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 o
ptical 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 ge
t the right
product to market on time, should Cisco build its own external venture or just a
cquire
somebody outright?
NETWORKING OPPORTUNITY: PIPELINKS
For the previous two years, Cisco had been preaching about the promise of a mult
iservice
network a single network capable of transporting data, voice, and video. Cisco s
service provider customers agreed that network convergence would ultimately impr
ove cost
effectiveness and allow them to expand their service offerings. However, most se
rvice providers
were saddled with huge investments in their existing circuit-based voice network
s. This situation
implied a market need for optical (Sonet/SDH) routers that leveraged the existin
g infrastructure
while enabling a transition to multi-service networks: the market needed a produ
ct capable of
simultaneously transporting circuit-based traffic and routing IP (Internet Proto
col) traffic.1
Discussions with representatives from Cisco s Service Provider Line of Business
(SPLOB) indicated that developing optical routers internally was not a viable op
tion. A brief
search for potential acquisition targets had failed to identify any attractive c
ompanies none of
1 Sonet/SDH (synchronous optical network/synchronous digital hierarchy) is a pro
tocol for data transmission over
fiber optic lines.
Research Associates James McJunkin and Todd Reynders prepared this case under th
e 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 Marg
ot 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 94305-501
5. 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 th
e Stanford Graduate School of Business.
Version: (A) 02/25/00
GRADUATE SCHOOL OF BUSINESS
STANFORD UNIVERSITY
CASE NUMBER: EC-15
FEBRUARY 2000
CISCO SYSTEMS: A NOVEL APPROACH TO
STRUCTURING ENTREPRENEURIAL VENTURES
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, h
owever,
entrepreneur Amit Shah had recently approached Cisco with an idea for a Sonet/SD
H 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 contem
plated 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 struct
uring the
venture, but much had been done right. Volpi dug up the Ardent file and contempl
ated 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.
2
BACKGROUND ON CISCO SYSTEMS INC.
Cisco Systems was founded in 1984 by Leonard Bosack and Sandy Lerner, husband an
d wife
computer scientists at Stanford University who invented a technology to link the
ir disparate
computer systems together. Bosack and Lerner developed the first multi-protocol ro
uter 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 t
echnology
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, offer
ing high-end
routers. The devices were the traffic cops of cyberspace they directed network t
raffic to its
final destination via the most efficient, least congested network path. As the g
lobal 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 equipme
nt market
the plumbing of the Internet. By 1997, approximately 80% of the large scale router
s that
powered the Internet were made by Cisco. Although routers, LAN switches, and wid
e-area
network (WAN) switches would remain Cisco s core products, the company expanded it
s
product line to include a broad range of other networking solutions, including W
eb 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 th
e number one or
number two position in nearly every market in which it competed. In addition, Ci
sco s
Internetwork Operating System (IOS) software was increasingly becoming the de fa
cto 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 co
rporation with
over 10,000 employees in 54 countries. By 1997, revenues had increased over nine
ty-fold since
2 Excerpts taken from Cisco Systems, Inc. Acquisition Integration for Manufacturi
ng , Case # OIT-26, Graduate
School of Business Stanford University and Harvard Business School, revised Janu
ary 1999.
3 Cisco s IOS Software was the industry leading internetworking software, like Mic
rosoft Windows for networking.
IOS is a platform that delivers network services and enables networked applicati
ons. IOS enables interoperability
connections between otherwise disparate hardware, and accommodates network growt
h, change, and new
applications. It also contains security features, including access control, auth
entication, 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. (Exhi
bit 1) In June
1997, Cisco s market value totaled $46.3 billion.
Two highly respected CEOs have led the company: John Morgridge, and his successo
r, 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 presentat
ion on frugality to
the Cisco sales force, after being appalled by reports that salespeople were fly
ing first class on
business trips. Equipped with slippers, earplugs, and eye covers, Morgridge disp
layed 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-compe
titive nature.
