Cisco is one of a small handful of companies whose importance in business looms even larger than its market cap (which was, for a while, the world’s largest at the height of the dot-com boom and currently totals a hefty $159 billion-plus). Renowned as a manufacturer of advanced routing and switching technology that provides the foundation for the Internet, Cisco is also a business case study example of growth through acquisition (its adeptness at integrating acquisitions is legendary).

Like other tech companies, Cisco’s stock (and business) suffered when the bubble deflated: the stock declining from a high of near 100 (adjusted for splits) to its present range in the mid-twenties. Times were tough. But unlike many of its Silicon Valley neighbors, Cisco now finds itself well-positioned for the coming wave of Internet growth that analysts foresee with the explosion of video and voice.

As a result of the dot-com bomb experience, Cisco “got a chance to be a great company, not just a good one,” says Eugenia Corrales, Vice President, Product Operations. “We got the opportunity to break away from our competitors because everybody was hurting. And, we got the chance to really ask, ‘How robust are our processes?’”

Out of those process conversations emerged a strategic vision that-should it succeed-stands a good chance to also get written up in business school history. As a key differentiator for both its products and its enterprise, Cisco has embarked in the past several years on ground-breaking supply chain initiatives intended to make it the industry supply chain leader. To hear the leaders of the effort describe it, they had little alternative. 

In Cisco’s case, the effort becomes particularly noteworthy because the bulk of the company’s production, approaching 90%, is conducted by independent, globally diffused contract manufacturers. Approximately 2,000 Cisco people are involved directly in manufacturing; the contract manufacturers employ six to seven times as many. “You can argue,” says Jim Miller, Vice President, Advanced Manufacturing Technology, “that the core function where the Cisco manufacturing people add value is in supply chain management.”

To get a first-hand sense of how the company is re-engineering these global supply chains, I recently traveled to the outskirts of San Jose-through what used to be orchards along the Great American Parkway, past intersecting streets with names like Patrick Henry and Old Glory-to Cisco headquarters. The unexpected size of the campus, some three dozen buildings, bespeaks the scale of the ventures underway within. “Mx” is what they call the manufacturing piece, Manufacturing Excellence, the cornerstone of which is the supply chain.

Over the day, in interviews with the company’s top manufacturing executives (uncharacteristically all in town for quarterly meetings), a consistent theme emerged: supply chain flexibility was critical. To accomplish future growth, Cisco’s plan is to focus investments in proprietary technologies, move into new areas (the day I was there they announced a high-end dedicated video conferencing suite, complete with cameras and office furniture) and aggressively market their brand (complete with a consumer-friendly new logo, a stylized Golden Gate Bridge) as ‘best in class’ to give their products the kind of image that sustains maximum pricing power.

At the heart of this strategy, as it unfolds in future product generations, is Manufacturing Excellence. And at the heart of Manufacturing Excellence is a new approach to the supply chain.

Corrales had as her mission to ‘improve quality,’ which entails a revised approach to product development designed to enable new generation products while simultaneously cutting cost out of production. She offers an example: the design for a product intended to provide high performance at high speeds was prohibitively costly; “Through the Mx approach, we were able to isolate those elements which needed to run on high performance and those which could run on standard tech, thereby allowing us to cut the price by using more ordinary tech components.”

Being able to do this very much depends on what she calls “a proactive” supply chain. This began with procurement and a dramatically reduced number of approved vendors that the entire company must use. The impact of reduction in the supply base has been incredibly powerful. “We’re able to look at mutual technology roadmaps with our suppliers. That gives us more bandwidth to concentrate on our core functions while the suppliers provide the context functionality.” Technology roadmaps, the future direction of Cisco’s products, are jointly developed in Commodity Councils that include engineering teams from both the company and these outside vendors. She cites circuit boards (“lines and spaces”) as an example of how Cisco profits from these Councils: “We tell them what we’re going to want, and they tell us what they’re going to be able to deliver. From that we do a ‘gap analysis.’ If the conclusion is ‘no problem,’ the implication might be that we’re not being sufficiently advanced in our product requirements to be a leader (a violation of the Mx directive to be ‘best in class’). On the other hand, a huge gap becomes a problem that could mean we are using ‘bleeding edge’ technology, which is not where we want to be in terms of proven reliability. The more we can do with the lowest risk technology, the better.”

