presGRADUATE SCHOOL OF BUSINESS STANFORD UNIVERSITY
CASE NUMBER: EC-15 FEBRUARY 2000

CISCO SYSTEMS: A NOVEL APPROACH TO STRUCTURING ENTREPRENEURIAL VENTURES
Mike Volpi, vice president of business development at Cisco Systems, was in his office in San Jose at Cisco’s headquarters on June 27, 1997. He was considering a set of strategic questions that he had faced many times since joining Cisco’s business development group in 1994. Volpi’s colleagues had recently identified a new networking opportunity in optical routers, and Volpi wondered how Cisco should pursue the opportunity. Should Cisco develop the product internally, or should they pursue external talent that was more familiar with the technology and market segment? If the external route was the best strategy to get the right product to market on time, should Cisco build its own external venture – or just acquire somebody outright?

NETWORKING OPPORTUNITY: PIPELINKS For the previous two years, Cisco had been preaching about the promise of a multiservice network – a single network capable of transporting data, voice, and video. Cisco’s service provider customers agreed that network convergence would ultimately improve cost effectiveness and allow them to expand their service offerings. However, most service providers were saddled with huge investments in their existing circuit-based voice networks. This situation implied a market need for optical (Sonet/SDH) routers that leveraged the existing infrastructure while enabling a transition to multi-service networks: the market needed a product capable of simultaneously transporting circuit-based traffic and routing IP (Internet Protocol) traffic.1 Discussions with representatives from Cisco’s Service Provider Line of Business (SPLOB) indicated that developing optical routers internally was not a viable option. A brief search for potential acquisition targets had failed to identify any attractive companies – none of

1

Sonet/SDH (synchronous optical network/synchronous digital hierarchy) is a protocol for data transmission over fiber optic lines.

Research Associates James McJunkin and Todd Reynders prepared this case under the supervision of Professor Garth Saloner and Professor A. Michael Spence as the basis for class discussion rather than to illustrate either effective or ineffective handling of an administrative situation. The development of this case was managed by Margot Sutherland, Executive Director, Center for Electronic Business and Commerce, Stanford Graduate School of Business. Copyright © 2000 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. To order copies or request permission to reproduce materials, email the Case Writing Office at: cwo@gsb.stanford.edu or write: Case Writing Office, Graduate School of Business, Stanford University, Stanford, CA 94305-5015. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means - electronic, mechanical, photocopying, recording, or otherwise - without the permission of the Stanford Graduate School of Business. Version: (A) 02/25/00

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 2

the existing players had what Cisco was looking for in terms of people, products, technological innovation, ownership concentration, and location. In a fortunate coincidence, however, entrepreneur Amit Shah had recently approached Cisco with an idea for a Sonet/SDH router that was very similar to the product that Cisco envisioned. Shah’s company, Pipelinks, was still in the “idea” stage – so an outright acquisition was not yet appropriate. As Volpi contemplated this situation, he realized that it exhibited many similarities to a situation he had faced one year earlier. In that case, Cisco had created a custom, made-to-order company called Ardent Communications to fill a market void. Plenty of mistakes had been made in structuring the venture, but much had been done right. Volpi dug up the Ardent file and contemplated possible strategic and structural improvements that could be made, in the hope that some incarnation of the spin-in model would be an effective way to serve the current market need. BACKGROUND ON CISCO SYSTEMS INC.2 Cisco Systems was founded in 1984 by Leonard Bosack and Sandy Lerner, husband and wife computer scientists at Stanford University who invented a technology to link their disparate computer systems together. Bosack and Lerner developed the first “multi-protocol” router – a specialized microcomputer that sat between two or more networks and allowed them to “talk” to each other by deciphering, translating, and funneling data between them. Cisco’s technology opened up the potential of linking all of the world’s disparate computer networks together in much the same way as different telephone networks were linked around the world. Cisco began by competing primarily in the LAN (local-area network) market, offering high-end routers. The devices were the traffic cops of cyberspace – they directed network traffic to its final destination via the most efficient, least congested network path. As the global Internet and corporate Intranets grew in importance, so too did Cisco. With an early foothold in this rapidly growing industry, Cisco quickly became the leader in the data networking equipment market – the “plumbing” of the Internet. By 1997, approximately 80% of the large scale routers that powered the Internet were made by Cisco. Although routers, LAN switches, and wide-area network (WAN) switches would remain Cisco’s core products, the company expanded its product line to include a broad range of other networking solutions, including Web site management tools, dial-up and other remote access solutions, Internet appliances, and network management software. Despite the breadth of its product offerings, Cisco held the number one or number two position in nearly every market in which it competed. In addition, Cisco’s Internetwork Operating System (IOS) software was increasingly becoming the de facto industry standard for delivering network services and enabling networked applications.3 Cisco received its initial funding from the venture capital firm Sequoia capital, who helped to recruit John Morgridge as CEO in 1988. The company went public in February 1990 with a $222 million market value and never looked back, growing into a multinational corporation with over 10,000 employees in 54 countries. By 1997, revenues had increased over ninety-fold since

2

Excerpts taken from “Cisco Systems, Inc. Acquisition Integration for Manufacturing”, Case # OIT-26, Graduate School of Business Stanford University and Harvard Business School, revised January 1999. 3 Cisco’s IOS Software was the industry leading internetworking software, like Microsoft Windows for networking. IOS is a platform that delivers network services and enables networked applications. IOS enables interoperability connections between otherwise disparate hardware, and accommodates network growth, change, and new applications. It also contains security features, including access control, authentication, firewall, and encryption.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 3

