Thanks guys,

I understand completely what you are saying and James, yes you are right in
that what I'm getting out of this is a way of improving keywords etc.
However, I'm try to correlate the searches used to find the best blend of
keywords, and which pages to place keywords on for external searches, and
make the vistors on the site see the information easily without needing
internal searches. Visitors coming from searches outside are giving keyword
hits a hammering, however, when I look at internal searches, very few
keywords are hit in fact the search results seem to run at about 3-5% of the
numbers from the outside searches. Even though the search page is actually
hit more times than we get referring visits.

For me, the perfect solution would be for Analog to amalgamate the search
results by ignoring search requests where the refering URL and IP are the
same. This would reduce the external search word count down to a more
realistic figure.

Interesting that Aengus pointed out the internal search can be treated as
external. In fact, I already have a similar entry in the SEARCHENGINE param
that does this. What I actually do at present, is run analog twice. Once
with the internal search included in the SEARCHENGINE list, once with just
the internal search in the SEARCHENGINE list. This at least gives me the
correlation between what's searched internally and externally on an even
par. It also let's me get an idea of the factor of difference between the
INTSEARCHENGINE results and the SEARCHENGINE results which is where I
derived the 3-5% from.

Thanks for the comments, I only got into Analog about 3-4 months ago, taking
over the responsibilty from another member of staff and have made some
fairly important discoveries on our site as a result. Probably one of the
easiest tools out there to use and just love the computer readable output
format.

Matt

-----Original Message-----
From: James Riemermann [mailto:James.Riemermann@;state.mn.us]
Sent: 21 October 2002 21:35
To: [EMAIL PROTECTED]
Subject: Re: [analog-help] INTSEARCHENGINE Inconsistency


Matt,

Beyond the points Stephen and Aengus made, I'm not sure from your message
that
you're thinking about the INTSEARCHENGINE report in a meaningful way. This
tells
you nothing at all about how anyone finds your site, but can reveal things
about
their behavior once they get there. Analyzing the SEARCHENGINE report can
tell you
something about what keywords on your web site are corresponding with words
users
are typing in on various search engines. You only get info on people who
start on
that search engine and are led to your site. And of course, it tells you
nothing
about people who might have been looking for info your site contains, but
didn't
find it.

The INTSEARCHENGINE report, on the other hand, offers no info on how people
find
your site, but can tell you a fair amount about their way of looking for
things
once they get there, regardless of whether those searches provide them with
helpful
results. For instance, if none of the top ten internal search phrases used
are
showing up in your home page navigation, you might want to reconsider your
navigation scheme. Or, if the most commonly used phrases deliver unhelpful
results,
you might want to work with that. (The INTSEARCHENGINE report can also be
used for
purposes entirely unconnected with searches of any kind, but that's another
subject.)

Basically, SEARCHENGINE provides info similar to the search engine reports
other
analysis tools provide, and there's no reason to look to INTSEARCHENGINE for
that
sort of information. Additionally, when you speak of visitors clicking back
to, for
instance, Google's search results, I think there's a pretty good chance
their next
click will be to some other site entirely--if your site was what they were
looking
for on google, my guess is they're more likely to move forward than
backward. If
you don't have many INTSEARCHENGINE hits in comparison with INTSEARCHENGINE
hits,
the most likely explanation is that not many people are using your internal
search
engine. Which is fine, as long as they're finding what they need on your
site in
some manner.

james


============================================
James Riemermann
MN Office of Tourism
651/297-2077
[EMAIL PROTECTED]

For travel info: www.exploreminnesota.com
============================================

>>> [EMAIL PROTECTED] 10/21/02 09:46AM >>>
Hi,

I've got Analog 5.24 working just great including the INTSEARCHENGINE
function but thought it important to point out an interesting anomaly to
note. 

INTSEARCHENGINE reviews the actual URL passed whereas SEARCHENGINE reviews
the referring URL. This has a curious but expected result when reviewing the
statistics. One thing to remember, is that when someone searches your site,
they may be presented with several hits. If they click on a link, they get
directed to the page in the search list, this causes a line to be logged in
your site log. For external hits, this causes a search word and search query
hit, if the user clicks back and clicks another link into your site, you get
another line since it reviews the referring URL in the log. For internal
searches, you get a single hit for the initial search query, but no extra
hits for clicks to subsequent pages, clicking back and clicking another
result gives extra log lines, but since it's not reviewing the referring
URL, no extra search word hits.

The net effect is that the internal searches yield very few results, whereas
the external search can ramp up many times the number of results when in
reality, it could be just a single search. Difficult to say what is right or
wrong here, should analog try to amalgamate results from outside search
engines to counter the extra hits?

Any comments?

Matt

+------------------------------------------------------------------------
|  This is the analog-help mailing list. To unsubscribe from this
|  mailing list, go to
|    http://lists.isite.net/listgate/analog-help/unsubscribe.html 
|
|  List archives are available at
|    http://www.mail-archive.com/analog-help@;lists.isite.net/ 
|    http://lists.isite.net/listgate/analog-help/archives/ 
|    http://www.tallylist.com/archives/index.cfm/mlist.7 
+------------------------------------------------------------------------
+------------------------------------------------------------------------
|  This is the analog-help mailing list. To unsubscribe from this
|  mailing list, go to
|    http://lists.isite.net/listgate/analog-help/unsubscribe.html
|
|  List archives are available at
|    http://www.mail-archive.com/analog-help@;lists.isite.net/
|    http://lists.isite.net/listgate/analog-help/archives/
|    http://www.tallylist.com/archives/index.cfm/mlist.7
+------------------------------------------------------------------------
+------------------------------------------------------------------------
|  This is the analog-help mailing list. To unsubscribe from this
|  mailing list, go to
|    http://lists.isite.net/listgate/analog-help/unsubscribe.html
|
|  List archives are available at
|    http://www.mail-archive.com/analog-help@;lists.isite.net/
|    http://lists.isite.net/listgate/analog-help/archives/
|    http://www.tallylist.com/archives/index.cfm/mlist.7
+------------------------------------------------------------------------

Reply via email to