same 6-year period of the company's cash position at the beginning, $0.6 billion as you remember and now being over the $1 billion mark over the 6 years. During that period, $2.2 billion of cash inflow has been generated by the company. So very good yielding cash business that we've got. Where did that go? $300 million went in to capital spending. Again this is only related to our computing environment, the infrastructure to run massive simulations with our tools just as our customers do is compute intensive but to put it in perspective, it's $300 million over the 6 year. From the M&A side, $1 billion has gone in, and I'll also you where that's gone over the past few years, but as you can see, a significant commitment to continue to put the balance sheet to work, to grow the top line of the business and improve the profitability during that period. Over on the net share repurchases, that's $400 million. We just note in your notes that on a gross level, the company spent $1.2 billion buying back the company stock but the employees, of course, this is not pre-stock, the exercise price of that represents about $800 million but it's also coming to the company. So $400 million is gone, out of pocket for the company to buyback the stock position. And I'll talk more about that buyback as we go-forward. So people say, where do you use the cash? This is the number. M&A, the biggest piece of it, the net share repurchase is the next area and it's some of the capital that addresses the whole free cash flow for the company.
So let's talk about that M&A perspective. These are the -- where we spent $500 million over the last 2 years, most of it was in FY '10. We want to be very disciplined buyers. We recognize our cost of capital is about 10%. So all of our returns have to be better than that. We got to focus on NPV, we got to focus on acquisitions we put in place, continue to grow top line, grow market share and grow profitability. So the largest investments of that $500 million have been in IP as both [indiscernible] here, Virage and nSys, we disclosed. Also in terms of systems, CoWare, Synfora and VaST in the past 2-year time frame. We have some investments in the quarter, which really finds some better algorithms of doing things, better, faster, cheaper and where we can embed that into our core technology and just provide an overall product that's better and then John identified the ORA, the optical investment we made addressing the LED business. So that part continues to grow.
So let's just talk a minute about the balance sheet itself and where it's invested. So good level of cash, you know that it's grown over time, but again as you look at the sources of where the cash has been deployed, you can see the U.S. column of cash here has declined over that same period. When you look at M&A, most of the M&A comes out of U.S. dollars you look at that significant amount of the buyback activity to drive the share count down so much and that is only and exclusively using U.S. cash. When we look at how cash is generated, when you look at over $400 million this year, it's split 50-50. So $200 million comes into the U.S., about $200 million of that is from offshore, which our net effect of repatriation cost is about 20% to the P&L. So that's the, again, the penalty of repatriating -- pay tax off-shore and then paying another 20% to repatriate. So at this point, you can see a significant amount of cash, so what do you do with that? You got to look at how you can look at acquisitions off-shore and also continue to work with Congress on repatriation as every other tech company is doing. So that's where it's invested and then just to clarify the point, it's in the notes as well as. We have a $300 million line of credit, that also holds $150 million -- well, holds a $300 million requirement for maintaining cash level, so it's $1 billion and what we typically like to do is keep $150 million of that in the U.S. and also keep about $150 million as a cushion on our offshore market. So that's what we like to hold in terms of roughly $300 million of that in reserve, okay? So again, in terms of debt, we do have a line of credit, rates are very good and we would consider putting that at the right space for the right M&A opportunity that comes up. So again, we're very strongly positioned, have a great balance sheet relative to our industry and continue to have the opportunity, lot of dry powder.
Okay, one of the areas around our focus on reducing the share count, and we've reduced it nicely over the past year. We were up to about 152 million shares last year, we have 147 million in place as of Q3. So a nice focus on getting the share count down and then some people say, gee, Brian, you spent so much on that buyback over that period of time, where is the share count, it's only down by several million. One of the things that continue -- we present on our financial statements, of course, is the amount of stock options overhang. So part of the plan is we've been reducing from 25% overhang to the basic diluted -- basic share count. You've got 25% more stock options outstanding if you go back to '05. We've cut that in half. So much of the share count activity of buying back is also gone to reduce the share count down in half. So now, we're at about a 13% level compared to 25% back in '05. So that's where the share count has gone. Again, nicely the shares for the most part are in the money and they're being exercised and as they're being exercised, we buy back more. So again, relative to grants, we buy back more, we have decreased the number of stock option grants that we make. We have converted, again, several years ago reduced the life from 10 years to some of the use were granted back in 2000 even and we reduced that down to 7 years. Our RSU transition is basically a roll of 3 stock options for one RSU and then again, it's just for – they're given to fewer people and we've reduced the term. So all of it is focused on getting the share count down to drive up that EPS as we indicated earlier.
