Dr. Alex Shubat is President and CEO of Virage Logic, a company based in Fremont, California. Shubat helped found the company in 1996, and since that time has served as CTO, VP of R&D, and COO before taking over at the helm of the organization.
Prior to Virage, Shubat spent 10 years at Waferscale Integration. Along with a BSEE and MSEE from the University of Toronto, Shubat also has a PhD in EE from Santa Clara University, an executive MBA from Stanford, and holds 20+ patents. Who better then to address a fundamental in Business 101: How to grow the product portfolio.
Q: Tell me about Virage and the technology.
Alex Shubat – Virage Logic is a broadline IP provider serving customers who are SoC manufacturers, or designers, or chip companies.
Q: Who’s your competition?
Alex Shubat – We are a physical IP company, primarily competing with other physical companies, companies like Artisan, which was bought by ARM, or ASPIC, which is no more, or the physical group at Mentor Graphics, which also is no more. At one time, the list included 8-to10 companies in that space, but it’s far shorter now.
Q: Your strategy within that environment?
Alex Shubat – We observed, as a physical IP provider in memory and logic libraries, that we were pretty much saturated in TAM [total available market] when we reached $50-to-$60 million, alternating between us and Artisan at the time.
So, when I became COO and CEO in 2007, we outlined two distinct strategies, a type of Business 101: Increase the Product Portfolio.
We knew we needed to expand beyond memory and logic. We had done very well with system houses, but had not penetrated the IDMs; those IDMs were still building their own IP. So, we started to expand our portfolio – that was our strategy – which we announced to our investors, our customers, and our partners.
We got into the DDR space, acquired some AMD IP [2008/9], productized that IP, took it to market, and broadened the portfolio. We acquired a division of Impinj [June 2008], which got us into the logic market. Recently, we acquired ARC [August 2009] and got into the embedded microprocessor space. And finally, we acquired the internal IP team from NXP [October 2009], and began selling IP back to them, and to the world.
This strategy has allowed us to fuel the growth of the company in two ways: a) by broadening the product portfolio; and b) by attacking the IDMs, starting with converting NXP to a customer.
Q: So, who’s your competition now?
Alex Shubat – In the area of high-speed interfaces, it’s Denali and Synopsys. In the general high-speed interface market, there are also a lot of smaller players. On the physical IP side, memory and logic, it’s still Artisan, an IP division of ARM. They’re really a premier supplier. On the processor side, we compete with ARM and Tensilica.
Q: How do you keep track of all those competitors?
Alex Shubat – To understand that, you have to observe what we’ve done as a corporation as we’ve expanded into this scenario. We re-organized into business units, which act like smaller companies keeping track of their own competitors.
We formed, for instance, a high-speed interface unit to compete with Denali and Synopsys. We formed a processor unit to compete in that market, and more recently through our NXP acquisition, we also added an analog business unit.
All of our business snits are lead by very seasoned managers – people out of Cisco, Actel, and Broadcom, for instance – who are used to running organizations with hundreds of people.
If you look at our executive management team now, you’ll see that it has been revamped and upgraded. That’s also an important step in Business 101.
Q: The strategy was clearly stated, so no one was surprised by your success?
Alex Shubat – No one was surprised, and we’ve been able to execute. Our board, our customers, and our investors all knew we would be expanding our broadline IP portfolio.
Q: You mentioned penetrating the IDM market; that‘s got to be really tough.
Alex Shubat – Yes, absolutely. The toughest!
The way we explain the whole semiconductor ecosystem, we believe there is disaggregation today. At one time, people were building everything under one roof; there used to be in-house EDA departments, internal fabs, and, of course, IP development and design capability.
The disaggregation, however, began with the semiconductor equipment makers, ASML, Applied, and so on. That was before my time, but I can relate to when EDA companies [began to] spin out of National, LSI, etc., which is where the birth of Synopsys and Cadence began in the 1980s. Then, the foundries began to serve the emerging fabless community, but not until the last 5 years have the IDMs converted to the fablite model.
It was interesting in 2007, to see a company like Texas Instruments announce they would go with the fablite model; LSI had already converted completely to that model. Then there was the whole NXP-Freescale alliance. They were going to build their own fabs, but the strategy collapsed, even though NXP may still have capacity there in the analog space.
For us at Virage, as soon as a company becomes fabless – with Virage having a lot of productized IP and customer friendly datasheets – the IDMs will switch over to buying from us. We know that, because we’ve seen it happen.
Q: Yet even with all of that, isn’t NIH still a problem with many customers?
Alex Shubat – Yes, however it’s really a control problem. We argue that, if you can let go of manufacturing control – the key piece in the chip supply chain – if you’re comfortable outsourcing manufacturing, [outsourcing IP is a natural extension].
Remember that before manufacturing was outsourced, EDA was outsourced. People were able to be in business by developing key products using EDA tools, not just developing their own tool, and that became true all the way up to tier 1 semiconductor companies.
It’s the same today with IP. We educate potential customer by helping them understand that silicon-proven IP offers a low-risk prospect. We say to our customers:
You want to build internal IP only if it helps you differentiate your end product, only if you have to prove to your customers that your chip is great because you developed the IP internally.
But, if you can’t differentiate on it, if your own customers don’t care about the graphics IP, or the signal-to-noise ratio in a particular block, then if you can get your team to accept it, [third-party IP is the solution].