the IPO, from $69.8 million in fiscal 1990 to $6.4 billion in fiscal 1997. (Exhibit 1) In June 1997, Cisco’s market value totaled $46.3 billion. Two highly respected CEOs have led the company: John Morgridge, and his successor, John Chambers. Morgridge helped to shape the Cisco culture from day one, focusing on customer satisfaction, product quality, and frugality. He once gave a legendary presentation on frugality to the Cisco sales force, after being appalled by reports that salespeople were flying first class on business trips. Equipped with slippers, earplugs, and eye covers, Morgridge displayed how to fly coach and make it seem like first class. John Chambers, who joined Cisco in 1991 and succeeded Morgridge in January 1995, was well known for his fair but ultra-competitive nature. Chambers, a former IBM and Wang Laboratories marketing and sales veteran, fostered Cisco’s strong customer focus and was credited with continuing Cisco’s striking success in the networking industry. Corporate Strategy Throughout the 1990’s, organizations of all sizes were beginning to recognize the value of their information networks and the Internet as a source of business advantage. As a result, more of Cisco’s customers sought end-to-end networking solutions. Building on its expertise in routers, Cisco strove to deliver a wide range of new products, expand their offerings through internal and external efforts, enhance customer support, and increase presence around the world. The main element of Cisco’s strategy during this expansion phase was to maintain a passionate customer focus and consistently work toward exceeding customer expectations. To deliver on that goal, Chambers realigned the organization along lines of business specifically targeted to the three key markets Cisco served: Enterprise, Service Providers, and Small/Medium Business. The new organization enabled Cisco to provide market specific, end-to-end solutions that included integrated software, hardware, and network management. It also allowed Cisco to customize its sales, support, and business programs to each market. One of the keys to the company’s success was the Cisco brand, which was recognized as a leading name in networking. Customers associated the Cisco brand with a secure, reliable, highperformance network. Chambers wanted to enhance and expand the brand in the future, and increased Cisco’s marketing efforts to include television, Internet, and print advertising. The ongoing deregulation of telecommunications and technology convergence were driving the trend toward integration of voice, video, and data networks. Historically, there had been three separate types of networks: phone networks for transmitting voice, computer networks for transmitting data, and broadcast networks for transmitting video — but advances in digitization allowed these forms of communication to be translated into binary computer language. This, in turn, made it possible to transmit voice, data, and video over one network in a more efficient and economical manner than using three disparate networks. As a result, phone companies were beginning to transform their archaic voice networks into unified, multi-service networks. Chambers believed that this transition to the “New World” of communications would create exciting opportunities for Cisco to capture share in the $250 billion telecom equipment market, which was previously dominated by huge, well-capitalized companies such as Lucent Technologies and Northern Telecom. These competitors were so large that Chambers was able to instill a “David vs. Goliath” mentality within Cisco. While expanding into these new markets,

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 4

Cisco strove to maintain its product leadership in each of the market segments it served. The product leadership strategy involved the innovation of Cisco’s engineering teams, complemented by alliances, acquisitions, and minority investments. Building Shareholder Value through Acquisitions, Investments and Alliances As the networking space became more competitive, and as minimizing time to market became increasingly important, Chambers realized that Cisco could not keep up with the changing market needs solely through internal development. Using acquisitions and alliances to gain access to world-class technologies and people became a defining component of Cisco’s strategy. This strategy was relatively unique in the high-tech world at the time — many companies viewed looking to the outside for technological help as a sign of weakness. Chambers commented on the acquisitions and alliances strategy: They are a requirement, given how rapidly customer expectations change. The companies who emerge as industry leaders will be those who understand how to partner and those who understand how to acquire. Customers today are not just looking for pinpoint products, but end-to-end solutions. A horizontal business model always beats a vertical business model. So you’ve got to be able to provide that horizontal capability in your product line, either through your own R&D, or through acquisitions.4 Although Chambers and Ed Kozel, Cisco’s chief technology officer, were a key driving force behind Cisco’s business development strategy, many in the industry regarded Mike Volpi as the man responsible for shaping Cisco’s legendary business development practice.5 When Volpi joined Cisco in 1994 after graduating from the Stanford Graduate School of Business, Cisco had completed only one acquisition, Crescendo Communications. Two more acquisitions closed soon after Volpi arrived, but he was deeply involved in all subsequent acquisitions. Before pursuing a new market opportunity, Volpi’s group assessed the merits and downsides of the “buy vs. build” strategies. If Cisco did not have the technological capability, engineering talent capacity, or time to develop the product internally, the business development group would often opt to acquire or partner with an external player. Although the acquisitions made headlines, licensing, partnering, and investing were equally important to Cisco’s strategy. Cisco was very active on the minority investment front, which gave the company insight into new technologies without the risk of deploying internal development resources. Volpi used a simple chart to put companies into the right context (Figure 1):

4 5

“The Art of the Deal”, Business 2.0, October 1999. Volpi initially reported to Charles Giancarlo, who joined Cisco through the Kalpana acquisition in 1994 and served as VP of business development until March 1997, when Volpi assumed that title.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 5

Figure 1: Range of Cisco’s Business Development Activities
Acquire Invest Joint R&D/ Marketing/Sales

Degree of Strategic Value

OEM/ License

Level of Commitment The public equity markets were the principal exit strategy for hot high-technology start-ups, but a Cisco acquisition represented an appealing exit alternative for many networking companies. Cisco reigned supreme in the technology industry as the most effective company at identifying, acquiring, and successfully integrating companies into their culture. By June of 1997, after the Ardent deal closed, Cisco had acquired 19 companies for an aggregate total of roughly $7 billion. (Exhibit 3: Summary of Cisco’s Acquisitions as of June 1997) What were the keys to Cisco’s success? Why did they do this better than the competition? “We made every mistake in the book,” Volpi frankly stated, “but we learned from these mistakes, and they have helped us in subsequent transactions.” Cisco made predominantly “small” acquisitions of private companies, with typical deal sizes of $200 million or less, instead of acquiring large, established, public companies.6 The primary inclination towards smaller acquisitions was the relative ease of integration – large, established companies with strong corporate cultures presented greater integration challenges. Further, Chambers asserted that Cisco did not acquire to gain short term market share, but was instead looking for technology and talent for the future: When we acquire a company, we aren’t simply acquiring its current products, we’re acquiring the next generation of products through its people. If you pay between $500,000 and $3 million per employee, and all you are doing is buying the current research and current market share, you’re making a terrible investment. In the average acquisition, 40 to 80 percent of the top management and key engineers are gone in two years. By those metrics, most acquisitions fail.7 Charles Giancarlo, Cisco’s vice president of business development from 1994 to 1997, reiterated the importance of acquiring and retaining key people: When you are buying a company… it’s obviously not for today’s products. That means keeping the people in place who can create that growth. We won’t do a deal if a company has ‘golden parachutes’ – accelerated vesting for employees