Okay, so I'll just to give you insight on how we've been able to do that. Now of course, we have to continue to work on repatriation opportunities and be able to use the significant amount of cash that we have offshore. So with that, hopefully we've just about to lead into some Q&A and turn it over to Aart for questions and you can see a perspective of how we've got a very strong business model, a lot of growth across the segments we've identified, and again continued focus on increasing shareholder value that we've laid out with an aggressive putting the balance sheet to work approach that we've laid out here this morning again, okay? So with that, turn it over to Aart.
Aart J. de Geus
Thank you, Brian. So I hope that we've been able to give you sort of a snapshot of the company from multiple perspectives from the perspective of the market that we're touching and why it's so dynamic and why therefore there's so much stress on the field to serve the customers, but why it's also exciting to be in the middle of the stress, because that means they count on us and they count on us for technology reasons, for support reasons and for the very fact that we are well financed and therefore, will be around no matter what happens around us and do well for them. Hopefully, we were able to give you a good overview and a complete overview of where we are financially, how we think about things, how we balance an engine that has many global complexities but is actually I think in very, very good health compared to virtually every other company that has in this type of a position, and how we last but not least, are dialing up our sensitivity and commitment to look at the shareholder returns side. We think that the company has done actually -- has executed very, very well under a broad set of let's say, variations in the market in the last few years and therefore, we should now drive towards seeing some hard returns on that. With that, we like to open it up to Q&A. We'll asked the speakers to be here.
Lisa has a couple of technical do's and don'ts.
I do. If I can make 2 requests. One, if you could please introduce yourself and give your full name and secondly, if you could please start with a single question, Jay, and then we'll dial back around for follow-up questions after that.
Aart J. de Geus
Okay, well, Jay, you're fast on the draw.
You guys are going to let Jay ask the first question?
Unknown Analyst -
[indiscernible] Howard Securities. So I'd like to ask 2 related questions about the industry. In your slide, you talked a lot about the market drivers and for those of us who've been following the industry for some length of time, those are the culmination of what we've seen evolve and drive for the last 30 years or so and yet, core EDA is a low-single-digit growth. The question is, how do you -- the systemic or structural limitations at the customer side, which is to say relatively small, not quite fixed number of users, the propensity of customers who share their EDA spending, and so again the question is how do you possibly overcome that and relatedly, in one of your slides, you talked about the extent of your IC design tools business in absolute terms being significantly larger than for Cadence or for Mentor but as a proportion of your business, it's larger than for your peers. So you're more highly correlated for IC. The question is how do you reduce that tendency? And what do you think the TAM opportunity is for [indiscernible].
Aart J. de Geus
Let me summarize it as how do you grow, I think, is bottom line of your question. And fundamentally, we have, at the minimum, follow 2 strategies. One is to say, "Hey, not only do we want to lead in terms of the products that are in that area, but also increasingly respond to the fact that design is so complex, that having just individual tools will not cut it over time." And therefore, the correlation, the interaction of the tools are actually very important. And by the way, that includes the human support to make the customer successful. The economic version of that is to say, "How do you gradually grow share in a market that is subject to whatever the vagrancies are in the semiconductor market?" The second answer, as we need to say, to recognize that until there happens much more consolidation, that's the way it is until we'll grow share, that to recognize that there are great opportunities in the surrounding. And the first surrounding that I think we can now finally claim success in is to say we were able to grow an IP business from scratch that has good growth opportunities, that does not sell to the same buyers that we have built business out and sell directly to the project to the engineering managers, that has a natural destiny to move towards significantly more outsourcing for both economic terms, but also technical terms. It's actually difficult to do these blocks really well, and in which in our segments, we are the #1 leader. Therefore, we can aspire to both growth and good profitability. And the reason we can say, "Yes, we were able to do that," is because it's past the 20% point of our business. Therefore, it is material and it's doing well. We will continue that line of thinking, saying, "Well, what can we do more in the systems, what can we do in the multi-physics surroundings?" So in other words, we are diversifying the portfolio while not ignoring for a second that the core position is a profit engine and is an engine that gives us better opportunities in these adjacencies. So in that sense, I think we have responded to the characteristics of the markets that are given around us, and so far, are executing, I think, reasonably well. Do you want to add anything to it?
Joseph W. Logan
Just to clarify something, Jay, we intentionally, we want to be focused on the areas that help our customers. We'd like that. We like developing those relationships. If you talk about larger proportion of our business is on the design side than, say, our competitors, the majority of the nondesign side business in EDA that we don't currently participate in is in PCB. And that's not exactly a wonderful growth segment there. It's essentially kind of a small business that's very mature, that has a disadvantage that it sells to someone else. So you're not able to bring that $1.5 billion of power to bear on helping someone design their very critical chip.