In refining this process, it turned out that there were no standardized sets of measurements to evaluate product development processes over time and across the breadth of the company. To rectify this, Cisco is creating a system of New Product Introduction (NPI) metrics, a home-grown tool that affords real-time monitoring and “the ability to see everything that is under development in one system.” As these metrics develop, the expectation is that “it will be increasingly possible to manage product development against standards.” To this end, company-wide implementation is merging NPI with project management software. “We want to build project management into the tool so that people have to fill out only one data set and are tracking same data. Then we can see where a project is in its life cycle management.”

Another Mx initiative, Value Engineering, incorporates lifecycle management into the product at the earliest stage. Products can be designed so that ‘cost reductions’ can be easily introduced during the last phases of product lifecycle when margins shrink. Or, so that as new technology comes on line, it can be utilized in an existing product.

Jim Miller, Vice President, Advanced Manufacturing Technology, came to Cisco during the ‘re-creation’ phase in 2002, as the focus on new products intensified. “Product innovation is the nexus of what makes the Cisco model work,” he explains. “The company doesn’t scale without new products.” But, unlike companies in industries where change moves slower, there is little margin for Cisco to miss an opportunity in the introduction of new product. “What makes the situation so challenging for us is that, where the clock speed for Aerospace, say, might be 15 years, for us it’s 18-20 months.”

Mx is conceived to provide an underpinning to help respond to such time pressures. “The current industry norm,” says Miller (who previously worked at IBM, Intel, and in supply chain operations at Amazon) “is to focus on bringing the idea from concept to market but not looking at the full lifecycle of the product.” This is most definitely not the Cisco approach.

“More than 90 percent of the cost of the product is locked in at the end of the design phase. In many companies, engineers don’t speak to manufacturing; the question for us is how to integrate them, how to get people out of functional silos and to think holistically.”

The supply chain for these products assumes added complexity since Cisco relies almost entirely on contract manufacturers. In order to oversee these processes, operating 24x7 throughout the world, an innovative management system called Autotest was developed. It captures real-time data from facilities-globally disparate manufacturers with disparate operating systems-to provide a ‘one window’ view of the state of global production lines. Intelligent agents built into the system can even trigger corrective actions long-distance when required.

“You have to ask, ‘Where do we add value?’ in the whole supply chain management piece. ‘What does our contribution entail?’ observes Miller. “We think it means figuring out what is the right connectivity and providing the right visibility.”

But, as Miller says, “Autotest is old news. The ‘new’ news is the ‘design for test’ capability.” Deploying what he calls “very sophisticated stuff,” the supply chain can be even further accelerated by having the ‘boxes’ in effect test themselves for operational capability. “In the value chain, testing is a non-value add,” he explains. “So, for a box to test itself is a big advantage. And, it affords an opportunity to find flaws and problems in prototypes.”

K.C. Wu, Senior Director, Supply Chain Management, came to Cisco in 1994 and participated in the company’s ‘lean’ transformation when it switched from doing its own manufacturing, warehousing and shipping to employing contract manufacturers. Viewed from her supply chain perspective, she regards the overarching challenge of Mx to be “supply chain visibility, real time monitoring, control and execution.” The challenging piece, she explains, “is understanding the kind of interaction you want to resolve with customers and with supply chain partners.” Maintaining alignment with those partners, she emphasizes, depends on standardized processes rather than “prescriptive directives” (such as, for example, directing your CM to use a specific system). “Standards-based partner interfaces can be leveraged and are scaleable.” Implementing such interfaces, however, requires a sharing of common goals, common processes and a common vision. “They have to see value; we have to see value.”