Chambers, a former IBM and Wang Laboratories marketing and sales veteran, foster
ed 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 re
sult, more of
Cisco s customers sought end-to-end networking solutions. Building on its expertis
e in routers,
Cisco strove to deliver a wide range of new products, expand their offerings thr
ough internal and
external efforts, enhance customer support, and increase presence around the wor
ld.
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 specifica
lly 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 soluti
ons 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 adv
ertising.
The ongoing deregulation of telecommunications and technology convergence were d
riving 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 netw
orks for
transmitting data, and broadcast networks for transmitting video but advances in
digitization
allowed these forms of communication to be translated into binary computer langu
age. 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 compan
ies 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 cr
eate
exciting opportunities for Cisco to capture share in the $250 billion telecom eq
uipment market,
which was previously dominated by huge, well-capitalized companies such as Lucen
t
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 i
t 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 marke
t became
increasingly important, Chambers realized that Cisco could not keep up with the
changing
market needs solely through internal development. Using acquisitions and allianc
es to gain
access to world-class technologies and people became a defining component of Cis
co s strategy.
This strategy was relatively unique in the high-tech world at the time many comp
anies viewed
looking to the outside for technological help as a sign of weakness. Chambers co
mmented 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 drivi
ng force
behind Cisco s business development strategy, many in the industry regarded Mike V
olpi 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 Busin
ess, Cisco had
completed only one acquisition, Crescendo Communications. Two more acquisitions
closed
soon after Volpi arrived, but he was deeply involved in all subsequent acquisiti
ons.
Before pursuing a new market opportunity, Volpi s group assessed the merits and do
wnsides 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 develop
ment group would
often opt to acquire or partner with an external player. Although the acquisitio
ns 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 The Art of the Deal , Business 2.0, October 1999.
5 Volpi initially reported to Charles Giancarlo, who joined Cisco through the Ka
lpana 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
OEM/ License
Joint R&D/ Marketing/Sales
Invest
Acquire
Degree of
Strategic
Value
Level of Commitment
The public equity markets were the principal exit strategy for hot high-technolo
gy 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 a
t 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 ro
ughly $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 help
ed us in
subsequent transactions.
Cisco made predominantly small acquisitions of private companies, with typical dea
l 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 la
rge, established
companies with strong corporate cultures presented greater integration challenge
s. Further,
Chambers asserted that Cisco did not acquire to gain short term market share, bu
t 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 199
7, 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 ho
le 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 opp
ortunities in
virtually every sector of networking. Cisco s unique vantage point allowed the com
pany 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 c
ompany s localarea
network to the wide-area network.
Cisco identified two principal customer needs. The first need was to simplify an
d improve
management of network access equipment. The conventional approach to public netw
ork 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 believe
d that these
solutions would remain expensive especially in comparison to LAN bandwidth where
Ethernet technology10 dominated. The company expected that high speed network ac
cess
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 aggre
gated 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 Cisco s Secret: Entrepreneurs Sell Out, Stay Put , Inc. Magazine, March 1997.
9 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, s
uch as voice, video, or data, are
conveyed in small, fixed-size cells.
10 In 1996, Ethernet technology penetrated every corner of the Enterprise networ
k 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 c
onsidered how
to address the market that Ardent Communications would eventually serve. Buildin
g 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 thi
ngs on a day-today
basis, but did not have idle human resources to devote to the new project. Diver
ting
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 mar
ket a
priority. The business development team concluded that Cisco had neither the tim
e 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 thi
s 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 ac
quisition target
would be a difficult task in all cases Cisco would have to spend lots of time an
d effort
modifying the product set and integrating the newly acquired company into the Ci
sco
organization. In addition, retaining key employees post-acquisition was always a
challenge.
After considering the buy and build alternatives and finding neither to be satis
factory, Volpi
mused, Why not custom make a start-up to build exactly the product we want, and t
hen buy
them later if they succeed? In essence, this solution would entail the creation o
f a new venture
designed as a spin-in from day one build to buy, a hybrid of the buy and build app
roaches.
