June 14, 1999
Web Service Numbers Just Don't Add Up
By Bob O'Donnell
Compared to radio, television and other broadcast media, the Internet
and World Wide Web, in particular, is supposed to be a marketer's/seller's
dream come true. In fact, even though the Internet is still in its
infancy, it's viewed by many in these areas as being the ideal media for a
number of different reasons.
In addition to the greatly-hyped notion of one-to-one marketing, the
'net offers the promise of incredibly precise counts of
viewers/visitors/users/buyers or whatever term is most appropriate for a
given web site. Given the sophisticated web reporting tools now available,
in fact, you can be easily led to believe that it isn't unrealistic to get
numbers that are accurate to the person. Compared to the sampled
approximations used in the other media, this is an enormous improvement.
But despite this supposed precision, I have a really hard time
believing the numbers that many web sites tout as their customer base.
Ironically, my instincts tell me that they're even less believable and
much less reliable than the surveyed numbers used for things such as
television ratings.
For example, when I see a new company offering a web site with web page
hosting or free e-mail or any other service that claims they have millions
of users, I just shake my head. Where could they possibly be getting these
numbers from?
According to some research I did, the total number of Internet users
worldwide is estimated to be between 150 and 170 million total. Of those,
a little less than 2/3 come from the US and another decent percentage of
the remaining third come from English-speaking countries (important only
because most of the web site services now online are only available in
English, which presents a language barrier to non-English speaking web
surfers). Of course, it's not clear that even these numbers represent
unique individuals-I suspect a lot of double counting for people who use
the web at work and at home. But just for sake of argument, let's say
there's about 120 million potential customers (that is, unique individuals
who can read and understand the content of a particular site) for various
web site services. For any web site to get even four million customers is
saying that about 3.5% of all the English-speaking web surfers in the
world are customers of their site. Now, on the one hand, that may not
sound like a big deal, but if you really think about it, you'll realize
that's ridiculous. For any web site that isn't spending tens of millions
of dollars on marketing to get their name out there, reaching this level
of market penetration would be almost impossible.
And it gets worse if you think about some of these numbers in
aggregate. If you actually started combining all the "customer"
counts of the different web site services, you'd start to realize that
virtually every person on the web would need to be using several of these
services for the numbers to add up. Call me crazy, but I don't think the
typical web surfer has five different free e-mail accounts, three
different free home pages, and subscriptions to and/or accounts with
several other sites.
In fact, I have yet to find a single web site service that I have felt
is really worth my time and effort (let alone my money!). Sure, I've
checked out quite a few and even signed up for and tried a couple, but I
have found nothing compelling enough to keep using on a regular basis.
And this brings me to around to my original point, which I guess boils
down to the equation that precision does not equal truth. While I'm sure
many of these web service-type sites receive millions of visits from web
surfers-which is where they derive their numbers from, my guess is that
regular customer counts are several orders of magnitude less than what
they claim.
I think the potential for providing valuable services on the web is
very real, but the overenthusiastic claims of many current web operations
are far from it.
©
Copyright 1999, by InfoWorld Publishing Corp., a
subsidiary of IDG Communications, Inc. Reprinted from InfoWorld,
155 Bovet Road, San Mateo, CA 94402. Further reproduction is
prohibited.