How is Cisco doing with this alignment? “The ordering space,” she says, “is more mature. We are aligning with partners to transform logistics and transportation. We will continue to expand the end-to-end capabilities on order management.” When we spoke, one type of transmission that had been aligned is work order dispatch. Still to come were change orders, such as schedule date, configuration, quantity changes, etc.

The priorities Mx poses to the supply chain fall into several areas, explains Ms. Wu. 

In terms of lean implementation, it is holistic optimization and expansion of leveraging partner capabilities. In logistics and transportation, it is how to leverage information across networks to optimize delivery. In product costing, Ms. Wu says, the mission is to make the request-for-quote process more efficient and integrated with total product cost management.

When thinking about NPI processes, “usually the focus is on the technical aspects,” she explains. “But to be effective, we must design for source, fulfillment and delivery in parallel to achieve the Cost/Quality/Delivery goals.”

Visibility across the entire network is the ultimate goal with ‘single source of truth.’ “It’s closer than most people think,” says Ms. Wu, “it’s not just a hypothetical.”

The destination is clear. The question is how to get there. The foundation pieces now being laid, she explains, are “the new functionality.” The managerial challenge is “how to give value to the business fast without compromising the business and system architecture.”

Steve Darendinger, Vice President, Advanced Sourcing and Supply Chain Strategy, is the executive charged with overseeing the supply chain architecture risk management, emerging market and technologies and compliance.

The importance of supply chain innovations in infrastructure and technology is to help enable sales cycle velocity, provide faster lead time and delivery to customers and offer efficiencies to the company’s supply chain operations. Consolidating and controlling the supply chain becomes even more critical, explains Darendinger, as Cisco enters “new geographies, which is integral to our growth.” ‘New geographies’ translates into emerging markets, with the tough questions regarding how to integrate new customers and markets with capability and efficiency.

In implementing initiatives, the on-going challenge is to develop supply chain platforms capable of sustaining the aspirational goal of being “world class.” The first stages have been consolidating and rationalizing the supply base and cutting costs (Cisco had some 2000 commodity suppliers at the beginning of 2001; that has now been reduced by 70%).

The next stage, as an example, is to have “methodology in place that allows technology to be integrated throughout the supply chain.” With such an integrated supply chain, Darendinger foresees leveraging the supply base capabilities to drive technology across multiple product solutions.” He expects this to become ever more critical as voice, video and data increasingly converge across the same platforms. 

Risk management is another example that assumes heightened importance within such a supply chain. Indeed, Darendinger predicts that “risk management will be the next evolution in the supply chain.” Making the Cisco supply chain-from component all the way to the customer-as risk free as possible is the goal. To do this, efforts are underway to enhance predictive capability to foresee and prepare for vulnerabilities. That can mean multiple sources in different parts of the world, alternative sites within a single provider or, beginning at the design level, seeking to use more widely available alternatives to customized components. In addition, having business continuity plans in place at all suppliers is key.

It comes down to “how to measure risk in terms of ROI,” says Darendinger. “How responsive is the supply chain when an event such as a site fire or natural disaster occurs? What level of investment versus risk is any organization or company willing to take in their supply chain? Cisco has spent significant time on building a robust risk process, and we continue to make improvements each year. This is clearly a journey but one we believe is important for the future.”

The processes entailed started in new products, which are now given risk ratings on such factors as where the component technology resides on the end-of-life cycle, the supplier’s history in terms of supply and quality, and the availability (“continuity”) of components, given their wide use throughout the sector.

The quest is underway for what Darendinger calls “a systemic approach” of risk that consists of data supported decisions. “We want to have systemic capabilities to anticipate and respond to the severity and the periodicity of a crisis so we can predict outcomes and responsiveness to those situations.” This translates into being able to quickly calibrate the supply chain to changes in the market as well. “We’re always looking at developing our supply chain to be as agile and responsive to market conditions as possible.”