6

The $4.6 billion acquisition in April 1996 of StrataCom, which filled Cisco’s hole in WAN switching products, stands out as an exception. 7 “The Art of the Deal”, Business 2.0, October 1999.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 6

that kicks in once a company is sold. The minute you buy the company, they all get rich. We prefer ‘golden handcuffs’, which are applied with two-year noncompete agreements with key executives and technical personnel at the target companies, and the provision of Cisco stock options that vest over time.8

LOOKING BACK TO 1996: THE MARKET OPPORTUNITY FOR A NEW ACCESS PRODUCT In 1996, the rapid evolution of network infrastructure was creating business opportunities in virtually every sector of networking. Cisco’s unique vantage point allowed the company to rapidly identify these new markets. By early 1996, Cisco believed that an unmet need existed for a product that could inexpensively carry voice, data, and video traffic from a company’s localarea network to the wide-area network. Cisco identified two principal customer needs. The first need was to simplify and improve management of network access equipment. The conventional approach to public network access required that disparate hardware components (such as leased line modems, channel banks, etc.) be cabled together, creating a complex hardware puzzle. Companies incurred high maintenance costs and trouble-shooting nightmares since each component was controlled by its own management system. The second customer need was to optimize use of expensive WAN bandwidth. Despite the industry buzz about high-speed ATM trunks,9 Cisco believed that these solutions would remain expensive – especially in comparison to LAN bandwidth – where Ethernet technology10 dominated. The company expected that high speed network access solutions, running at T3 (4.5Mbps) or OC-3 (155Mbps), would be confined to niche markets for the foreseeable future. Most customers would choose the slower, more economical T1/E1 (1.5Mbps) link to the WAN. These factors highlighted a market opportunity for an access solution that aggregated LAN-based data, voice, and video traffic over the low cost T1/E1 ATM trunk. This solution would be sold to service providers, helping them to: • Provide an integrated T1/E1 access solution that was cost-effective enough for wide deployment • Contain costs by using a single product in multiple applications • Contain upgrade/conversion costs by using a remotely configurable product • Contain support costs by using a product with an interface that both customers and service providers were familiar with. This product concept was the genesis of Ardent Communications.

8 9

“Cisco’s Secret: Entrepreneurs Sell Out, Stay Put”, Inc. Magazine, March 1997. A trunk is an access line that connects remote offices or central sites to the service provider network. Asynchronous Transfer Mode (ATM) is a data transfer technique where multiple service types, such as voice, video, or data, are conveyed in small, fixed-size cells. 10 In 1996, Ethernet technology penetrated every corner of the Enterprise network with 10BaseT (10Mbps), 100BaseT (100Mbps), and the coming Gigabit Ethernet (1000Mbps).

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 7

New Venture Strategy In 1996, Volpi contemplated the traditional buy and build alternatives when he considered how to address the market that Ardent Communications would eventually serve. Building the product in-house had several obvious advantages – most notably not having to integrate two different organizations. The multi-service access business unit had been doing similar things on a day-today basis, but did not have idle human resources to devote to the new project. Diverting resources away from current projects was not a viable option. After talking with the business unit VPs, it became clear that building the product in-house would take too long – recent competitive pressures from 3Com, Ascend, US Robotics, and Micom made time to market a priority. The business development team concluded that Cisco had neither the time nor the resources to go after the new market on its own. Buying a company whose products addressed this market was also an option. In this case, Cisco had a clear conception of the market need, but was unable to identify attractive companies that were focused on this space. Volpi’s experience suggested that finding the right acquisition target would be a difficult task – in all cases Cisco would have to spend lots of time and effort modifying the product set and integrating the newly acquired company into the Cisco organization. In addition, retaining key employees post-acquisition was always a challenge. After considering the buy and build alternatives and finding neither to be satisfactory, Volpi mused, “Why not custom make a start-up to build exactly the product we want, and then buy them later if they succeed?” In essence, this solution would entail the creation of a new venture designed as a spin-in from day one – “build to buy,” a hybrid of the buy and build approaches. At first glance, the spin-in model seemed to address three key issues: time-to-market, recruiting top talent, and integration with the relevant Cisco business unit. As the Cisco business development team thought more deeply about the new venture approach, they realized that the hybrid nature of the spin-in solution raised several difficult tradeoffs. What sort of structure would allow the start-up to leverage Cisco’s strategic assets without quashing the entrepreneurial feel? How should Cisco structure the venture to minimize the tradeoff between the virtues of independence and the need for smooth ex-post integration? Would it be possible for Cisco personnel to coach the new team without stifling creativity? Should Cisco invite other investors to participate in the financing? How large an initial ownership stake should Cisco take in the venture? Incentives would also be a major issue: How could Cisco implement a structure that would provide the right incentives for the new venture’s management and employees, without upsetting the current Cisco employees who would assist in the new venture’s integration? At the end of the day, the new venture would have to live within an existing Cisco business unit, at which point it would rely heavily on Cisco employees for success. Structuring the Ardent Communications Venture To develop a potential model for the new venture approach in 1996, Volpi and Kozel had reflected upon an even earlier deal that Cisco had considered. In the spring of 1996, Wu Fu Chen, a successful networking entrepreneur was working with Sequoia Capital and two Cisco employees to launch a new networking company. The idea for this company came from the Cisco employees, who intended to leave their jobs at Cisco to build a solution which they hoped their employer would want to acquire. The product concept had a great deal of potential, and the founding team was flush with engineering talent. Wu Fu Chen had an especially impressive