Saket Kalia - JP Morgan Chase & Co, Research Division
Saket Kalia from JPMorgan. So one question, we've seen some mixed news in the semiconductor industry this quarter. So I was wondering, just very high level, what could you share about how you're customers are viewing semi R&D going into 2012?
Aart J. de Geus
Sure. Well, the first thing is, of course, they read the same newspapers, we all do. And everybody is very cognizant about the overall macroeconomics just being fairly unstable. And so the first thing that people do is to try to probe their own value chain as well as possible, and make sure that they're not sitting on inordinate amounts of inventory. And so I think the good news part of that is as much as the customers are looking with caution forward, they're also not particularly worried about sitting on inventory that immediately is going to create a big supply-demand imbalance. And that's, I think, why we see a relative degree of stability. The other thing which is interesting is they, too, realize that there are a number of markets that are very promising and tend to be a winner-take-all. And if at all -- and maybe the field people can comment. If at all, it feels like an intensifying of pressure to run even faster, maybe also partially predicated by the fact that there are advances in silicon technology that were, at least by some, unexpected this year and are moving faster. But I think the other thing is that they understand that a number of the mobile-related products have RAMs and differentiations that are very strongly attached to the execution of technology. And so we have not seen any downturn whatsoever from that pressure.
Yes, I toured through Asia recently, in the early summer, and every customer I visited was looking to get a certain speed profile out of a processor core in order to hit hard the mobile market. And so you see a lot of folks saying, "I want to be the first, I want to do my winner-take-all strategy." And they're doing that to serve an Asia market that has different sets of characteristics. So I think we're seeing in soft times and in aggressive times, people are looking to design their way through that market time.
Aart J. de Geus
Joe, any comments on the...
Joseph W. Logan
I believe we may have conversations with customers about what's going on with Greece, and they have that as much knowledge as I do about what's going to go on with Greece. But our customers are intensely focused on design. They're intensely focused on their road maps. And if I look at the buying behavior this year, where the first 3 quarters were actually stronger and more stable than the last couple of years, and that's continuing through –- we’re now 2/3 of the way through Q4. So I put it that our customers are just intensely focused on their road maps and design and they need tools and IP to do that.
Paul B. Thomas - BofA Merrill Lynch, Research Division
Paul Thomas from BofA Merrill Lynch. If you look back in '08, '09 downturn, you guys had already completed your model transition year, highly ratable method. And now your main competitor is probably halfway through or in the process. One of the reasons you have done been in the past was to help pricing a little bit in terms of trying to book the deals and when you had to do them. When your competitor gets all the way through their transition, do you think that, secularly, there will be at least a little bit less pressure on ASPs in terms of the timing of when the deals get done? Or do you think there won't be any change from the environment you're in right now?
Aart J. de Geus
Maybe Joe can comment on the timing. I think, in general, there's no question that there will always the pressure on ASP, but it's often the reverse, which is it's pressure on putting more value for the same ASP because we are driving more as well. So that's just a given of the field that we are in. I think as companies move to more stable business models, they tend to reflect on how to improve those models because they can. Whereas when you're in these, "Oh, you have to meet the end of the quarter," and actually the customer says, "Well, it's 5:00, Friday, I want to go home," you know he's lying. Well, I shouldn't say it like that. But the fact that -- erase that from the record. The fact is the pressure is not in a healthy situation for having a really good working relationship only on Monday. But the fact is it's negotiation. And so I think both on the customer side and our side, there's a much bigger perception that we actually are in multiyear relationships that are absolutely vital to both parties. And I think as other EDA companies get into more of such a situation, I think, that can only be positive for us.
Joseph W. Logan
Sir, so the comment about timing. As you probably know, we've had this model for 7, 8 years now. And it's been a transformation in the relationship with our customers, in that we sort of have an agreement that Synopsys can't use this model to leverage them because it's a small industry and leveraging anybody doesn't work. It comes back to haunt you. And on the other side, customers understand that we're not really trying to or not at all trying to drive individual agreements to hit individual quarters. So what's that's meant from a timing perspective is if you look at the orders' linearity over the last, say, 7 years, we're often closing as much business in the first month of the quarter as we are in the last month. So we spread out our businesses if the engagements feel much more natural from a timing perspective. And I hope that, that's the case with our competitors as well.
Unknown Analyst -
Brian, just 2 quick half questions. Did you say that ops margin would grow in '12? Any sort of range on that? And then the second question is in your slide on cash flow, you talked about in '07, a big customer payment. Was there anything onetime with that customer or anything unique with that '07 customer payment that maybe when that customer renews, it wouldn't happen again in terms of the cash collection characteristic of that '07 signing?
