In an exclusive interview, the world’s largest ODM/EMS takes aim at BoM costs.

When we spoke last to Flextronics chief technology officer Nic Brathwaite, in April 2003, he mentioned the company’s development of camera modules as complementary components for its booming cellphone manufacturing business. Today, the company has become one of the largest producers of camera modules in the world.

It’s all in a day’s work for the world’s largest original design manufacturer and contract assembler. Under the lead of its Corporate Technology Group, which Brathwaite heads, the company is performing something of a case study for the ages on how to leverage its formidable lead in cellphone technology to drive out costs for its customers – and develop new markets for itself.

In July, Brathwaite spoke with Circuits Assembly about Flextronics’ direction.

 

CA: Nic, what’s happening on the technology front at Flextronics?

NB: A lot is actually happening, I think, certainly since the last time we talked. Let me start by recapping our strategy. Our technology strategy is actually a simple one which, if executed well, will create significant value for our customers and an enviable competitive advantage for Flextronics. In the Corporate Technology Group, the fundamental basis of that strategy is actually the thesis that creating value and a competitive differentiation in the EMS industry today requires an independent ability to affect product cost, beyond the normal transformation cost. Everything we do from a technology perspective is grounded by that basic principle.

 

CA: That ties into the idea of influencing as much of a percentage of the BoM cost as possible?

NB: Absolutely. That ties into a lot of things. If the only way of differentiating yourself as a company in terms of cost is your ability to convince a supplier to give you a slightly better price than he gives somebody else, that is not a sustainable competitive advantage. That’s my personal belief. The only real way, in my opinion, to create that independent ability to impact cost is to figure out how you can impact the BoM cost by significantly taking cost out of the BoM. You have to be able to basically attack every element of cost. In some cases that means very creative and strategic partnerships. In others it means leveraging Flextronics’ own capabilities to create components and subsystems and modules.

 

CA: Let’s look at that in one particular aspect: Flextronics’ selling of its semiconductor operation.

NB: The semiconductor operation was acquired when we did the Dii acquisition. We acquired a company called ASIC International. That is really semiconductor design, and most of it is ASIC conversion, and then we have a mixed signal group that is doing very well.

 

CA: Was there a technology-related reason behind the decision?

NB: I wouldn’t call it a technology-related reason. In my opinion, we have never been able to fully capture the complete value of the semiconductor business within Flextronics, and that means both on Wall Street and internally. We continue our strategy of looking for opportunities to reduce cost, protect IP and create differentiation through silicon, and we will continue to use either this team or other silicon design services where it’s necessary.

 

CA: When you talk about getting beyond the transformation costs, where does that sale fit in? Wouldn’t that have taken you beyond the transformation costs?

NB: That is definitely true, but a lot of the attacking of costs at the BoM level actually doesn’t necessarily reside at the silicon level. Remember, the silicon capabilities we have are mainly in design services, we’re talking about FPGA conversion, we have mixed signal which we’ve been able to leverage a number of applications, but most of this is not where Flextronics’ greatest value is in terms of costs. I’m not suggesting there aren’t opportunities. Take for example, cellphones. We got into the camera modules business primarily by looking at camera modules as a cost element for a cellphone and going out and attacking that. Other areas of a cellphone where you have to attack cost are plastics and batteries and power supplies. You don’t have to have this semiconductor group to do that. And even if you do have a semiconductor group, leveraging all these areas is very difficult because many have specialized intellectual properties to develop silicon that can provide some kind of advantage. Even if Flextronics has semiconductor design services, it would never have all the capabilities and all the IP for all the pieces. That is a factor, and there is something there potentially, but I don’t believe so. Remember, it’s not necessarily having the group that gives the competitive advantage but the ability to leverage it to get the competitive advantage.

 

CA: When we talked last you were doing several analyses of total cost from design through distribution. What if anything did Flextronics change in its logistics and its thinking based on what you learned?