As a chip company, you have to be able to tell your end customers that you’ve developed internally the key features they need, but you bought the standard capabilities that are not important to them.
Today, this [attitude] is a win-win for Virage and our customers.
Q: A lot of customization is still needed. How much of the IP business is about product and how much is about services?
Alex Shubat – If you are an IDM, at 130 nanometers, you have your own SPICE models, your own unique processes. But if Virage wants to build memories for you, we have to modify it to meet your specific process target, so half [of the offering] is product that we’ve developed and half is customization. That was [traditionally] the case with IDM customers.
As soon as they lined up with TSMC at 90 nanometers, however, we could actually ship a product to them on Day 1. Now [similarly], we have a very broad 40-nanometer portfolio. Anyone who wants it can download it with zero customization needed. Why would internal teams do it from scratch [when we can provide it].
Q: Providing better IP to the market today means more companies are willing to accept it?
Alex Shubat – Obviously, we had to drive the quality initiative up, so the customers felt comfortable, and could buy from us.
Q: On the subject of leadership: How does one learn the craft in high tech, and is it critical to be a technologist to succeed?
Alex Shubat– Dan McCranie [Virage Executive Chairman] says that the IP business is the most complex in the industry, which means a lot coming from a semiconductor veteran.
Given that complexity, in my personal view as a CEO I believe you have to be a technologist at heart. A technologist by training would probably do better than someone coming in from other channels.
Q: Technologists, however, may not understand the subtleties on the business side of things.
Alex Shubat – Yes, young technologists often make mistakes, because it’s hard to know what you don’t know. Which is why it can be good to get a mentor who’s 100-percent from the business side.
Coming into my role as CEO, with Dan McCranie being Executive Chairman and signed up to lead the transition with me – we all learned, as a company, from Don’s mentoring.
Q: Can you summarize the principle skills it takes to be an effective leader?
Alex Shubat – It’s not really possible to have just a single summery, although there are some fundamentals that help every leader. For one, you definitely have to have a goal, and a strategy for how to get there, which you have to keep reiterating to your employees. That strategy means marshaling all available resources to reach the goal. You can’t just have a disjointed strategy and be moving in multiple directions.
You also have to keep communicating the strategy to your customers and your partners, so nobody is surprised [by the direction the company is taking].
And, it’s important to continuously measure the success of the strategy.
Q: What are those metrics, and how do you measure them?
Alex Shubat – You can track metrics in a lot of things. For example, one strategy is to be early in the technology. We monitor closely when a new process rolls out, and whether we have products there. As one of our internal goals is to be early, this is an important metric to track.
Another one: We want to expand our portfolio in an organic way; we have MBOs [management by objective] for our key executives.
Also: We do not just do acquisitions for acquisition’s sake. We evaluate first. Does this acquisition give us a new capability? Is it synergistic with our current portfolio? This evaluation produces a very strict model for acquisitions.
We monitor all of these things closely.
Q: You make it sound very easy.
Alex Shubat – It’s easy to set strategies, and measure their success. That’s the business experience part. That’s Business 102!
Early on [in the life of the company], we did not have a lot of metrics. We found it was not easy to track a lot of these things, but as we transformed our management team, we actually began to exercise a lot of these practices. We now do quarterly operations reviews. Every department has metrics, which is important.
If we don’t put these things in place now, how will we do it after we pass the $100 million mark?
Q: What’s the exit strategy? IPO? Being acquired?
Alex Shubat – We are already a public company, so that exit strategy is gone. As a public company, the strategy is to build shareholder value. People coming in at $2.50 per share, for instance, want to see it go to $7, and then to $9. Whatever comes into the portfolio, we monitor how that new product affects that value.
Q: Is the ability to integrate an acquired team and technology into Virage a measure of success?
Alex Shubat – If you see how many acquisitions are done, and how many go foul, you would understand why we have so carefully educated ourselves about acquisitions.
Integrating a new entity [into an acquiring company] is a process, and you better have that process figured out [if the acquisition is going to succeed]. We’re not world experts, but we’ve done more than 5 acquisitions in the last 5 years. You can call it a capability, or you can call it secret sauce, but you better pay a lot of attention to it.
One of the simple things that helps with the integration: It’s better to be serving the same customer base, something we watch very carefully [in evaluating an acquisition] – adding another product to go through that same channel.
Q: Better to look before you leap?
Alex Shubat – It’s not at all trivial. It’s far more than just look before you leap. It’s do a complete evaluation before you leap!
Q: Where did you develop your own sophistication and ability to understand the business of technology?
Alex Shubat– There’s something in the DNA about becoming an entrepreneur. I attribute my capabilities from having emigrated through multiple countries to get where I am today. I needed to learn new languages, new environments, new cultures. These things helped with my upbringing.
I’m from Latvia, by way of Israel, France, and Canada. When I learned about Silicon Valley, I came out here for a look-and-see trip, and ended up staying. That was 25 years ago, and it’s been a great place. Quirky, but a great place nonetheless.
Editor's Note: Upcoming topics in the Business 101 Series include:
* Serial Entrepreneur
* Sustainable EDA
* VC Funding: 10% of nothing versus 100% of something
* After the M&A: Integrating the new team and technology
* Patent Prosecution: Is the answer always yes?
Peggy Aycinena is Editor EDA Confidential, and a Founding Partner of EDAMarket, hosted on GABEonEDA. She can be reached at email@example.com