Yes, sure. Yes, there is a commitment on margin expansion. We wanted you –- when you’re here today to know that we look ahead to obviously '12, '13 and '14 in our planning models. And that we felt and give you guidance around a point that where we are to-date, that looks okay. It's in the ballpark for us, we're going to work through that. As part of that assessment, it's also to say, and margins should expand next year as well. So that's built into the plan. When you looked at this year, there was a bit of a decline in operating margin, but then as Aart explained, so much of that was just related to the acquisitions that we do. So you buy a company, you don't get all the revenue. Because of purchase accounting, that deferred revenue goes away. The companies typically don't have the level of operating margin that we do. So we need to go in and look at the G&A spending, look at the sales and marketing spending, fine-tune the R&D piece and get that operating margin back up to a level that we have. So a year of significant acquisition like we saw this year, while we still got top and bottom line growth, the margin dipped for this year. And again, we committed to improving that again for next year, and we'll give you more details of that at the end of November when it's our year-end call. So that takes care of the short discussion on '12. Back to cash flow for '07. A significant customer contract was renegotiated, it was a onetime event of when that particular contract was renewed relative to payments that we owed at that point. Typically, in terms of cash flow, we have cash in advance. So there will be quarterly cash in advance or annual and advanced payments from our customers. So there is an element, as we look at this year, of significant amount of increased business activity for us. The order side came in very nicely through Q3 and for the full year. And as we explained, when we see over $400 million of cash this year, it can be lumpy just based on when those contracts are renewed and customers pay in advance. So specific to '07, there was a onetime element related to that. As we look forward through this year and into next year, this year is very strong, very strong business level of activity, very strong collections. The profile this year, relative to our cash, was much further in advance than what we had anticipated just because the order profile improved from what we had originally anticipated. So we got more, we got it faster and the quality was much better than what we had earlier anticipated. So it's good.
And for those of you who aren't really knowledgeable on the company, in '07, we had 2 payments from the same large customer. And that's what caused that spike.
Raj Seth - Cowen and Company, LLC, Research Division
Raj Seth from Cowen. Just a question on the context of the earlier discussion on operating margins around R&D. And Aart, you're doing about 30% R&D spend, in 2007 it was 29%. There hasn't been really any leverage on that line, as I believe there has been on G&A line and a couple of the others. In this business, is there any opportunity to see R&D as a percentage of revenues decline over time as you grow going forward? Or is there something about your own R&D intensity and the breadth of your business to think that's difficult?
Aart J. de Geus
It's a complex question because obviously, if you can grow share, if you can grow your business, you leverage the R&D better. And that's why being able to have a strong position in any subfield is important. Otherwise, you just keep investing, do not get the revenue for it. And so we do anticipate that we can, over the coming years, gradually grow more share, therefore, there will be more leverage. The reason that makes this question difficult is because simultaneously, as to an earlier question from Jay, we said, "Hey, we also are trying to expand our TAM." And so there's 2 ways of doing that. One is to say, "Okay, let's invest organically in new areas." And the other is we'll work it through M&A. Well, if you do it organically, then obviously it will pass the R&D end. It's typically a good return, but it's a long return. If you go through M&A, it taxes the P&L for the reasons I gave before, the haircuts, the ops margin of the incoming company, the moving to ratable model. But you are often immediately in a market position that gives you customer feedback and revenue. And so just to reiterate what -- attached to the ops margin question, the answer that Brian gave, is our plan right now is to improve the ops margin. If there's big M&A -- and that's an if, there's no yes or no. It's just when it happens, you need to realize that, that has impact. But it doesn't take away from our desire to step-by-step move the ops margin back to the mid-20s because we think that, as a company, we should do that and can do that. Our main businesses are clearly already there.
Thomas Diffely - D.A. Davidson & Co., Research Division
Tom Diffely of D.A. Davidson. Hoping on talking about margins, hoping you could give a little more color on the IP systems business and why it's lagging today beyond just the fact there's new acquisitions there? And where do you think that business can ultimately go on a margin basis?
Aart J. de Geus
Well, I'll defer to Joachim to explain a little bit how we build that. It is a different business than the tools business because it tends to be more subsegment-specific. But it has exactly the same characteristics as I just tried to explain to Raj, which is, "Hey, if you can sell more of a given title, absolutely the leverage goes up very, very rapidly." And so Joachim, why don't you give a sense of how do we actually think about that as we choose, which titles to go after?
Yes. So the way that we look at the business is relatively simple, in fact [indiscernible] building blocks.