At first glance, the spin-in model seemed to address three key issues: time-to-m
arket, 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 diff
icult tradeoffs. What
sort of structure would allow the start-up to leverage Cisco s strategic assets wi
thout 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 own
ership stake should
Cisco take in the venture? Incentives would also be a major issue: How could Cis
co 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 succe
ss.
Structuring the Ardent Communications Venture
To develop a potential model for the new venture approach in 1996, Volpi and Koz
el 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 fro
m the
Cisco employees, who intended to leave their jobs at Cisco to build a solution w
hich they hoped
their employer would want to acquire. The product concept had a great deal of po
tential, and the
founding team was flush with engineering talent. Wu Fu Chen had an especially im
pressive
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 Cha
mbers
believed that funding Cisco employees to go out and build new networking compani
es 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 c
all 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 simpl
y, Make this
product and we ll give you lots of money. After a series of discussions, Wu Fu agre
ed 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 spe
cification for the
first Ardent product, tentatively called Ardent 101. For Cisco to buy the new co
mpany, 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 mi
lestones that
would set expectations for the product timeline. (Figure 3)
Figure 2: Ardent 101 Functional Requirements
1.
Ability to accept data, voice, and video traffic
2.
Aggregate up to 2Mbps traffic on the WAN side
3.
Support ATM, Frame Relay, and TDM trunks
4.
Support standard office environments
5.
Support Bridging, IP Routing for LAN data traffic
6.
Support Circuit Emulation for Voice and Video Applications
7.
Support voice and data compression
8.
The target list price for the base configuration is about $5,000; Cost of
goods target is $800 or lower
9.
Will consider support of data encryption
10.
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
2.
Twelve months after the Effective Date of the Agreement, the Company shall
have begun integration of the Product
3.
Fifteen months after Effective Date of the Agreement, the Company shall have
begun the beta program for the Product
4.
Eighteen months after the Effective Date of the Agreement, First Customer
Shipment shall have occurred
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 fo
r 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 financi
ng since this would
create more of a start-up feel. To foster an entrepreneurial environment with st
rong employee
incentives, Cisco agreed to give the founding team and employees a large ownersh
ip 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 reque
sting in their
spin-ins and spin-outs (arguing that their intellectual property, brand name, an
d 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 rig
ht 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 simu
ltaneously,
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 P
referred 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 ro
und. 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 fir
st 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 milli
on shares of
Series B Preferred Stock on September 20. Seven days later, the founders purchas
ed another one
million shares. The remaining equity capitalization consisted of 9.25 million sh
ares 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 mil
lion. 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 Arden
t 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 ves
ting
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-spec
ified price, but
would also obligate Cisco to purchase the company if the new team succeeded in b
uilding 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 in
to a strategic
investment. The model intrigued both John Chambers and Ed Kozel, who viewed it a
s 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 outst
anding
equity securities of the Company, or all of the Company s assets, in Cisco s discret
ion,
for a purchase price of $232,500,000, payable either in cash or equity securitie
s of
Cisco.11
Since Cisco would also write a put option, the shareholders in the new venture c
ould 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 s
ame 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 s
ell its
Securities to Cisco, and Cisco shall be obliged to and will purchase such Securi
ties, in
accordance with the purchase price and other terms of purchase 12
Cisco believed that although the put/call structure truncated the upside for inv
estors and
employees, it mitigated enough risk to make the investment or employment decisio
n attractive
from a risk/reward standpoint. The option agreement turned out to be a very effe
ctive recruiting
instrument. If the product requirements and milestones were met, the 15-20 perso
n engineering
team would share a $30 million payout less than 15 months out. The five person f
ounding team
would do even better: delivering on the product would allow them to share more t
han $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 C
isco 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 ad
ding the
technologies of ATM and Frame Relay over a T1/E1 connection, circuit emulation f
or digitized
voice over ATM or Frame Relay, voice compression, and telephony capabilities. Th
ese changes
were not on the official evolutionary path of IOS, though they were similar to s
ome 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 a
s 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 Pref
erred 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 v
alue-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 te
am throughout
the development process. This was not unusual, because Cisco had provided consul
ting services
from time to time in the past. Ardent paid the standard fee for these engineerin
g resources,
$250,000 per engineer per year. Since this was regarded as a temporary assignmen
t, 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 unti
l 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 ge
nerally 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 pr
oduct
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 appropr
iate because
the product specification had already been narrowly defined, minimizing the degr
ees 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 Ci
sco product
line, although this was not a stated requirement in the product specification. I
nstead of designing
a tailored box for Ardent 101, Wu Fu decided to adhere to the existing Cisco pro
duct line and
imposed an additional requirement on his engineers: they would adopt the form fa
ctor for the
Cisco 2500 series. This solution was compact and familiar to Cisco s carrier custo
mers.
