[Previous] [Next] - [Index] [Thread Index] - [Previous in Thread] [Next in Thread]

Subject: Re: UKNM: Web Usage Stats
From: Robb Masters
Date: Tue, 25 Aug 1998 14:36:03 +0100

Sean Clark said:
> A while ago I remember people asking about up-to-date figures for Web
> browser and computer platform usage. Well, I've added a section to the
> Headland home page called "WebWatch" that supplies this information based
> on visits to 10 of our most popular sites.

While it's admirable of Sean to publish these figures, and I for
one am always interested in such statistics, I can't help pointing
out that to infer anything about what users in general use from
such information is inherently flawed.

Let's say for example a site is built using technology X and
technology X is only supported by browser Y. Then it is likely that
virtually all accesses to the site will be from users with browser
Y. This would lead some to assume that the majority of the
*potential* audience have browser Y. The truth may be that only a
very small minority have browser Y but users of all other browsers
gave up when the home page didn't work for them.

This is obviously an over-simplification and over-exaggeration, and
I wouldn't claim that Headland's sites have done anything so foolish.
However, the sites I glanced at do provided limited support for
text-only users - such as users of legacy systems and the visually

This isn't meant to be a criticism of Headland, by any means - very
many, if not most, commercial sites are at least as "guilty".

But before I put my techie hat on and say how easy it is to create
a site that has all the "bells and whistles" but can still support
the "lowest common denominator" users, I'll remind myself that this
is a marketing list and say just two things:

* Take all browser statistics with a large dose of salt - unless you
know they come from sites that will appeal equally to all users and
which provide appropriate support for all browsers (quite unlikely)

* Exclusion of part of your potential (and desirable) audience is
generally a bad idea. To use Headland's site statistics for a
moment, 5% of users aren't using the top four browsers, yet many
sites (not theirs) seem to assume that everyone is (or even worse),
and alienating 5% of your potential Internet user base strikes me
as a poor marketing policy.

Robb Masters rmastersatredkite [dot] com
Project Manager

Red Kite New Media http://www.redkite.com/

  Re: UKNM: Web Usage Stats, Daniel Bambach

[Previous] [Next] - [Index] [Thread Index] - [Next in Thread] [Previous in Thread]