So we start looking at different areas. First, the particular market segment, like mobile or home entertainment or automotive. Then we look at what's the commonality in terms of functionality between all these subscripts are being made, then we pick the kind of the important things, the ones that we find on many, many chips. Once you've done this, then you have to get into all of these chips, right? That becomes you potential, your TAM for that particular protocol. Let's pick -- I don't want to pick USB all the time. Let's just pick something in the mobile space, like the MIPI standards, for instance, where we have been very active over the last couple of years. This is an initial development of the protocol, so that it works on a particular process, maybe on TSMC. But then there's limits used to put your IP onto chips, which are being manufactured at TSMC. Obviously, TSMC does not manufacture all the chips in the world, right? So you want to be able to get that IP onto every single other process. And that's where most of the leverage actually come from. The initial development is like N, and then getting that IP to another process is about 10% of that N. So the more processes you are able to attack, the more you're able to start generating a return in a profitable business or a business with a higher margin. So our fundamental strategy is pick an area, pick the titles and then immediately strive to work towards becoming the #1 provider across the whole process universe. Then we move on and then we attack the area. And that's working very well. I didn't show you any detail, but in most of these segments, we're actually the #1 provider. And that's the way that you achieve profitability in IP.
Thomas Diffely - D.A. Davidson & Co., Research Division
So longer term, would you expect margins being above corporate average for the IP?
Well, slightly towards getting to the corporate average.
Aart J. de Geus
Yes. And maybe the way to think about it -- and I don't want to be -- hooked too strongly to an equation, being an engineer. One way to look at it is to say look of your profitability and ops margin, look at your growth rate and add those 2 things up. And there's a tradeoff -- or multiply, if you prefer that equation. IP is clearly a case where today, there's more demand than we can fulfill. And so it's clearly a question of, "Hey, why are we not investing more there because we'll grow faster?" Yes, but we also like it to, as Joachim just said, move it towards the corporate average as soon as possible. And so it's a good balancing act. And the job of Joachim is to grow the highest value business within Synopsys. And if push comes to shove, growth is more important than profitability. But the business without really good profitability is not just not a business, certainly not when it's 20% of our overall corporate profile. And so we, on purpose, communicated to you that we are pushing the possibility up because we think we can go. But we do see it as a good grower. And so that's why looking at the sum of the 2 numbers is probably a good indicators for the thinking of anything.
Maybe another way to approach it would be that it is structurally a little bit more resource-intensive because of the porting to the various processes.
It's James Pan from CP&E Partners. Just a couple of questions, I want to follow up on that one. But first, on the acquisitions that we made in the last couple of years, $500 million last year, that's $1 billion over 6 years, what would you guess is our weighted cost of return or weighted return on an after-tax basis, on a leveraged basis? And the second question is if your margins are lower in IP, 20% in your business, and your corporate margins are 25%, that means your core EDA margins must be, what, 35%?
Aart J. de Geus
Well, my point is that when the IP margins go up, won't your overall core margins go up or your overall target margins go up?
Aart J. de Geus
Well, let me go backwards on that. The objective, as much as it's exciting to talk about margin, is the growth rate of the actual profit produced. And so even if you have a business that has a lower margin than the corporation but it's going really well and it's possible, go, go, go, right? I mean, there's no question that we will keep investing in IP even if it's at a lower margin than the core businesses. At the same time, our general direction is still to gradually move the ops margin up because if you look at many, many years, you can see that a company like Synopsys can absolutely trends to the mid-20s because we are in a strong position in that business, we should be able to execute against that. But there are many variables in play at any point in time. The second part, which is, and maybe Brian can elaborate on that, the return is not traceable in the first 2 years of an acquisition. It takes much longer to look at which acquisitions work. Now some tend to move very quickly because you're immediately adjacent. You have a sales channel that's ready to sell, the product is ready. Others, the acquisitions themselves, there is a reason we think they're for sale is because they have all kinds of challenges, but there's an interesting opportunity to move into a field. And it takes a bit of time to get that. That's a long way of saying that M&A itself has many, many variables. And ultimately, one has to be convinced that, A, there's a strategic reason to go in that direction. Secondly, that over time, the return on investment is as good, better or at least similar to what one can achieve internally. But it has a very different timeline because one is in action immediately, but the return is actually more visible more over time.