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 8

background – he had co-founded four companies since 1986, including Cascade Communications and Arris Networks. Despite the promise of this networking idea, the Cisco business development team declined to invest. The decision came from the top – Chambers believed that funding Cisco employees to go out and build new networking companies would set a dangerous precedent. At the time, Mike Volpi and Ed Kozel recognized that Wu Fu would be an excellent person to recruit as President and CEO of the proposed spin-in venture, which they would call Ardent Communications. Kozel contacted Wu Fu and outlined the Ardent business idea and Cisco’s spin-in concept. Volpi later characterized the initial message to Wu Fu as simply, “Make this product and we’ll give you lots of money.” After a series of discussions, Wu Fu agreed to head up the Ardent venture. Defining the Ardent 101 Product In early June of 1996, Kozel, Volpi, and Wu Fu outlined the basic functional specification for the first Ardent product, tentatively called Ardent 101. For Cisco to buy the new company, Wu Fu and his team needed to develop a traffic aggregation device for data, voice, and video with certain functional requirements. (Figure 2) The group also developed a set of milestones that would set expectations for the product timeline. (Figure 3) Figure 2: Ardent 101 Functional Requirements 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Ability to accept data, voice, and video traffic Aggregate up to 2Mbps traffic on the WAN side Support ATM, Frame Relay, and TDM trunks Support standard office environments Support Bridging, IP Routing for LAN data traffic Support Circuit Emulation for Voice and Video Applications Support voice and data compression The target list price for the base configuration is about $5,000; Cost of goods target is $800 or lower Will consider support of data encryption Support European requirements (E1)

Figure 3: Ardent 101 Milestones 1. Six months after the Effective Date of the Agreement, the Company shall have completed the specifications for function, architecture, and design for the Product Twelve months after the Effective Date of the Agreement, the Company shall have begun integration of the Product Fifteen months after Effective Date of the Agreement, the Company shall have begun the beta program for the Product Eighteen months after the Effective Date of the Agreement, First Customer Shipment shall have occurred

2. 3. 4.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 9

Capital Structure By June 14, 1996, Cisco and Wu Fu’s team had agreed on a preliminary term sheet for the new venture. (Exhibit 4: Memorandum of Terms For Private Placement.) Kozel and Volpi decided to invite the venture capital firm Sequoia Capital to participate in the financing since this would create more of a start-up feel. To foster an entrepreneurial environment with strong employee incentives, Cisco agreed to give the founding team and employees a large ownership position – over 55% on a fully diluted basis. Cisco sought an equity stake of only 32% for itself. This was a major departure from the large equity shares other parent companies were requesting in their spin-ins and spin-outs (arguing that their intellectual property, brand name, and other resources entitled them to “free” equity). Sequoia Capital also took a relatively small equity position of 11%. All parties agreed that a balanced board of directors would deliver the right degree of control over the company’s direction. Initially, the Board would consist of Wu Fu, Ed Kozel, and Sequoia’s Mike Goguen. Unlike most venture deals, the Series A and Series B rounds were negotiated simultaneously, with closing dates less than two months apart. In the A round (July 11 closing) Wu Fu and the other members of the founding team would purchase 3 million shares of Series A Preferred Stock at $0.333 per share. The low share price was analogous to cheap founders’ stock in an entrepreneurial venture. Neither Cisco nor Sequoia would participate in the A round. The implied post-money valuation as of July 11 was $2.4 million. For the B round, the new company decided to issue 11 million shares, up from the 9 million shares outlined in the initial term sheet. On August 30, Ardent received the first cash infusion of the B round, in which Sequoia Capital purchased approximately 2.5 million shares at $1.00 per share. Cisco also made its investment at $1.00 per share, purchasing 7.535 million shares of Series B Preferred Stock on September 20. Seven days later, the founders purchased another one million shares. The remaining equity capitalization consisted of 9.25 million shares of common stock. Of this total, approximately 3 million shares would go to the engineering team in the form of option grants. The implied post-money valuation as of August 30 was $23.3 million. The rough capitalization table used by Cisco is described in Exhibit 5. Retaining Key Employees Volpi knew that even though Cisco was creating Ardent to produce a very specific product, it was the people, not the product, that represented a significant portion of Ardent’s value. With this in mind, Cisco laid out a four-year vesting period for the options granted to employees – 25 percent would vest after the first year, with the remainder vesting monthly over the next three years. Upon a change in control, like the planned acquisition by Cisco, none of the employee vesting would accelerate – the only exception being made for Wu Fu Chen. Wu Fu’s vesting would accelerate such that at most one year of vesting would remain, but he was subject to a oneyear lock up agreement which kept him from leaving Ardent upon acquisition. Facilitating the Spin-in: The Put/Call Feature Cisco needed to create a legal mechanism that would allow Ardent to cleanly “spin-in” at some point in the future. After a series of discussions, the founding team proposed a simple put/call structure that would give Cisco the option to purchase the company at a pre-specified price, but would also obligate Cisco to purchase the company if the new team succeeded in building the

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 10

product. This was the first time that Cisco had integrated a put/call feature into a strategic investment. The model intrigued both John Chambers and Ed Kozel, who viewed it as an innovative mechanism for developing a made-to-order company. The “Option” section of the term sheet explained the call option:
Until the earlier of fifteen (15) months from the closing or one (1) month after the first customer shipment, Cisco shall have the right to acquire either all of the outstanding equity securities of the Company, or all of the Company’s assets, in Cisco’s discretion, for a purchase price of $232,500,000, payable either in cash or equity securities of Cisco.11

Since Cisco would also write a put option, the shareholders in the new venture could force Cisco to purchase the company at the pre-specified price, so long as the ten specific functional requirements were met. To keep matters simple, the put and call would have the same strike price. The put option read:
…if First Customer Shipment occurs within (15) months after the functional requirements for the Product are first defined, and in Cisco’s reasonable judgment, the product meets the specifications set forth, each of the security holders shall sell its Securities to Cisco, and Cisco shall be obliged to and will purchase such Securities, in accordance with the purchase price and other terms of purchase…12

Cisco believed that although the put/call structure truncated the upside for investors and employees, it mitigated enough risk to make the investment or employment decision attractive from a risk/reward standpoint. The option agreement turned out to be a very effective recruiting instrument. If the product requirements and milestones were met, the 15-20 person engineering team would share a $30 million payout less than 15 months out. The five person founding team would do even better: delivering on the product would allow them to share more than $100 million. Leveraging Cisco’s Assets IOS The Ardent product would complement Cisco’s existing multi-service access products (called the 3800 product family). To facilitate interoperability, Volpi decided to license Cisco’s IOS software to Ardent free of charge until Cisco’s option to buy the company expired. IOS was to be used as the architectural foundation for Ardent 101. Ardent would focus on adding the technologies of ATM and Frame Relay over a T1/E1 connection, circuit emulation for digitized voice over ATM or Frame Relay, voice compression, and telephony capabilities. These changes were not on the official evolutionary path of IOS, though they were similar to some development work being done within Cisco. In addition, IOS had been created and modified by numerous Cisco employees for use in the various products Cisco sold, but not to be sold as a shrinkwrapped software product. The procedures to use and adapt IOS were not well documented, which would present a challenge to the Ardent employees. Under the terms of the licensing deal, Cisco would retain all ownership of IOS, including software developed by Ardent to interact directly with IOS.
11

