|

|

|

|
|
Google
Search Engine "Florida Update"
On November 16th, 2003 the Search Engine Google commenced
an update (the "Florida update") on how it ranks
web sites for the results of keyword searches. This new update
in how google ranks web sites has had a catastrophic effect
for a very large number of web sites and has turned web site
marketing and the fledgling new industry of search engine
optimization on its head.
In a nutshell, a vast number of web site pages, many of
which had ranked at or near the top of the results for specific
search queries for a very long time, simply disappeared from
the results altogether. Also, the quality (relevancy) of
the results for a great many searches was reduced. Many have
speculated that Google is attempting to force web sites to
use its AdWords pay-per-click (PPC) advertising program for
certain commercially lucrative key word searches.
While it is not totally clear exactly what Google is trying
to accomplish or how its new ranking algorithm works, it
is clear that only certain keyword searches (deemed commercialized
by Google) are affected by this algorithm. The chart below
depicts the current best guess as to how the new Google web
site indexing algorithm works.
"Florida
Update" Filter Chart - (explanation
follows below)

|
|
|

|

|

|
|

|

|

|
|
List
#
|
Decision
Box
|
Description |
|
-
|
Search
Phrase Entered |
A 2 or 3-word phrase is entered into the Google Search
Box, with a given site in mind.
All the tests apply to the INDEX page of "Some Given
Site" that you are looking at. |
|
1
|
English
Test |
Currently, only English language sites are subject
to the filter.
If the site is NOT in English, then there is no
filtering (one French word exception noted).
END
If the site is in English, go to the next test. |
|
2
|
Dictionary
Test |
Is the 2 or 3-word search phrase listed in
the Florida Dictionary of "Poisoned Commercial
Words"? (which many feel are taken directly from the
AdWord Database)
http://www.scroogle.org/hitlist.html
If they are not, then there is no
filtering.
END
If the words ARE in the dictionary, go to the next
test. |
|
3
|
Commercial
Test |
Is the site listed in a commercial category of the Google
Directory (derived from the ODP (Open Directory
Project) at www.dmoz.org), whether it is a commercial
site or not?
http://directory.google.com/Top/Shopping
http://dmoz.org
If the answer is yes, then the next test is, "Are you
in Froogle?"
http://froogle.google.com
If the answer is yes, the site is FILTERED no
matter what.
END
If the answer is no, go to the next test. |
|
4
|
Hilltop
Test |
The new HILLTOP ALGORITHM is applied.
Is the site itself an "Expert" site, as defined
by the Hilltop dissertation?
http://www.cs.toronto.edu/~georgem/hilltop
or, for the condensed version -
http://www.cs.toronto.edu/~georgem/hilltop/poster.html
If the answer is yes, there is no
filtering.
END
Does the site have many links from the same domain?
If the answer is yes, the site is FILTERED no
matter what.
This is an attack on Affiliates and Link-Farms.
The following work in combination:
Do you have a lot of links from "Expert" sites?
Do you have lots of links from -edu, -gov and -org sites?
Are your links very diverse - many links from many
DIFFERENT sites?
If the answers are yes, in some aggregate form, then
there is no filtering.
END
If the answer is no, go to the next test. |
|
5
|
SEO
Test |
Has the site been "over-optimized" FOR
THIS PHRASE, according to the old rules of Search Engine
Optimization?
the basis for which are here -
http://www.google.com/webmasters/guidelines.html
along with "best practice", Google-conforming,
NON-SPAM SEO techniques.
These are the previous text factors that we are the
most familiar with - Title, description meta, keyword
meta, body repetitions, H1, anchor, alts, etc.)
If the answer is no, there is no
filtering.
If the answer is yes, the site is FILTERED.
END OF DISCUSSION |
|
|
|
|
|
|
|

|

|

|
|
|