James, maybe to quantify some of the points there. We have a disciplined process inside the company both for identifying targets, disciplined buying mentality of we know many of the companies in the industry, if not all of them. And at some point, look at evaluations against those, and some are out of reach and some are negotiable and some aren't. We just work it over time, so we don't overpay. Then we look at some of the metrics. We look at an internal rate of return. So I mentioned our cost of capital was 10%. So internally, we say, "Let's aim for a 20% return on our investments," knowing that not all M&A hits the actual expected targets. But given that they're greater than our cost of capital, still a net return to shareholders above that cost, when we look at accretion-dilution, and again, in that regard, we typically look at companies being accretive within the first year. Our most recent acquisitions are on track to be accretive as we set forth for this year. And then we also have a process that we both internally look at for the company, but also present to the Board of Directors every quarter on all of our acquisitions over the past 2 years, relative to the financial performance, revenues, bookings and so on, and when we're winning, how the product is coming, what are the financial metrics that go with it. So there's again a commitment on continuing to grow through M&A, disciplined about how to evaluate, as well as follow up after the process and continue to generate that. So those are the metrics we look at, and we're happy with the returns.
Evan Bloomberg - Sigma Capital Management, LLC
Evan Bloomberg from Sigma Capital. I just had a question about the margins regarding the acquisitions you've done this past year. I'm trying to understand what the impact was for a onetime perspective versus the fact that those businesses were lower margins, the onetime impact should I assume help you next year?
Yes, that's where we said the commitment we saw this year are just in operating margins as expected. And given the companies, and some were public, so you know that the profile of margins that were there and anticipate it. So as that comes in, when it comes over to the company, typical purchase accounting, deferred revenue, look at what's on the balance sheet, we lose 85% to 90% on whatever is in deferred revenues. All the expense comes over at day 1, except for those areas that we're able to reduce, whether we've got some synergies and spending in the G&As and some of the sales teams and marketing. So we look at some of that on the R&D, as I mentioned. So overall, there is a negative impact. And as I said, we are expecting margin expansion in '12 given that we're through that acquisition significant year last year. We're through that and expect that. Now all the discussions about '12 also assumes that there's no further M&A in the '12. I'm just giving you the organic perspective on the company's [indiscernible].
Unknown Analyst -
[indiscernible]. Question on the IP side. What is the focus area for growth on IP side? You have your [indiscernible], you have some presence on memory now. So is [indiscernible] possible you can find [ph]? And also related to that, now you are competing with ARM on the [indiscernible], would you ever consider interrupting [ph] their royalty part about the IP side?
Yes, so in terms of average of focus for IP, our key areas so far historically has been the area of interface, so some interface IP. We still see a tremendous amount of runway to continue growing that business. That story is not over by any sense of imagination. On top of this, we have layered one very exciting area, which is one of embedded memories, which has become a large component of our IP business. And the smaller fraction of our business, which is the area of standard social libraries. And our intention is to continue growing the embedded memories and also the standard social library business. And to the extent that it's going to have an impact on the business model, it might have, because if you're familiar with the business model for embedded memories and for standard social, there's something called foundry loyalties, which are based on wafers and on wafer shipment. At this point in time, that royalty element is not a significant element of our revenue. It may grow a little bit going forward.
Unknown Analyst -
[indiscernible]. Could you talk a bit more on acquisitions, prioritization of what you're looking at and where -- what sort of things?
Aart J. de Geus
Well, most people don't realize that at any point in time, we probably look at about 30 companies nonstop. Secondly, that the looking and the buying actually is often has a very long time difference because the number of windows is actually quite small. Market is up, everything is too expensive, there's no buyer. Market is down, everybody thinks they're undervalued, there's no seller. And there's actually only small windows of transition where it's possible. Secondly, you need to make sure that there's actually an alignment with the opportunity space at that time for the business and our ability to absorb and pay for it. I think in the last few years, especially for companies that are a little larger, we are starting to focus more on the financials of it. Maybe that is because, over the last 2 years, we've actually built fairly complete portfolio. And therefore, at this point in time, if we buy things, there's always small buys, typically technology buys, just to make the portfolio stronger to maintain or grow the possibility of the core engine. Secondly, the most interesting buys are really the ones that are in adjacencies where we immediately can move into action. And so the IP acquisitions, while difficult because many of these companies have all kinds of internal challenges, have been very good for us in broadening our position. And then the intriguing but also fairly open-ended in terms of the search space, are the things on the new adjacencies. And there, we typically look at defining adjacency in terms of these are technical adjacencies. So if you take, for example, the Optical acquisition we did, although it's not exactly same technology, it's 3 dimensional, very sophisticated simulation. We know that kind of stuff. We understand immediately what they're talking about. Secondly, is there are a channel adjacency? To what extent can we use our channel? And you heard Joe say that while we are selling in similar ways, he has a little bit of a different channel. So the question is that a broadening or is that a nonefficiency? Let's be careful with that. And the last thing is, is there a customer adjacency, meaning, can you sell to the same people? And to take the Optical example, what was intriguing is we were already selling to people that did LEDs, that did solar cells, beyond our TCAD business. And so this was a natural complement in an area that we already knew. And so obviously, the further field you go, the more you tax your adjacency criteria but the more you broaden your TAM. The closer you're in, the more efficient you'll be, in fact, in execution, but you're not necessarily broadening your TAM and then you [indiscernible] the question of what's the growth rate. And so that's how we think about it. And so it's really 3 buckets, strengthen the core or broaden the position there. Secondly, go into immediate adjacencies that we see have big opportunity, and third, which is can we broaden the TAM overall.
