Commentary 4Mobility - Cometa Wholesaling Wi-Fi Access
By Andrew M. Seybold <andy@outlook4mobility.com>
9 December 2002
Cometa Wholesaling Wi-Fi Access
Wi-Fi hotspots are becoming the next gold rush for the computer and
telecommunications industry. But is that gold in the hills or just fool's gold? Last week
three big players, AT&T, IBM and Intel, along with two investment companies formed
a new company called Cometa. Their plan is simple: Deploy over 20,000 Wi-Fi
hotspots in the 50 largest cities and wholesale wireless data to whoever wants to buy
it and resell it.
Since no terrestrial network has ever made money offering data-only services, what
do these folks know that the rest of the industry does not?
Okay, let's see if I have this correct. Intel, IBM, AT&T (not wireless), Apax Partners and
3i have formed a new company called Cometa that is going to put into service and then
wholesale Wi-Fi access in the top fifty U.S. cities so that Wi-Fi access is only a
five-minute walk away from any city dweller and a 5-minute drive away from a
suburbanite. With 20,000 Wi-Fi access points?
Let's play with some math here. Each access point today can cover a roughly circular
area out to about 300 feet from the center. That works out to about 283,000 square
feet per access point. 20,000 of these access points will provide about 203 square miles
of Wi-Fi coverage spread out over the 50 largest U.S. cities or 4 square miles of
coverage per city.
To put this in perspective, a typical 850-MHz urban cell site covers a radius of three
miles or 28 square miles. This means that Cometa plans to provide coverage equivalent
to less than one-fifth of a cell site per city. Access points will be scattered around the
cities in locations where groups of people want and need high-speed wireless access.
The announcement didn't mention anything about airports so I assume it will leave
airports to others and T-Mobile already provides access at Starbucks coffee shops. That
still leaves plenty of territory for Cometa. But it will probably run out of access points if
it unwires a single university in each city. So I have to ask what the big deal is.
There are ten or so hotspot providers today with more than 4,000 hotspots located in
more than 300 cities in 43 states. Adding another 20,000 is a significant undertaking
but I don't see anything in Cometa's announced plans about aggregating access to
hotspots already in place. If this is the case, Cometa will become one of ten or more
players that wide-area wireless operators will have to deal with when putting together
their own wide-area/Wi-Fi offerings.
Cometa admits that the backhaul is one of the most costly parts of putting together a
Wi-Fi hotspot system. But since AT&T (the long distance company) is part of this new
company it will be able to make use of AT&T's existing high-speed backhaul and thereby
save a lot of money. What this means to me is that AT&T will underwrite the cost of
the backhaul and many of the hotspots will be located where AT&T already has wired
high-speed connectivity, not necessarily where there is demand. But until we see how
and where Cometa rolls out its own hotspots it won't be possible to fully understand its
business model.
There are two established types of hotspot users: Those who travel and want and need
access back to their corporate information (using some type of secure Virtual Private
Network or VPN) and those who are local to a hotspot location and visit a Starbucks or
other caf in order to take advantage of higher-speed access than they have at home.
These two groups have very different usage patterns. Local folks want to sit in a caf
sipping coffee for several hours while surfing the Internet while business travelers, for
the most part, want to set up their system, get to their corporate LAN, download their
email, check their calendar, perhaps retrieve a file and get off the system.
Cometa appears to be aiming at both of these groups and we will have to wait to see
whether it can be successful as a wholesale Wi-Fi service provider. With today's business
models it is difficult to figure how a wholesale company can make money. On its Web
site, Boingo states that hotspot operators get $1.00 per connection through Boingo. If
they sign up Boingo users they receive an additional finder's fee. At $1.00 per
connection how can a hotspot provider that doesn't have high-speed backhaul in place
for some other purpose stay in business?
One keys to Cometa's success or lack thereof will be how cost effectively it can set up
this business and how lean it can run. The company will be located in two locations, one
on the East coast and one on the West. Even if AT&T gives away its backhaul (which I
doubt), I can't see how Cometa can make a profit. None of today's hotspot providers are
making any money there have already been a number of failures (T-Mobile bought
MobileStar at pennies on the dollar). I have to wonder what secret sauce Comenta has
that the others don't.
Yes, the demand for Wi-Fi access is growing and yes, wide-area operators are trying to
figure out their own play in this space. But the number of mobile users who are willing
to camp out at a location, boot up their notebook computers, log onto the provider for
that location and then use the service (at a reasonable price) is a small fraction of the
notebook user population. I believe that the only way to justify the cost of an extensive
public Wi-Fi hotspot network is if there is a value proposition beyond the pay-for-usage
model. The benefits realized by the company that bills the customer for Wi-Fi service
must go beyond the fees they receive from customers. Intangible benefits include
keeping a wide-area voice-and-data customer from jumping ship to another network
with a combination wide-area and hotspot system, adding revenue per user (ARPU) that
can be quantified or perhaps even selling more coffee.
I believe in hotspots and I use them often (but only the ones with reasonable pricing). I
also use my wide-area wireless data PC Card often. I do want a combination Wi-Fi and
wide-area wireless data service but I want to receive a single, reasonable bill. Cometa's
idea of becoming a Wi-Fi wholesaler is interesting and if it were the only company a
wide-area wireless network had to deal with in order to obtain a large enough Wi-Fi
footprint I might believe that it had broken the code.
Someone at Intel or IBM or AT&T or all three must believe that there is money to be
made as a Wi-Fi hotspot wholesaler. I welcome any of them or the folks at Comenta to
share their business model with me and prove to me that it can make money as a
standalone business for the partners that have invested in it.
In the meantime, the Wireless LANd grab continues. Wide-area wireless companies are
trying to figure out how to put together the best locations at the best pricing for their
own customers. They are trying to figure out how to charge for both wide-area and
Wi-Fi data access because they know there is a demand. But it appears to me that Wi-Fi
hotspots may become similar to the dot-com frenzy: lots of folks believing all of the
hype that surrounds this new, hot set of services getting into it without filling in the
blanks on a spreadsheet.
Those who move ahead because there is a business case, those who can prove to their
investors that they can make money here and those who have stepped back and looked
at the total size of the market and how much of that TAM they can capture should be
moving full speed ahead. Those who are jumping into the Wi-Fi hotspot market because
the industry is looking for the next big thing and Wi-Fi is it had better take a few days
off, walk around the block or on the beach, and then sit down with their spreadsheets
and figure out how they are going to make money at the Wi-Fi game!
Andrew M. Seybold
The Outlook 4Mobility provides its news summaries and analyses free of charge. Outlook
4Mobility products and services include Consulting Services, Mobiltorials, Newsletters,
Customized Proprietary Research, Wireless Tutorials and Conferences. Please visit our
web site at www.outlook4mobility.com for additional information.
Copyright 2002 Outlook 4Mobility |