NB: We are definitely continuing to do this. I am personally not driving that kind of initiative. One thing we have learned is that in many cases there’s a lot more savings that can be accomplished at the back-end through distribution and the logistics chain than you can find in manufacturing. We’ve been able to get our customers to pay more attention to cost savings in those areas than we might have previously done, and to recognize the value that Flextronics brings in that area and leverage that value more. From those perspectives, that’s been very successful. And the logistics business has continued to grow for us; well over $1 billion.

 

CA: What is the Corporate Technology Group is doing?

NB: Let me first talk about what our responsibility is. Our responsibility is to leverage technology and create value and competitive situations. The group includes process technology development, product technology development (ODM, component technology) and product assurance. Part of our responsibility is to drive the technology roadmaps of the company and make sure all of the businesses have technology roadmaps that are complementary and that serve the purpose of the end-products and applications of the customers that Flextronics is going after. That has continued to play a very important role in the company. I think that the Corporate Technology Group continues to drive a lot of very significant initiatives. The ODM initiative, the cellphones, came out of these activities. The camera module business, one of the most successful businesses in the history of Flextronics, came out of that. We have a lot of other initiatives in things like power supplies and Ethernet switches. I’d say that’s working very well. We’ve been able to make sure that our technology strategies and roadmap of the different businesses are aligned and our customers are starting to give us very high marks in terms of our quarterly customer service. We have seen improvements in our technology scores which have consistently been among the highest scores in all the categories valued by our customers.

 

CA: Have you reached the point where you are driving roadmaps for your customers, instead of the other way around?

NB: I wouldn’t categorize it that way at all. Our roadmap is our response to our customers’ needs and directions. Obviously we influence our customers’ roadmaps but by no means do we drive it.

 

CA: So you don’t see the time whereby a customer would come to Flextronics and say, Hey we’re tapped out on ideas; what have you got?

NB: Well, we exchange ideas today but I don’t expect it to happen that way. We have regular technology reviews with our customers. We share ideas, we look at initiatives that we have and we show them how some of these things could be valuable to them in their applications. In some cases they adopt them, sometimes they want more data and in other cases they might reject them outright. Those things are continuing and will always happen. But I would say the level of collaboration in some of these areas with some customers is improving.

 

CA: Is the Corporate Technology Group playing any role in targeting and auditing potential Taiwanese manufacturing partners?

NB: Yes and no. When you are talking potential Taiwanese partners, several groups within our company would play a role in auditing and monitoring, whether it’s Taiwanese or other partners. I wouldn’t like to answer that question specifically to Taiwanese partners because we have technology partners all over the place. When it comes auditing and monitoring and doing due diligence on technology-related partners, we do play a role.

 

CA: I honed in on the Taiwanese because that seemed a particular area of interest in some recent company discussions on the level of revenue that Flextronics was driving from ODMs in Taiwan and the anticipation that those revenues would grow at a faster pace than the overall company.

NB: But we should take that in context. A lot of the Taiwanese ODM partnerships that we are involved in today are revolving around PC strategies. It is based on collaborating around PCs and leveraging the strength of the particular companies. If you look at Taiwan, it has significant strength and experience in PC design and motherboard design and manufacturing. And Flextronics has great strength in things like plastics, moldings, sheet metal, stamping, enclosures, design and manufacturing, power supplies. We have a world-class low-cost global manufacturing footprint. We have great logistics capabilities. Especially in PCs, it makes absolute sense for us to look for ways to collaborate and leverage our strength and the strength of some of the Taiwanese ODMs to create additional value for our customers, and in some cases, the services. Obviously these areas can expand beyond PCs into other areas. You have to bear in mind that our ODM initiatives are always in response to our customer requirements for added service and new product development. It is because our customers are asking us for these services that we are looking at ODM platforms. Before we do that we always consider the competitive match up, we always consider what capabilities we have to bring and do a GAPP analysis before we embark on any ODM initiatives. And, if it makes sense in the analysis we do collaborate rather than go it alone. This is not a new strategy, it’s just gotten a lot of press lately.

