[Sigia-l] ROI/Value of Search Engine Design - Resources?
Richard Wiggins
rich at richardwiggins.com
Tue Feb 18 12:16:02 EST 2003
Of COURSE whether the customer arrives seeking a known item is highly
relevant to the question of whether she searches first. The Thomas Friedman
Duct Tape column and the Linksys Wifi are known item searches. The content
of today's NY Times obviously are not. You could even distinguish between
known generally (Linksys Wifi) versus known specifically (Linksys WRT54G);
if I've got a model number, I'm definitely headed for the search box first.
The whole debate about what the percentages are is something of a
distraction. We've identified lots of reasons why the percentages vary:
nature of site, nature of user, what user seeks today (known item versus
not), quality of site search engine, quality of site categories/links,
intrusion of graphics, hard to decipher links, whether some event has
occurred (e.g. Sapphire worm, new product announcement), etc. etc. ad
nauseam.
The point is that a certain percent of your customers will do searches. If
you're Amazon and you have a million customers a day, and let's say "only"
20% of your customers search, that's 200,000 customers to keep happy.
You can't satisfy all searchers, but if you understand the Zipf curve and
the Best Bets idea, you can satisfy at least half with close-to-perfect
precision.
/rich (no LIS degree but I'm married to someone with one...)
On Tue, 18 Feb 2003 09:04:45 -0800 (PST), "Ken Bryson" wrote:
>
>
> >
> > So, yes, we do see a combination of site design effect and the user's
> > knowledge of the genre playing a role in the strategy that users take.
> >
> > Jared
> >
>
>
> At this point I'm surprise nobody has brought up the issue (albeit
> librarian-centric) of known item vs. subject searching. That is, depending
> on whether you know exactly what you're looking for (Linksys Wifi) or just
> the general category/subject of items (wireless routers), you will likely
> take a different route to finding your information.
>
> Debating the usage of category browsing vs. searching based primarily on
how
> people use ecommerce or online news sites, seems to limit the debate a bit
> too much.
>
> Anybody who's actually completed a LIS degree (unlike myself) care to chime
> in on this? Lou?
>
>
> -kb
____________________________________________________
Richard Wiggins
Writing, Speaking, and Consulting on Internet Topics
rich at richardwiggins.com www.richardwiggins.com
More information about the Sigia-l
mailing list