Unknown Analyst -
This is a geography question, having nothing to do with Greece. You highlighted, non-Japan, Asia-Pacific as a growth market for you and for the industry for the last many years. By my calculation, you have about 40% of that market already, well above your corporate average. The question is could you talk about the investments that you're making to sustain that position or grow that share in that market? And does that geography in any way influence your M&A strategy? Conversely, could you talk about your prognosis for Europe as a market for EDA? And then a follow-up.
Aart J. de Geus
At least you're predictable. These are great questions because first, of course, Asia-Pacific, which by the way, for us is made up primarily of Taiwan, Korea, China, India, Singapore and a few other small things around that, as a region, obviously, have gone through a revolutionary change in the last 2 decades, and especially in the last decade for China. And revolutionary change both in terms of the market size, but also the competence in high-tech. Maybe Deirdre can comment in a minute the difficulty of dealing with that rapid change from a support point of view, and I think why we've done well. But we realized this early on. And both through the literal diversity on our team and therefore, they're feeling very comfortable in that region, but also from the decision-making of early on putting high-quality people in the regions to grow, including being in China already in '95, for example, in Taiwan and Korea before that. We have been able to carve a very strong position. I'm not sure if it's 40% or not, it sounds about right. And the opportunities, of course, grow with those territories. And in that context, maybe the other comment would be from Joachim on specifically IP because with that region also comes a slightly different new design methodology, which is much more IP-based. So we've been able to move very quickly in those areas. And there's no question that the hunger for participating in the global markets is now shifting again. Specifically, China is not going, is starting to focus on its own internal market for many economic reasons you understand well, but also for the very sheer fact that there is a growing middle class that will buy modern products, maybe not in the same price point. And so we can see a very high degree of emphasis of things like tablets at a different price point, maybe with not quite the same feature set but still using very advanced chips. And so in just the last 4 years, having been there many times, it's been amazing to see how quickly the design community has improved the design strength. And by the way, how many international companies are there? From an investment point of view, there's no question that we continue to invest heavily. And in that context, the situation of repatriation or non-repatriation of cash has material impact, I think, on the United States because companies such as ours and many of the high-tech companies have a large amount of cash abroad. Finally, on Europe. Europe for us has been relatively speaking very stable because it's driven by a few key engines such as Germany, France, the Nordic countries and the Benelux and England. And those have been relatively speaking quite stable. You can add to that for us, we put Israel in that bucket, which has been growing quite well. And so from a fundamental technology, they have strong systems companies. There's no question there's a lot of issues right now in Europe. And I think you made the joke that Germany is acquiring Greece without Alex Brown [ph]. But the fact is there are lot of question marks, but we have not seen it's changing the advanced design push.
I remember we just celebrated our 20-year anniversary in Taiwan, having a direct channel and a massive R&D investment there in China and every other company virtually in Asia. But about 10 years ago, I visited a small company in Taiwan that was small then but now is well over $1 billion. They issued a challenge to me and said that, "We're not -- don't think about us as a second-tier geography, as a second-tier class of design. We're competing against the large U.S. firms, the large European firms. We expect first-tier equipment and technology, first-tier support because we're competing on a global basis." The gauntlet was thrown and the challenge was issued, and I think we've responded well with that. Similarly in China, it's amazing that many of our customers are doing 20-nanometer investigation. They're active in 28- and 32-nanometer designs. They have quickly accelerated to very advanced node and very advanced deployment of technology that wasn't the case probably even 5 years ago. So the acceleration in that market is extraordinary. And not only do we have sales and support in the regions, we also have a very large R&D investment. So I think we're very well positioned in these regions to respond to customers' demand.
Aart J. de Geus
Just one more comment...