Excerpt from Memorandum of Terms for Private Placement of Series A and B Preferred Stock of Ardent Corporation, June 1996. 12 Ibid.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 11

Licensing IOS software to Ardent for free was a very contentious issue – it upset several Cisco personnel who felt that the company was giving away the crown jewels, the real value-add in Cisco’s solutions. Disgruntled Cisco employees argued that Cisco was giving away a prized possession and then paying to buy it back. Engineering talent Ardent would also need engineering help to integrate IOS into their new product. To address this issue, a small group of Cisco employees were selected to work with the Ardent team throughout the development process. This was not unusual, because Cisco had provided consulting services from time to time in the past. Ardent paid the standard fee for these engineering resources, $250,000 per engineer per year. Since this was regarded as a temporary assignment, these people would continue to be employees of Cisco under the same terms as they had before Ardent surfaced. Testing and certification facilities Other resources provided by Cisco included testing and certification facilities. Cisco decided to allow Ardent to use its testing and certification facilities free of charge until the option period expired, after which it would charge Ardent a nominal fee. Business unit expertise One key issue that Volpi and his colleagues debated was the extent to which the multi-service access business unit should coach Ardent through the development process, and generally stay informed as to what progress had been made. Cisco did have some similar products in the pipeline, but none overlapped significantly with Ardent 101 as defined in the product requirements document. After some debate, Volpi and Kozel decided not to involve the business units until after the Ardent product was completed. This approach seemed appropriate because the product specification had already been narrowly defined, minimizing the degrees of freedom and therefore the need for frequent coaching or updates. Hence, Ardent operated in stealth mode through the end of 1996 and early 1997. Cisco form factor Both Cisco and Ardent wanted the new product to have the look and feel of the Cisco product line, although this was not a stated requirement in the product specification. Instead of designing a tailored box for Ardent 101, Wu Fu decided to adhere to the existing Cisco product line and imposed an additional requirement on his engineers: they would adopt the form factor for the Cisco 2500 series. This solution was compact and familiar to Cisco’s carrier customers. Squeezing Ardent 101 into the 2500 box would be an engineering and manufacturing challenge, but the Ardent engineers felt confident it could be done. Accounting issues The Ardent spin-in model had implications on the accounting methods Cisco could employ to account for the venture and complete the spin-in. Many of Cisco’s acquisitions had been done using the pooling-of-interests method. For acquisitions where the price Cisco paid was far in excess of the assets’ book value, the pooling method was generally preferable.13 Using pooling of interests accounting required, among other things, that the acquisition occur in a single
13

Net income is generally lower under the purchase method because significant goodwill, an intangible asset which represents the excess of the purchase price over the assets’ book value, must be amortized over a defined period.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 12

transaction where over 90% of the company was acquired, and that there be no prior control exerted on the company. Since Cisco took an initial 32% stake in Ardent, and the call option translated into significant control, pooling was never a possibility – but the strategic value of the venture outweighed the accounting effects. Volpi commented, “We certainly look at the accounting impact in our decision process, but I don’t think that accounting issues should ever dominate the strategic issues in making decisions.” The accounting for Cisco’s investment in Ardent was relatively straightforward. Since Cisco owned more than 20% but less than 50% of the company, GAAP required that it use the equity method of accounting. Hence, the Ardent investment was recorded on Cisco’s books at acquisition cost plus its pro-rata share of Ardent’s earnings (or losses). The Ardent Acquisition In late June 1997, Volpi and Ed Kozel discussed whether Cisco should exercise its option to purchase the outstanding shares of Ardent. In many ways, the decision was immaterial, because Wu Fu and his team were likely to meet the acceptance criteria and exercise their put option, obligating Cisco to purchase the company. For this reason, the key consideration was timing: Should Cisco wait until the end of the option period to spin-in Ardent, or was it better to do it now? Volpi and Kozel determined that doing the deal sooner rather than later would deliver two principal benefits. First, Cisco could begin the task of integrating Ardent into the Cisco family, speeding time to market for the Cisco branded solution. Second, purchasing early would help to avoid confusing the marketplace. Ardent’s marketing people had begun to put their own spin on the product, now called “Integress.” However, Cisco might want to use a different marketing tact. If Ardent had the opportunity to educate the marketplace, Cisco would either be forced to continue the same marketing program or re-educate potential customers. Historically, the reeducation approach had not worked well. In fact, many competitors were highlighting Cisco’s “inconsistent messages” in their white papers and marketing materials. On June 24, Cisco announced its intention to acquire Ardent. The press release stated: “Under the terms of the acquisition agreement, shares of Cisco common stock worth approximately $156 million will be exchanged for the outstanding shares and options of Ardent.” (Exhibit 6) This was consistent with the agreed upon total acquisition price of $232.5 million, since Cisco already owned 32% of the company. On a per share basis, Cisco paid approximately $10.00. This worked out well for the founders, who received approximately $102.3 million – more than 100 times their initial investment. Sequoia Capital received a payout of $24.6 million, a relatively small number but still 10 times money invested in less than 12 months. Although the Ardent deal had several flaws, Cisco had learned a lot about how to structure future deals from the acquisition. Volpi turned his attention away from the Ardent acquisition and back to the Pipelinks opportunity.