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 pa
id was far in
excess of the assets book value, the pooling method was generally preferable.13 U
sing 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 g
oodwill, an intangible asset which
represents the excess of the purchase price over the assets book value, must be a
mortized 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 pri
or 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 str
ategic 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 issue
s should ever
dominate the strategic issues in making decisions.
The accounting for Cisco s investment in Ardent was relatively straightforward. Si
nce 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 a
t
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 it
s option to
purchase the outstanding shares of Ardent. In many ways, the decision was immate
rial, because
Wu Fu and his team were likely to meet the acceptance criteria and exercise thei
r 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 i
t better to do it
now? Volpi and Kozel determined that doing the deal sooner rather than later wou
ld 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 ma
rketing
tact. If Ardent had the opportunity to educate the marketplace, Cisco would eith
er be forced to
continue the same marketing program or re-educate potential customers. Historica
lly, 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 s
tated: Under
the terms of the acquisition agreement, shares of Cisco common stock worth appro
ximately $156
million will be exchanged for the outstanding shares and options of Ardent. (Exhi
bit 6) This
was consistent with the agreed upon total acquisition price of $232.5 million, s
ince 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 inve
stment.
Sequoia Capital received a payout of $24.6 million, a relatively small number bu
t 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 acqu
isition 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 tran
sporting
circuit-based traffic and routing IP (Internet Protocol) traffic had been identi
fied 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 servic
e provider
customers who were struggling to bring their networks into the New World of unifie
d
networks, but wanted to make the transition without scrapping their existing cir
cuit-based TDM
infrastructure.
One of the factors Volpi considered was Cisco's expertise in the domain of optic
al 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 ran
ge) than Cisco s
traditional products. Because this was a new and unfamiliar market for Cisco, Vo
lpi and his
team looked externally to pursue the market opportunity, but none of the existin
g players met
Cisco's criteria.
Volpi's team believed, however, that an idea that successful entrepreneur Amit S
hah had
approached Cisco with might address this market need. Shah, who had sold his fir
st 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 t
he access points
of the Internet infrastructure the metro space near large population areas. Shah c
onceived 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 immed
iately 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) success
fully 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
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
$
$
$
$
1990
35,842
21,102
15,874
3,701
1,673
78,192
4,114
-
367
82,673
4,973
6,290
1,976
13,239
123
13,469
-
69,204
82,673
1990
$
$
$
$
1991
40,323
51,104
34,659
6,078
8,797
140,961
12,665
-
519
154,145
7,743
17,965
542
26,250
436
26,686
-
127,459
154,145
1991
$
$
$
$
1992
39,955
116,477
61,258
9,142
20,244
247,076
28,017
46,866
1,974
323,933
16,262
46,953
15,108
78,323
-
78,323
-
245,610
323,933
1992
$
$
$
$
1993
27,247
61,738
129,109
23,500
26,702
268,296
48,672
274,260
3,985
595,213
24,744
77,492
17,796
120,032
-
120,032
-
475,181
595,213
1993
$
$
$
$
1994
53,567
129,219
237,570
27,896
59,425
507,677
77,449
457,394
11,174
1,053,694
31,708
130,846
42,958
205,512
-
205,512
-
848,182
1,053,694
1994