Sidebar: "A Conversation With Cisco's Angel Mendez"

Angel Mendez, Senior Vice President, Worldwide Manufacturing, is a bona fide star in the supply chain world. He joined Cisco in 2005, after helping to lead the turn-around at Palm where he was in charge of Global Operations. In 2003, he was named by Business 2.0 magazine to the “Dream Team All-Star Executives Any Company Would Kill For.”

He spent an hour with World Trade Editorial Director Neil Shister for a rare ‘on-the-record’ interview.

WT: Is Cisco making a ‘new statement’ with its supply chain?

Mendez: I’m not so sure we’re making a new statement. Cisco has already been quite proud of its supply chain model. Go back in time and its fair to say that the company took a fundamental in-house manufacturing strategy and through pretty creative use of out-sourcing built a very competitive machine that was able to grow at a high rate of speed with very great consistency. So if you analyze how that worked, using our IT infrastructure we were able to bolt on lots and lots of different configurations of manufacturing nodes or supply chain nodes in order to scale at a very high rate of capacity. Had they not done that, what you would have found was a lot of factory creation, which while it could have perhaps led to the same supply chain capability, would have taken longer and the company would have missed an enormous opportunity to create its franchise that it is now, quite frankly, resting on.

So, I’m not sure that we’re saying we’re any more or less supply chain centric. What we’re saying is that we recognize the company is going through another transformation as we take on market transitions yet again. We want to benefit from the designs and the disciplines that we currently run, but we want to invent a few new ones and deploy new tactics and get smarter in some areas where we think we can improve. The collection of those two things form what is known as the MX agenda. 

That’s what MX is about. It’s a program. It’s a series of initiatives and a governance model and a series of very aspirational goals based on a clarity around where we are in the benchmarks of the world. But, it’s also a culture shift.

There are very few companies that are organized in a functional model like we are. We’re a large corporation organized-with some exceptions-where everyone is managed by one supply chain function. So we have to have the capability to range from low complexity-higher volume products to a much higher complexity-lower volume products and the spectrums of those works are quite broad. In many companies, this would be multiple divisions. The other commentary is that we are doing things in some areas where we like to think we’ve got a lot of innovation going on with processes you put in no matter what your supply chain looks like. 

Take risk management. We’re trying to be more anticipatory in terms of potential disruptions in our chains and how we would recover and how we put together redundant capabilities. Demand management would be another area-how we look at the demand environment of a service provider segment is going to be very different than how we look at the demand management of a consumer segment and how you position your supply chain response. 

Basically, you take the same procurement team, the same demand team, and they just organize themselves differently.

WT: I haven’t heard of many companies organizing their enterprise supply chains with this much sophistication.

Mendez: It’s hard! This is not an easy task we’re talking about. 

WT: What’s the toughest thing about the job for you?

Mendez: The breadth of our product line makes it difficult, that very complex portfolio. And, the fact that we innovate at an incredibly high rate. If you step back and look at our supply chain model, by and large we are a high velocity configurative order model. Everything we make from low complexity to high complexity with the minor exception of our retail business-the Linksys product that goes on a shelf in a shrink-wrap box-the lion’s share of what we make is a configurable product. When one does that across the spectrum of product families that we have, that’s hard enough; when you do it at a very high rate of new product introduction, that’s ‘very interesting.’ And, the other thing that makes it challenging is that we’re shifting our technologies quite rapidly. 

The lion’s share of Cisco’s history has been spent growing very successfully around switching and routing. The last four or five years our emphasis has been complementary to switching and routing around a number of things we call our advanced technologies. Some of these advanced technologies are very successful large markets today; some are just beginning to roll out.

WT: When you came to Cisco was manufacturing system broken in the wake of the dot bomb? Was your mandate to reinvent the process?

Mendez: No. I wouldn’t call anything here ‘broken.’ What there was, however, was two things: One, an opportunity to take a hard look at ourselves in terms of the rest of the world. We had been leaders for a long time and it felt like we were not leading in some areas. So clearly, we spent some time benchmarking to validate that. And, the second one was to put ourselves on a path that was consistent with the requirements of a changing company. A company that is transforming itself and as a result its supply change needs to grow and evolve. We felt collectively that we could do more of that at a faster rate of speed. o, I certainly wouldn’t call it a ‘broken’ problem. But, neither would I call it a ‘pure optimization’ problem. 