On IP was really interesting what happened because the dynamics was, the adoption dynamics was very different from what we see in the rest of the world. In the rest of the world, most of our customers have internal capability for doing the kind of things we do as IP in this space. And very often, the tradeoff, should we continue doing it internally, should we be outsourcing this? And with all sort of concern in additional external dependency, are those guys trustworthy? Yes, they are. But will they be really able to execute and deliver on time? When we went to China, we stated an explicit strategy for IP for China about 5 years ago. And of course, there were some [indiscernible] which did well. I mean, they're going to do the low-end type of products, in which we just have some special work at the low price, et cetera et cetera. We decided to go the other way and play very much to what they wanted to be and wanted to become, and deployed a lot of people there, not to be able to support customers directly. And what they did is they just absorbed the IP because there was no internal capability available in those companies and their only desire was to get to market quickly in a predictable way with the product. So get-to-market with a chip that does work. And we could pretty much stand for that guarantee for that. The other thing that happened is that at the same time that we're seeing an opportunity of getting the IP from us and not to design those chips, they also liked the idea of having a single point of accountability in terms of the chip really making it, which is what brought the tools into the whole discussion. And we have a number of companies in China buying tools in IP at the same time, not because we were trying to capture this from a contractual point of view, but because they saw us as the guarantee for getting the chips to market quickly. Once we're in that game, of course, we have to deliver, and we did. And that started feeding on itself, success usually feeds on itself. And that's why there is a fantastic opportunity in IP, with very deep relationships also in China these days.
Unknown Analyst -
Last follow-up. At your last analyst meeting 4 years ago, one of the metrics that you highlighted in addition to run rate, which you introduced then as a key metric, was account profitability. And I'm wondering, Joe can jump in on this, is this something that you still measure or can, in fact, use as you highlighted a few years ago?
Joseph W. Logan
Sure. We do still measure it. And I'll talk about how I use it and Deirdre can comment on how she uses it. It does provide -- we do it, we report twice a year internally in the company. And it provides a perspective of where you're investing or perhaps where you're overinvesting. And I think I use it to redirect resources to places that are growing faster, away from places that perhaps are growing less. So for me, it's a very straightforward tool for redirecting resources.
Yes. And when you got 1,000 field employees across 3 continents, you need to make sure that you're very nimble in terms of where you're putting those resources. So we use it as a mechanism. Are we investing consciously? Or are we just in a position where there's inertia and we're there because we've been there? And I think this gives us a good tool and a good mechanism to be very conscious about our investments, where we are from a product standpoint, where we are from a geography standpoint, and that those investments are paying off, because it's strategic to us from an R&D collaboration standpoint or it's strategic to us because this is a growing customer, and we think we can grow with them.
Okay, we have time for one more question. Anybody?
Unknown Analyst -
One follow-up for Brian. So at this time last year, I think you started cash flow at about $220 million because of differences in collections and disbursements from your prior. You've obviously, outperformed that by almost 2x because of stronger orders and collections. And I think you pointed it to a 3-year average several times. Is there any reason to think that in FY '12, the interplay of collections and disbursements will significantly change your historical cash flow profile?
To be specific, the impact of collections is the most significant part. The largest element on our P&L and our balance sheet is that level of cash coming in from customer account levels. As we said, it can be lumpy, though, right, when those scheduled collections are anticipated to come in. If customers say, "This is a great opportunity to renew, I need new licenses, I need to extend," in a lot of instances, we'll say, "Well, let's redo the whole amendment, let's update the whole contract." In that case, we'll look at a new revenue profile and also a new cash flow profile to build on top of what we've already done. So as we're just about to wrap up another 6 weeks of the year, we'll have that final profile. We did highlight that the numbers are a little bit lumpy. And based on when customers are scheduled to renew as we see it today on that quarterly profile by customer of revenues and cash, we build out a very solid plan. And in this key year, we found out more orders came in. The timing of that relative to the whole year of when it would come in was earlier. And we just did really, really well in the cash collections for the year. We'll get all the number lined up. It's one of the hardest elements to forecast really tightly. And so again, as we do that, we'll let you know at the end of November with the year-end call how it's going to come up.
Aart J. de Geus
So with that, I guess, we're wrapping it up. First...
So I now pass it [ph] to Aart.
Aart J. de Geus
First, thank you for your time and your participation. Secondly, I hope that we were able to give you a balanced, complete and hopefully, positive perspective of the company. We certainly feel that the company is in a very strong situation. And after many years of investing and broadening the portfolio and strengthening it, I think we see many positive points. And so if the market is great, we will run with it. If the market is difficult, we will absolutely be the safest place for our customers to work with, and we'll benefit from that. So on the EBIT scenario, I think we are hopefully, a good investment in your eyes. And if nothing else, we are an investment in a free lunch for you following right now.
We're sticking around for lunch.
And yes, for that lunch. Lunch is actually across the hall. If you want to grab lunch, come back, we'll have each of our executives at a different table. So you're free to sit with whomever you'd like. And we'll say goodbye to our webcast audience.