 

CA: How do you protect the IP of your customers from one company to another?

NB: That’s a very good question and something that has to be taken seriously by all of us in this business. You have to protect IP. You have to create suitable firewalls to make sure that doesn’t happen. In the software areas, for example, we have teams that are dedicated to customers and in some cases in completely different buildings and even locations. In many cases we have what are called “cool off” periods where there are several months between when a person working on one project for a company can work on a project, similar or not, for a competitor. You have to make sure you have the appropriate sensitivities within the design centers and monitoring of these capabilities. We’ve actually published our position on IP and how we go about protecting IP on our Website. It is not trivial, by any means.

 

CA: Tailing onto that, lead-free initiatives have reached a fever pitch. How is Flextronics developing and distributing data on lead-free processing and initiatives throughout the company?

NB: Let me start by saying Flextronics is one of the pioneers in pushing this in the industry.

 

CA: And there are those who love you for it and those who hate you for it.

NB: That is exactly right. We started doing lead-free manufacturing in the mid 90s. We recognized once the EU directives were launched that this was going to be a challenge. We launched initially the EMS Forum to help get other EMS companies to understand the impact of this and help us work together in the supply chain and address this in the appropriate way. This is a huge job. It’s more than just developing the processes, which we’ve had and optimized and implemented in our factory for quite a while. We’ve had the processes ready. But the whole supply-chain management is very, very difficult. Take RoHS compliance: This is more than just saying this is lead-free or a reduction of cadmium and hexavalent chromium in materials. This is fundamentally ensuring that the supply chain understands what the directives mean and ensuring that your product has none of these hazardous substances, because ultimately the OEMs have started to push the liability onto the EMS companies and it’s putting us in a tough, tough situation. It’s a challenge, and primarily not so much a technical challenge, because we’ve done a good job addressing most of those technical challenges. There are other issues around product assurance, like lead-free solder joints, like ensuring we have data to write test methodologies to ensure we have reliable interconnects. In general, I think the challenges facing us today are primarily supply chain and factory implementation.

 

CA: The Corporate Technology Group is never far away from these supply chain issues. There’s no R&D wall. Does that suggest that the technology group is searching out solutions for logistics problems?

NB: We definitely are a key member of the corporate steering committee that drives these initiatives. It doesn’t mean we have ownership, but we are an active participant. As a company, we don’t believe in driving our organization through these boxes. There’s something of a joke in this company that we don’t like organizational charts. It’s not that they aren’t useful or necessary, but one of the reasons we don’t like them is that people tend to use them to put themselves in their box and they use that box to define what is not their job. That has always been something that our corporate culture has tried not to do. If you have the ability to implement something, and the skills and ability to do something, we believe you should get the ball. To see people from different organizations working collaboratively to solve problems is not unique to our company.

When I started the Corporate Technology Group, I came from the background where we had our own manufacturing operations and I’ve worked in technology development, and on both sides I saw a lot of the problems. When I worked in manufacturing I thought the technology guys were far removed from reality and threw things over the wall that never really quite worked. When I worked in technology I thought the people in manufacturing weren’t very smart. When we started this group in Flextronics we decided we were not going to have that kind of arrangement. In everything that the Corporate Technology Group does, you could almost call us project leaders that drive the development and implementation. On the project team are all these multidisciplinary elements, so when we are doing process development as a team most of the people actually come from the factories. The Corporate Technology Group is the leader and driver.

 

CA: How is the Corporate Technology Group constituted?

NB: It is made up of quite a lot of people, including what I would call direct members and others who work on the Corporate Technology roadmap and also report to the business units.

 

CA: Speaking of business units, a lot of our readers are customers of Multek. At its May 12 investor meeting, Flextronics said vertical integration is about 20% of total sales. A big piece of that would be Multek. How does Multek fit into the Flextronics picture going forward? Is the need for in-house PCB expertise changing?