$
$
$
$
1995
284,388
279,754
421,747
81,805
116,466
1,184,160
172,561
583,871
51,357
1,991,949
59,812
257,099
71,970
388,881
-
388,881
40,792
1,562,276
1,991,949
1995
$
$
$
$
1996
279,695
758,489
622,859
301,188
197,409
2,159,640
331,315
1,060,758
78,519
3,630,232
153,683
445,776
169,894
769,353
-
769,353
41,257
2,819,622
3,630,232
1996
$
$
$
$
1997269,608
1,005,977
1,170,401
254,677
400,603
3,101,266
466,352
1,630,390
253,976
5,451,984
207,178
656,707
256,224
1,120,109
-
1,120,109
42,253
4,289,622
5,451,984
1997
$
$
69,776
23,957
45,819
6,168
18,260
21,391
2,088
-
23,479
9,575
13,904
$
$
183,184
62,499
120,685
12,687
41,809
66,189
4,567
-
70,756
27,567
43,189
$
$
339,623
111,243
228,380
26,745
72,248
129,387
6,719
-
136,106
51,720
84,386
$
$
649,035
210,528
438,507
44,254
130,682
263,571
11,557
-
275,128
103,173
171,955
$
$
1,334,436
450,591
883,845
106,680
276,995
500,170
22,330
-
522,500
199,519
322,981
$
$
2,232,652
742,860
1,489,792
306,575
485,254
697,963
40,014
-
737,977
281,488
456,489
$
$
4,096,007
1,409,862
2,686,145
399,291
886,048
1,400,806
64,019
-
1,464,825
551,501
913,324
$
$
6,452,000
2,243,000
4,209,000
1,210,000
1,370,000
1,629,000
262,000
-
1,891,000
840,000
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

$$
2.00
$4.00
$6.00
$8.00
$10.00
$12.00
$14.00
$16.00
$18.00
Feb-90Aug-90Feb-91Aug-91Feb-92Aug-92Feb-93Aug-93Feb-94Aug-94Feb-95Aug-95Feb-96Au
g-96Feb-97
14 Prices adjusted for all splits since IPO, based on January 25, 2000 stock pri
ce.
Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15 p.
16
Exhibit 3
Summary of Cisco s Acquisitions as of June 1997

Company Date Purchase Price Description

Crescendo
Communications, Inc.
September 1993 $95 million High-performance work group CDDI
and FDDI switching solutions.
Newport Systems
Solutions, Inc.
July 1994 $93 million Software-based routers for remote
network sites of small/medium sized
networks.
Kalpana, Inc. October 1994 $240 million Manufacturer of modular and stackable
LAN switching products extend the
usability and data capacity of existing
Ethernet LAN s.
LightStream Corp. October 1994 $120 million 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.
Combinet, Inc. August 1995 $132 million Supplier of ISDN (Integrated Services
Digital Network) remote-access
networking products useful for
telecommuting and other networked
applications.
Internet Junction, Inc. September 1995 not public Developer of Internet gateway
software
connecting central and remote office
desktop users with the Internet.
Grand Junction, Inc. September 1995 $400 million Inventor and leading supplier o
f Fast
Ethernet (100BaseT) and Ethernet
desktop switching products.
Network Translation,
Inc.
October 1995 not public Manufacturer of cost-effective, low
maintenance network address
translation and enterprise Internet
firewall hardware and software.
TGV Software, Inc. January 1996 $138 million Internet software products for
connecting disparate computer systems
over local area, enterprise-wide and
global computing networks including
the Internet.
StrataCom, Inc. April 1996 $4.666 million 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.
Telebit Corp s MICA
Technologies
July 1996 $200 million Modem ISDN Channel Aggregation
(MICA) technologies will deliver highdensity
digital modem technology with
Cisco s dial-up and access product
lines.
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 Date Purchase Price Description

Nashoba Networks, Inc. August 1996 $100 million Token ring switching technologie
s
for providing users with a wide
choice of employing highperformance
switched workgroup
and backbone Token Ring
environments.
Granite Systems September 1996 $220 million Standards-based multilayer Gigabit
Ethernet switching technologies
for developing a wide choice of
backbone network technologies.
Netsys Technologies October 1996 $79 million Network modeling and design
software intended to help common
customers base design and plan for
networks ideally suited to their
unique business requirements.