WT: What area is ‘the longest’ way to go?

Mendez: I would say as a company our ability to manage demand well, shape it well, digest it well and execute to it at much higher rates of inventory turns is a clear area of opportunity.

WT: Why is demand management so tough?

Mendez: In the advanced technology areas, you’re dealing with early adopter markets that are far more difficult to predict. And now increasingly in areas like service providers, where the service providers are moving very aggressively to IP (internet protocol), they in turn don’t completely understand the demands on them. If you’re talking to a cable company or you’re talking to large telecom company, they are increasingly trying to understand their customer. They’re offering voice and data bundles and now video-those sorts of things are relatively new markets. 

WT: With early adapters or service providers with IP, how can you manage demand? What are some of the approaches and techniques you’re using?

Mendez: It’s really mainly the work of our business units. They spend a lot of time with the marketing people trying to understand those dynamics. From the supply chain perspective, the important part is how to take that input and then be the most effective at managing the supply chain with that input. In other words, they bring this forecast and we have to get a lot better at collaborating with that forecast. That whole exercise is a data management challenge. We want to do it quickly in terms of information speed so that it’s fresh information flowing into your supply chain. And then, commercially, you want to take supply information and be very effective at answering that demand signal at a very quick rate of speed. 

WT: Is the information now coming together?

Mendez: It’s tough with supply chains anywhere. Forecasting and demand management are the toughest things to solve. Expand it to our level of business-that makes it very challenging.

WT: The downside is that you’re going to miss an opportunity because you don’t have enough or alternatively you’ll be overloaded?

Mendez: The risk is that you either buy too much or too little. And by definition, you’re either forced to extend your lead times or have excess inventory. We manage to balance those two things pretty well, but there’s always opportunity to do it better.

WT: Are you inventing a new business model?

Mendez: It depends what you mean by business model. But if you asked me if I was inventing a new set of supply chain tricks, in some cases we might be improving some and in some cases we might be inventing some. I think what we are doing, which is somewhat unique, is driving an adaptive supply chain in a very large outsourced model across a very large spectrum of products and geographies. That combination is an interesting thing. 

There are a number of people doing things better than we are in outsourcing. A number…perhaps in product innovation or product introduction. I have no question that’s true and we look to those things to learn from. The challenge is doing it all in the environment that we’re driving in this organization. That’s what makes what we’re doing a little trickier. So when we get it right, there’s a high sense of satisfaction. 

WT: What ideas are you interested in that aren’t necessarily Cisco-related?

Mendez: I’m fascinated in the supply chain area about how people are trying to get better at risk management. Go back in time and you think of how supply chain gauntlets have been created. For example, you have Quality Control, which became Quality Assurance, then TQM, which became Six Sigma. You look back at JIT production becoming Lean Manufacturing. Purchasing became procurement and procurement became strategic sourcing and strategic sourcing some people now call value partnering. I’m finding that risk management has leap-frogged all those evolutionary steps. Suddenly, the Board of Directors is looking at the supply chain guy and asking, “What if I have a pandemic, what’s that going to do? Another terrorist attack?’”

More and more, the practitioners like me are getting those questions, yet underlying the discipline there hasn’t been the evolution of training and tools, and procedural work. You can’t enroll your new risk management person in a professional training program like, say, with Six Sigma. 

There are other things that go along with that. How many companies are going to finance risk management? For example, the way we typically do budgeting today: you have capital budgets, you buy machines and equipment, inventory, and you have operating budgets and travel and those things. Perhaps there ought to be a risk management conversation, there ought to be investments made specifically for the purpose of protecting your business. And when you bring it out in your financials and internal budgeting processes, there might be an opportunity to bring more attention to the preparedness of the company.