NB: I was in that meeting and I’m not sure I heard that statement exactly that way. Most people ask the wrong question. The question they ask is, What percentage of Multek’s business is internal to Flextronics? I understand why people ask it, but it’s not a meaningful question. To me, the real meaningful question is, What percentage of your internal supply comes from Multek or your camera modules? If Multek is a premier printed circuit board supplier, Multek is going to supply to everybody. The fact that 20% of Multek’s business goes internally is a good data point, but I’m not sure how meaningful that data point is. I think the bigger question is, Internally, what percentage of internal consumption comes from Multek? That’s actually very useful. I don’t know what the percentage is, but it’s pretty high.

 

CA: Is keeping Multek under the Flextronics umbrella central to what you’re trying to accomplish from a technology standpoint?

NB: There are two ways to answer that question. I think a company like Multek inside of Flextronics, if we leverage right – and I think we do a pretty good job of leveraging it – can give us a competitive advantage. If we understand the Multek technology so well that we can optimize designs and take advantage of having that relationship, then there’s tremendous benefit. If you can’t do that, if you can’t take advantage, then you don’t really get the full benefit. I think in Flextronics we’ve been able to really leverage having Multek internally. It doesn’t mean we don’t have more work to do, and that there isn’t more optimization we can do, but I think we really work hard to take advantage of having Multek inside of Flextronics. And there are cases where we see tremendous advantages. For example, there is one initiative where we have system architects working globally with process guys at Multek where we’ve been able to develop backplanes that may be among the highest performance electrical backplanes in the industry at a very competitive price. That’s just an example.

 

CA: Michael Marks has said he will step aside from day-to-day operations come January. The two of you have come across publicly as very much on the same page insofar as the vision of what Flextronics could be. Will his leaving the day-to-day operations affect the way you go about doing your job?

NB: I don’t think it will impact my job that much. First, Flextronics is a very collaborative company. A lot of the vision you see executed by Flextronics is actually a shared vision. Certainly Michael Marks and I have a very close working relationship and talk a lot. Michael is a visionary and has been a driver and key supporter of the technology strategy we have in place. But [current COO and CEO in waiting] Mike McNamara and I have worked a lot together. I started working at Flextronics 10 years ago reporting to Mike McNamara. I was responsible for driving the design services business, what is today called Flextronics Design. So Mike and I also have a very good working relationship. Michael and Mike and I are very good friends. Mike is committed to growing our company and is very competitive and will continue to do the things necessary to create additional value for our customers and a competitive business for Flextronics. That doesn’t mean there won’t be some change. But even if Michael stayed there’d be some change. The company has to change all the time in order to keep moving in the right direction. Mike has already demonstrated his commitment to a lot of the initiatives by being the biggest advocate for increasing our R&D spending and has actually started to drive additional improvements in our operational and business strategies to get further reductions in cost to help pay for some of the increase R&D. He is absolutely committed to finding ways to leverage technology to create a competitive advantage.

 

CA: Would you anticipate doing anything different in the way you manage that operation?

NB: I do not expect a lot of changes in the things we do. I spend about half my time talking to customers. It is based on those interactions with customers that we generate a lot of the ideas of things we need to do. I don’t think there’s any other company that has a Michael Marks. You’re not going to get another one. Do not underestimate Mike McNamara’s ability and his understating of the customer. They have different strengths but are very similar in a lot of ways.

 

CA: In the area of components, what is Flextronics doing in trying to influence the use of particular packages?

NB: We do not play as big a role as I would like. I think with some key semiconductor guys, we get involved early with several at the second-level package qualification. We would work to with them to evaluate new package development and new package design and technologies and perform the board/bump evaluation and qualification for them. But I would not consider that a major influence. We do influence them because we do these qualifications and analyses and come back to them and say, let’s make these changes, do this or do that, to get improvement. As we get more into the component business, as we have with camera modules, we are starting to be able to influence some of the semiconductor guys a bit more. In camera modules we definitely play a major role in influencing their packaging roadmap.