JUNE 1997: VOLPI CONSIDERS THE PIPELINKS OPPORTUNITY The new market opportunity for a Sonet/SDH router capable of simultaneously transporting circuit-based traffic and routing IP (Internet Protocol) traffic had been identified by several of

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 13

Volpi’s colleagues at Cisco. The product would be targeted at many of Cisco’s service provider customers who were struggling to bring their networks into the “New World” of unified networks, but wanted to make the transition without scrapping their existing circuit-based TDM infrastructure. One of the factors Volpi considered was Cisco's expertise in the domain of optical routing. In 1997, it was limited. High-speed Sonet/SDH networking solutions were much larger and more expensive (with price points in the hundred thousand to multi-million dollar range) than Cisco’s traditional products. Because this was a new and unfamiliar market for Cisco, Volpi and his team looked externally to pursue the market opportunity, but none of the existing players met Cisco's criteria. Volpi's team believed, however, that an idea that successful entrepreneur Amit Shah had approached Cisco with might address this market need. Shah, who had sold his first networking company to Cabletron Systems, a Cisco competitor, had been brainstorming for his next idea. During that process, he realized there was an increasing need for bandwidth on the access points of the Internet infrastructure – the “metro” space near large population areas. Shah conceived a Sonet/SDH router product very similar to that which Cisco envisioned. He called the idea Pipelinks. After a series of discussions, Volpi and Kozel determined that Cisco would like to work with Shah to bring his product to market. Volpi and Kozel began to think about invoking the Ardent spin-in model they had developed one year earlier because the Pipelinks concept was at an early stage. Shah was immediately intrigued by the idea – it seemed like an effective way to address many of the challenges he faced, namely: (1) raising funds, (2) recruiting the right people for the team, and (3) successfully executing with customers. Once Shah had agreed in principle to structure Pipelinks as a spin-in, Volpi sat down to outline the terms of the deal. As he dug through the license agreements, term sheets, and product requirements in the old Ardent file, Volpi identified several potential improvements and modifications that he would make.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 14

Exhibit 1 Cisco Systems Historical Financials

BALANCE SHEET FISCAL YEAR ENDING JULY 31, ANNUAL ASSETS (000s) CASH MARKETABLE SECURITIES RECEIVABLES INVENTORIES OTHER CURRENT ASSETS TOTAL CURRENT ASSETS NET PROP, PLANT & EQUIP INVEST & ADV TO SUBS DEPOSITS & OTHER ASSET TOTAL ASSETS

1990

1991

1992

1993

1994

1995

1996

1997

$

$

35,842 $ 21,102 15,874 3,701 1,673 78,192 4,114 367 82,673 $

40,323 $ 51,104 34,659 6,078 8,797 140,961 12,665 519 154,145 $

39,955 $ 116,477 61,258 9,142 20,244 247,076 28,017 46,866 1,974 323,933 $

27,247 $ 53,567 $ 284,388 $ 279,695 $ 61,738 129,219 279,754 758,489 129,109 237,570 421,747 622,859 23,500 27,896 81,805 301,188 26,702 59,425 116,466 197,409 268,296 507,677 1,184,160 2,159,640 48,672 77,449 172,561 331,315 274,260 457,394 583,871 1,060,758 3,985 11,174 51,357 78,519 595,213 $ 1,053,694 $ 1,991,949 $ 3,630,232 $

269,608 1,005,977 1,170,401 254,677 400,603 3,101,266 466,352 1,630,390 253,976 5,451,984

ANNUAL LIABILITIES (000S) ACCOUNTS PAYABLE $ ACCRUED EXPENSES INCOME TAXES TOTAL CURRENT LIAB OTHER LONG TERM LIAB TOTAL LIABILITIES MINORITY INTEREST SHAREHOLDER EQUITY TOT LIAB & NET WORTH INCOME STATEMENT FISCAL YEAR ENDING JULY 31, NET SALES COST OF GOODS GROSS PROFIT R & D EXPENDITURES SELL GEN & ADMIN EXP OPERATING INCOME NON-OPERATING INC INTEREST EXPENSE INCOME BEFORE TAX TAXES NET INCOME $

$

4,973 $ 6,290 1,976 13,239 123 13,469 69,204 82,673 $

7,743 $ 17,965 542 26,250 436 26,686 127,459 154,145 $

16,262 $ 46,953 15,108 78,323 78,323 245,610 323,933 $

24,744 $ 31,708 $ 59,812 $ 153,683 $ 77,492 130,846 257,099 445,776 17,796 42,958 71,970 169,894 120,032 205,512 388,881 769,353 120,032 205,512 388,881 769,353 40,792 41,257 475,181 848,182 1,562,276 2,819,622 595,213 $ 1,053,694 $ 1,991,949 $ 3,630,232 $

207,178 656,707 256,224 1,120,109 1,120,109 42,253 4,289,622 5,451,984

1990 69,776 $ 23,957 45,819 6,168 18,260 21,391 2,088 23,479 9,575 13,904 $

1991 183,184 $ 62,499 120,685 12,687 41,809 66,189 4,567 70,756 27,567 43,189 $

1992 339,623 $ 111,243 228,380 26,745 72,248 129,387 6,719 136,106 51,720 84,386 $

1993

1994

1995

1996

1997

$

649,035 $ 1,334,436 $ 2,232,652 $ 210,528 450,591 742,860 438,507 883,845 1,489,792 44,254 106,680 306,575 130,682 276,995 485,254 263,571 500,170 697,963 11,557 22,330 40,014 275,128 522,500 737,977 103,173 199,519 281,488 171,955 $ 322,981 $ 456,489 $

4,096,007 $ 6,452,000 1,409,862 2,243,000 2,686,145 4,209,000 399,291 1,210,000 886,048 1,370,000 1,400,806 1,629,000 64,019 262,000 1,464,825 1,891,000 551,501 840,000 913,324 $ 1,051,000

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 15

Exhibit 2 Cisco Systems Monthly Stock Price Chart: February 1990- June 199714

$18.00 $16.00 $14.00 $12.00 $10.00 $8.00 $6.00 $4.00 $2.00 $Aug-90 Aug-91 Aug-92 Aug-93 Aug-94 Aug-95 Aug-96 Feb-90 Feb-91 Feb-92 Feb-93 Feb-94 Feb-95 Feb-96 Feb-97

14

Prices adjusted for all splits since IPO, based on January 25, 2000 stock price.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 16

Company
Crescendo Communications, Inc. Newport Systems Solutions, Inc. Kalpana, Inc.

