Weekly Start-up
Ella Jacoby
Company Name: Charlotte's Web
Object: Internet router for telephony companies
There's a very popular children's story in the US
called "Charlotte's Web". Eyal Dagan, general
manager of the company with the same name,
says it is quite allegorical.
"We also spin a spider's web in setting up the
Internet network," says Dagan, "The parable
could be taken further, but we'll leave it at that."
The name, conjured up by one of the founders
residing in the US for many years, works. Dagan
reports that, at the last SuperCom exhibition held
in Atlanta, there were many American faces that
smiled at the name.
Business Card
Name: Charlotte's Web
Founded: March 1998
Product: Internet router designed for telephony
companies.
Employees: 15
Market: Telecom companies and large Internet
service providers.
Customers: None
Competition: Cisco, Lucent (after acquiring
Nexabit), Juniper, Pluris, Avici.
Ownership: Founders, employees and MRV
Charlotte's Web is likely to prove a legend on the
local scene, but it is far from child's play. This is
how it began: Dagan and his friend Eli Stein who
worked together at NBase (an Israeli operation of
MRV set up by three Israelis that develops
communications products and is traded on
NASDAQ) sat down together one day, about
eighteen months ago, and decided to do
something and form their own start-up to make
the strongest Internet router in the world. MRV
did not want to lose its developers. They
deliberated and decided that if they couldn't beat
them - they would join them. Dagan and Stein
departed from the company armed with enough
cash to enable them to develop the product right
up to the beta stage.
Dagan and Stein spent four to five months
interviewing scores of people. They recruited Ilan
Shimoni from 3Com, Eyal Evrech from Intel and
Gidi Kempfer, from the academic world. Dagan
had two modest requirements: the candidate
should make a good impression and excel in his
studies. Although the initiative began with Dagan
and Stein, the entire team is regarded as the
company's founders in almost every aspect.
After the team was assembled, they discussed
Charlotte's Web's mode of operation. "We asked
how we could make the highest, finest and
strongest tower," says Dagan. The tower is a
router at the center of the Internet, a sort of
refrigerator the size of a minibar, in effect a large
tap or valve, which directs cyberspace traffic.
How does one make routers stronger? Charlotte's
Web designed seven chips for installation on
hardware already existing in the market, and
software with sophisticated algorithms.
To understand how strong these routers are,
Dagan explains that four to five of Charlotte's
Web's routers are capable of networking the
entire State of Israel - every man, woman and
child in the Holy Land - if they are all hooked up
at 2 Mbps. And the price, accordingly, is $1.5-1.9
million each.
Charlotte's Web does not have a unique product.
The company is one of an admired elite in the
market, but Dagan stresses that his company is
"like a Formula 1 racing car". The market's major
player is without doubt Cisco. "When we noticed
that Cisco controls almost 100% of the market,
we could see there was some potential," says
Dagan. "When you see someone controlling the
market, it makes you feel uneasy. We decided
we could do better. It turns out that we were not
the only ones to think so, and other companies
thought they could do better too."
These other companies are Nexabit, sold to
Lucent a month ago for $900 million, start-up
company Pluris; and another start-up company
called Avici. Nortel, which has a minority stake in
Avici, recently announced it was withdrawing its
representative on the board of directors,
apparently in order to make the company more
attractive vis a vis competitors. Another company,
Juniper, recently had an IPO and climbed to an
Everest-like value of $6.5 billion.
"Globes": Aren't you a little late in entering
the market?
Dagan: "At the SuperCom exhibition, we
presented the world's strongest router. We didn't
think we would reach that stage. We will launch
the beta model by year end, and we hope to start
selling in the first quarter of the year 2000. Apart
from Juniper, no-one is actually selling yet. All
the competitors have more or less the same
timetable, and if we go on the market two months
later, it won't be disastrous. There are few
companies in this market. The real challenge is
to produce a working product, and whoever
succeeds will be a player in the market."
Published by Israel's Business Arena July 12,
1999 |