 

CA: Is component packaging an area where you see potentially huge gains for leveraging?

NB: I think it’s an area where there are huge opportunities. The issue is, How do you leverage to take advantage? Some people have suggested that one of the things the EMS companies need to be looking at is getting into semiconductor packaging to be able to do that. I believe that the EMS companies will over time have to get more involved in chip-and-wire and flip-chip technologies but for different reasons. I don’t necessarily believe that that means you have to become a semiconductor packaging company.

 

CA: One of the things you mentioned during the May 12 investor meeting was TV tuners. Why is Flextronics so hot on that technology?

NB: We are hot on looking at the components of some of the product categories that are big for us. Handheld, mobile communications is one of those things. We believe that mobile television is going to be a big area, not just for cellphones but laptops and mobile media players and even the automotive industry; instead of having a monitor hooked up to a DVD player we’ll start seeing TVs in cars. In general, I believe the market supports that position. I look at this as another module opportunity, just as we looked at camera modules a few years ago. There’s going to be a TV tuner in many products in the future, be it a notebook or cellphone or multimedia player. We think it’s going to be huge. That’s an opportunity for Flextronics to be able to provide a component into these products and again be able to influence BoM costs.

 

CA: So would we see a Flextronics branded tuner that would be used across multiple platforms?

NB: We don’t actually brand stuff, but just the same way today Flextronics is the largest provider of cameras for cellphones I would expect Flextronics would one day be the largest provider of tuners for mobile applications, which is expected to be a big market if you believe the analysts’ numbers.

 

CA: Could we see a day when a product with a Flextronics camera module or TV tuner comes with a label stuck on the box and that’s why the consumer would buy it, like Intel Inside?

NB: I have to admit, I don’t know. That’s not where we are today. Whether or not that may happen in the future I don’t know.

 

CA: Any other message you want to point out?

NB: There are a couple of messages I want to get out. I read a lot about how “Flextronics is going down this strategy of investing in design capabilities” and people call me and say, “None of your competitors are going that direction and they think you are crazy.” It is not uncommon to have multiple companies in an industry with slightly different strategies. Pay attention to those strategies: They are very simple. Does it make sense or not? To drive cost out of a system, and to create a way of independently impacting the cost; to focus on fundamentally reducing time-to-market, not just in a simple way that says, “We can turn around products one day faster than somebody else,” or “Because we have these capabilities we can take a week off your product development cycle,” but focus on taking months out of the product development cycle, focus on collapsing the product development cycle on an OEM product from 14 to 22 months down to about four to seven months. Instead of focusing on EMS vs. ODM, look at ODM services and platforms as a service that can be offered by any kind of company; it is more a business model than a business. And if you are going to do this, how do you do it without competing with intellectual capital, how do you do this without investing in product development engineering and growing your IP portfolio and partnering with technology providers? To me it’s a very simple and logical strategy. And look at the results. When we launched our ODM cellphone initiative, Flextronics was making 40 million phones a year. We launched that initiative in response to our customers who were saying, “We’re looking for this level of design capability and if you guys are not providing it we’re going to get it out of Taiwan and other places.” Flextronics today is doing at least twice that many phones if not more. Certainly, we have not lost business in general as a result of these strategies. Flextronics continued to grow revenues even during the worst downturn years, versus some of our competitors who were much bigger than Flextronics 10 years ago and today are half our size. Instead of looking at your competitors and saying that’s the wrong strategy, look at the result. I think if we continue to execute well, it’s a strategy that will continue to keep Flextronics as a key choice for many OEMs. It doesn’t mean it will be the only choice – although we’d like it to be.

 

Mike Buetow is editor in chief of Circuits Assembly.

• Prev
• Next