Exhibit 3 Summary of Cisco’s Acquisitions as of June 1997 Date Purchase Price Description
September 1993 July 1994 $95 million $93 million High-performance work group CDDI and FDDI switching solutions. Software-based routers for remote network sites of small/medium sized networks. Manufacturer of modular and stackable LAN switching products extend the usability and data capacity of existing Ethernet LAN’s. Jointly held company formed in 1993 by Bolt Beranek and Newman and UB Networks offers enterprise ATM switching, workgroup ATM switching, LAN switching and routing. Supplier of ISDN (Integrated Services Digital Network) remote-access networking products useful for telecommuting and other networked applications. Developer of Internet gateway software connecting central and remote office desktop users with the Internet. Inventor and leading supplier of Fast Ethernet (100BaseT) and Ethernet desktop switching products. Manufacturer of cost-effective, low maintenance network address translation and enterprise Internet firewall hardware and software. Internet software products for connecting disparate computer systems over local area, enterprise-wide and global computing networks including the Internet. Leading supplier of Asynchronous Transfer Mode (ATM) and Frame Relay high-speed wide areas network switching equipment transporting a wide variety of information, including voice, data and video. Modem ISDN Channel Aggregation (MICA) technologies will deliver highdensity digital modem technology with Cisco’s dial-up and access product lines.

October 1994

$240 million

LightStream Corp.

October 1994

$120 million

Combinet, Inc.

August 1995

$132 million

Internet Junction, Inc.

September 1995

not public

Grand Junction, Inc.

September 1995

$400 million

Network Translation, Inc.

October 1995

not public

TGV Software, Inc.

January 1996

$138 million

StrataCom, Inc.

April 1996

$4.666 million

Telebit Corp’s MICA Technologies

July 1996

$200 million

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 17

Exhibit 3 (cont’d) Summary of Cisco’s Acquisitions as of June 1997 Company
Nashoba Networks, Inc.

Date
August 1996

Purchase Price
$100 million

Description
Token ring switching technologies for providing users with a wide choice of employing highperformance switched workgroup and backbone Token Ring environments. Standards-based multilayer Gigabit Ethernet switching technologies for developing a wide choice of backbone network technologies. Network modeling and design software intended to help common customers base design and plan for networks ideally suited to their unique business requirements. Specialist in network product development in the enterprise marketplace, gives customers the ability to migrate from SNA to IP. Specialist in wide area network access products, gives telecommunications carriers a more cost-effective way to deliver high-speed data services for Internet and intranet access applications. Innovator of high-speed Synchronous Optical Networking/Synchronous Digital Hierarch technology to carry information to high-capacity backbone networks, such as those operated by telecommunications carriers and ISP’s. GISG is a pioneer in the Windows NT network security marketplace with its Windows NT Centri Firewall for small/medium businesses. Pioneer in designing combined communications support for compressed voice, LAN, data and video traffic across Frame Relay and ATM networks.

Granite Systems

September 1996

$220 million

Netsys Technologies

October 1996

$79 million

Metaplex, Inc.

December 1996

not public

Telesend

March 1997

not public

Skystone Systems Corp.

June 1997

$102 million

Global Internet Software Group

June 1997

$40 million

Ardent Communications Corp.

June 1997

$156 million

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 18

Exhibit 4 Preliminary Ardent Term Sheet, June 1996 ______________________________________________________________________________ Memorandum of Terms For Private Placement of Series A and Series B Preferred Stock of Ardent Communications Corporation June 14, 1996 ______________________________________________________________________________ This memorandum summarizes the principal terms of the Series A and Series B Preferred Stock financing of Ardent Communications Corporation. Offering Terms Issuer: Ardent Communications Corporation, a California corporation (the "company") 3,000,000 shares of Series A Preferred Stock and 11,000,000 shares of Series B Preferred Stock $.333 per share of Series A and $1.00 per share of Series B

Securities to be Issued:

Price:

Terms of Series A and Series B Preferred Stock Dividends: Annual $.03 and $.08 per share dividend, respectively, payable when and if declared by Board; dividends are not cumulative. For any other dividends or distributions, Preferred Stock participates with Common Stock on an as-converted basis. First pay cost plus accrued dividends on each share of Preferred Stock. Thereafter Preferred and Common share on as converted basis, until such time as the Preferred Stock has received an aggregate of two times cost, thereafter all proceeds shall go to the Common Stock. A merger, reorganization or other transaction in which control of the company is transferred will be treated as if a liquidation.

Liquidation Preference:

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 19

Exhibit 4 (cont’d) Preliminary Ardent Term Sheet, June 1996 Conversion: Convertible into one share of Common Stock (subject to antidilution adjustment) at any time at the option of the holder. Automatically converts into Common Stock upon consummation of underwritten public offering with a price of $5.00 and aggregate proceeds in excess of $7,500,000. Antidilution Adjustments: Conversion ratio adjusted on narrow weighted average basis in the event of a dilutive issuance. Proportional adjustments for stock splits and stock dividends. Votes on an as-converted basis, but also has series vote as provided by law and on (i) the creation of any senior or pari passu security, (ii) repurchase of Common Stock except upon termination of employment, (iii) any transaction in which control of the Company is transferred, and (iv) any adverse change to the rights, preferences and privileges of the Series A or Series B Preferred.

Voting Rights:

Terms of Preferred Stock Purchase Agreement Representations and Warranties: Standard representations and warranties by the Company. Assignment of Inventions and Confidentiality Agreement: Terms of Investor Rights Agreement Registration Rights: (a) Beginning earlier of June 28, 2000 or six months after initial registration, two demand registrations upon initiation by holders of at least 30% of outstanding Preferred Stock for aggregate proceeds in excess of $10,000,000. Expenses paid by Company. (b) Unlimited piggyback registration rights subject to pro rata cutback at the underwriter’s discretion. Full cutback upon IPO; 30% minimum inclusion thereafter. Expenses paid by Company. All employees and consultants shall enter into company’s standard form inventions and proprietary information agreement.

