[thelist] your research on search engines and the sites you work on

.jeff jeff at members.evolt.org
Wed Nov 6 15:45:01 CST 2002


ben,

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: Ben Dyer
>
> Thus far, I've had reasonable luck with the Spider Safe
> URL ISAPI filter from CFDEV:
>
> <http://www.cfdev.com/products/productdetail.cfm/id/1023>
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

fwiw, i've posted numerous times on other discussion lists about the
difference between pseudo-"search engine friendly" urls and true search
engine friendly urls -- aka, directory-style query strings.  here's a copy
and paste from a comprehensive email to another list that happens to be
talking about our very own site -- evolt.org:

><><><><><><><><><><><> begin copy <><><><><><><><><><><><

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: Pete Freitag
>
> > unfortunately, the pattern you mention is only
> > marginally search engine friendly.  true search engine
> > friendly urls look like this:
>
> Out of curiosity which search engines don't like url's
> like http://www.site.com/x.cfm/id/1 The only difference
> is that they have a dot in them, right?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

unfortunately i can't cite documentation for search engines.  i can only
speak from experience with a site (http://evolt.org/) that changed from the
slash delimited query string method to directory-style query strings.  after
we made the change we saw a huge jump in search engine trafficing and
rankings.  not much else changed about the site so the only logical
conclusion is that that was the culprit.

in my mind, using search engine friendly urls is about more than just being
search engine friendly.  it's about making your site user-friendly as well.
it's much easier, as a user to type in:

http://www.site.com/calendar/2001/12/02/

than it is:

http://www.site.com/index.cfm/fuseaction/calendar/year/2001/month/12/day/02/

which is only a far cry better than:

http://www.site.com/index.cfm?fuseaction=calendar&year=2001&month=12&day=02

if the application is designed right, lopping off the "directories" should
result in more and more broad views of the data.

it makes it ultra-easy to build in breadcrumb navigation in to the site as
well.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> I know google allows them (whats weird about google is
> that it caches and indexes dynamic pages [with ?] but it
> doesn't give them page ranks so they almost never come
> up in searches).
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

with the site i mentioned we had the same ranking problem with the
slash-delimited query strings.  the only thing i can figure is that google
sees the dot in the path before the slashes, figures the content might be
dynamic, and weights it accordingly.

the main site has been converted over to directory-style query strings and
the content has fantastic rankings.  dynamic content on one of the subsites
(http://lists.evolt.org/index.cfm/a/harvest) that still uses the
slash-delimited query strings has been around almost as long but is hard to
find in the search results cause it's so buried (yet it's the same type of
content found on the main site).  on this same subsite there are also
discussion list archives (http://lists.evolt.org/archive/) that are totally
static and they're also really well indexed and ranked.

the only disconnect between all of these sections is the paths.  those with
directory-style query strings or completely static content are ranked much
higher.

><><><><><><><><><><><> end copy <><><><><><><><><><><><><

there's something for ya to chew on.

btw, just as an aside, but making your site as search engine friendly as
possible will have another benefit in that it'll suddenly have more human
readable and memorable urls to content.

good luck,

.jeff

http://evolt.org/
jeff at members.evolt.org
http://members.evolt.org/jeff/






More information about the thelist mailing list