Metaplex, Inc. December 1996 not public Specialist in network product
development in the enterprise
marketplace, gives customers the
ability to migrate from SNA to IP.
Telesend March 1997 not public 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.
Skystone Systems Corp. June 1997 $102 million 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.
Global Internet Software
Group
June 1997 $40 million GISG is a pioneer in the Windows
NT network security marketplace
with its Windows NT Centri
Firewall for small/medium
businesses.
Ardent Communications
Corp.
June 1997 $156 million Pioneer in designing combined
communications support for
compressed voice, LAN, data and
video traffic across Frame Relay
and ATM networks.
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 Pref
erred Stock
financing of Ardent Communications Corporation.
Offering Terms
Issuer:
Ardent Communications Corporation, a California
corporation
(the "company")
Securities to be Issued:
3,000,000 shares of Series A Preferred Stock and
11,000,000 shares of Series B Preferred Stock
Price:
$.333 per share of Series A and
$1.00 per share of Series B
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.
Liquidation Preference:
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.
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.
Voting Rights:
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.
Terms of Preferred Stock
Purchase Agreement
Representations and Warranties:
Standard representations and warranties by the Company.
Assignment of Inventions and
All employees and consultants shall enter into company s
Confidentiality Agreement:
standard form inventions and proprietary information
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.
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.
Financial Information: 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 of Directors: 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.
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 shares 12.9%
9,000,000 shares 47.3%
6,250,000 shares 26.9%
3,000,000 shares 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.
Restrictions on Common Stock Transfers: (a) No transfers allowed prior to vestin
g.
(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.
Option: 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 $232,500,000
payable either in cash or equity securities of Cisco.
Closing Conditions: Closing subject to negotiation of definitive legal
documents and completion of legal and financial
due diligence by Investors.
Cisco Systems: A Novel Approach to Structuring Entrepreneurial Ventures EC-15 p.
22
Exhibit 5
Ardent Capitalization Table
CiscoSequioa Capital
Founders
Engineering Team
Total
Preferred A
-
3,000,000
-
3,000,000
Preferred B
7,535,000
2,465,000
1,000,000
-
11,000,000
Common
-
-
6,250,000
3,000,000
9,250,000
Total
7,535,000
2,465,000
10,250,000
3,000,000
23,250,000
Ownership
32%
11%
44%
13%
100%
Valuation ($/shr)
Valuation ($)
Cash Inflow
0.33 $ 1.00 $ 0.001 $
2,400,000 $ 23,250,000 $ 23,250,000 $
999,000 $ 11,000,000 $ 9,250 $
Option to Acquire
Acquisition Price $ 232,500,000
Return
Cisco
Venture Capital
Founders (5 employees)
Engineering Team (20 employees)
Cash Out Cash In Multiple
75,350,000 $ 7,535,000 $ 10.0
24,650,000 $ 2,465,000 $ 10.0
102,500,000 $ 1,002,250 $ 102.3
30,000,000 $ 3,000 $ 9,978
Return (Cash in - cash out) Per Employee
Founders $
Engineering Team $
20,299,550
1,499,850
Cisco Cost/Head
Cisco Cost
$
$
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 N
etworks
SAN JOSE, Calif. June 24, 1997 Cisco Systems, Inc. today announced it has signed
a
definitive agreement to acquire privately-held Ardent Communications Corp. Previ
ously, 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 vi
deo 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 opti
ons of Ardent.
In connection with the acquisition, Cisco expects a one-time charge against afte
r-tax earnings of
23 cents per share in the fourth fiscal quarter of 1997. The acquisition is expe
cted to be
completed by late-July 1997 subject to various closing conditions, including cle
arance 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 indust
ry, 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 3
800 series
within carrier service offerings for branch offices and remote sites by extendin
g leadership in
integration of voice, video and data. Based on Cisco IOS software, Ardent s low co
st 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-fou
nded
four other companies since 1986 including Cascade Communications and Arris Netwo
rks.
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 desi
gn. 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.