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 20

Exhibit 4 (cont’d) Preliminary Ardent Term Sheet, June 1996 (c) Unlimited S-3 Registrations of at least $1,000,000 each upon initiation by holders of 20% of the Preferred. Expenses paid by Company. Registration rights terminate (i) five years after initial public offering or (ii) when all shares can be sold under Rule 144, whichever occurs first. No future registration rights may be granted without consent of a majority of Investors unless subordinate to Investors’ rights. Right of First Refusal: Cisco Systems shall have the right to purchase all securities issued in subsequent equity financings of the Company, provided the Option, as defined below, has not expired. The Investors shall receive standard information rights including audited financial reports, quarterly unaudited financial reports, monthly unaudited financial reports and annual budget and business plan, as well as standard inspection rights. Board shall consist of four members. Board composition at Closing shall be Wu Fu Chen, Ed Kozel, and Mike Goguen. One other representative will be designated by a majority vote of the Series B Preferred Stock.

Financial Information:

Board of Directors:

Post-Closing Capitalization Series A Preferred Stock Outstanding Series B Preferred Stock Outstanding Common Stock held by Founders Common Stock Reserved for Employees (however, an additional 3,750,000 shares shall be available for grant after expiration of Option, held by Cisco): TOTAL : 3,000,000 9,000,000 6,250,000 3,000,000 shares shares shares shares 12.9% 47.3% 26.9% 12.9%

23,250,000

shares

100.0%

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 21

Exhibit 4 (cont’d) Preliminary Ardent Term Sheet, June 1996

Other Matters Common Stock Vesting: Common Stock shall vest as follows: After twelve months of employment, 25% will vest; the remainder will vest monthly over the following 36 months. Repurchase option on unvested shares at cost. No acceleration in the event of a Change of Control, except for Mr. Chen, whose vesting shall accelerate in the event of a Change of Control such that at most one year of vesting shall remain. (a) No transfers allowed prior to vesting. (b) Right of first refusal on vested shares until initial public offering. (c) No transfers or sales permitted during lockup period of up to 180 days required by underwriters in connection with stock offerings by the Company. Until the earlier of fifteen (15) months from the Closing or three (3) months after First Customer Shipment, Cisco shall have the right to acquire either all of the outstanding equity securities of the Company, or all of the Company’s assets, in Cisco’s discretion, for a purchase price of $232,500,000 payable either in cash or equity securities of Cisco. Closing subject to negotiation of definitive legal documents and completion of legal and financial due diligence by Investors.

Restrictions on Common Stock Transfers:

Option:

Closing Conditions:

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 22

Exhibit 5 Ardent Capitalization Table

Preferred A
Cisco Sequioa Capital Founders Engineering Team Total Valuation ($/shr) Valuation ($) Cash Inflow $ $ $ 3,000,000 3,000,000 0.33 2,400,000 999,000 $ $ $

Preferred B
7,535,000 2,465,000 1,000,000 11,000,000 1.00 23,250,000 11,000,000 $ $ $

Common
6,250,000 3,000,000 9,250,000 0.001 23,250,000 9,250

Total
7,535,000 2,465,000 10,250,000 3,000,000 23,250,000

Ownership
32% 11% 44% 13% 100%

Option to Acquire Acquisition Price Return Cisco Venture Capital Founders (5 employees) Engineering Team (20 employees) $ 232,500,000 Cash Out 75,350,000 24,650,000 102,500,000 30,000,000 Cash In 7,535,000 2,465,000 1,002,250 3,000 Multiple 10.0 10.0 102.3 9,978

$ $ $ $

$ $ $ $

Return (Cash in - cash out) Per Employee Founders $ Engineering Team $ Cisco Cost/Head Cisco Cost $ $

20,299,550 1,499,850 6,286,000 157,150,000

Conditions - No accelerated vesting for employees or founders, except for Wu Fu Chen - Commitment from Wu Fu to stay one year post acquisition - Right to future offerings in the company or direction of those offerings

Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15

p. 23

Exhibit 6 Press Release for the Ardent Communications Acquisition

CISCO SYSTEMS TO ACQUIRE ARDENT COMMUNICATIONS CORP. Further investment in Data, Voice and Video Integration For Public and Private Networks SAN JOSE, Calif. — June 24, 1997 — Cisco Systems, Inc. today announced it has signed a definitive agreement to acquire privately-held Ardent Communications Corp. Previously, Cisco and Sequoia Capital held minority equity stakes in Ardent. San Jose-based Ardent is a pioneer in designing combined communications support for compressed voice, LAN, data and video traffic across public and private Frame Relay and ATM networks. Under the terms of the acquisition agreement, shares of Cisco common stock worth approximately $156 million will be exchanged for the outstanding shares and options of Ardent. In connection with the acquisition, Cisco expects a one-time charge against after-tax earnings of 23 cents per share in the fourth fiscal quarter of 1997. The acquisition is expected to be completed by late-July 1997 subject to various closing conditions, including clearance under the Hart-Scott-Rodino Antitrust Improvements Act and Ardent shareholder approval. CISCO STEPS UP INTEGRATION OVER FRAME RELAY AND ATM NETWORKS With the continued pace of deregulation of the telecommunications service industry, carriers are increasingly offering services which integrate communication channels of voice, video, and data. As a result, the demand for low cost, easy to use, multiservice access products for new carrier services is rapidly expanding. The acquisition of Ardent will complement Cisco’s 3800 series within carrier service offerings for branch offices and remote sites by extending leadership in integration of voice, video and data. Based on Cisco IOS software, Ardent’s low cost platforms will natively support multiservice traffic and implement voice compression using high performance Digital Signal Processor (DSP) technology. Ardent’s early affiliation with Cisco has resulted in a complementary product platform offering superior interoperability with existing Cisco multiservice access and switching product lines. ABOUT ARDENT COMMUNICATIONS Ardent Communications was founded in 1996 by CEO Wu Fu Chen. Mr. Chen has co-founded four other companies since 1986 including Cascade Communications and Arris Networks. Ardent’s approximately 40 employees will remain in San Jose and become part of the Multiservice Access Business Unit led by Vice President and General Manager Alex Mendez within Cisco’s Service Provider line of business. Ardent Communications is on the leading edge of integrated access equipment design. Founded in 1996, Ardent designs, manufacturers and distributes advanced access products for integrating voice, video and data on public or private Frame Relay or ATM networks.