Does Javascript Affect Ranking?
by Jon Ricerca
http://www.SearchEngineGeek.com

Almost all SEO's agree that using too much javascript
can harm your rankings and might confuse the search
engines. Is it true? We decided to answer this extremely
simple question for the two leading search engines using
a simple statistical analysis.

The methodology: I gathered the results of the queries
that were naturally performed last month by myself and
three associates using the two leading search engines and
analyzed them. I had to visit each page and check the HTML
source code to see if javascript was being used. I counted
the number of pages found that utilized javascript for the
first 8 rankings. The results for each of the two leading
search engines were kept separate so that we could
discover any differences between the two leading search
engines for this factor.

The resulting graphs show the number of pages utilizing
javascript for each ranking. The Y-axis shows the number
of pages found utilizing javascript, while the X-axis
shows rankings 1 through 8. Here are the graphs for each
of the two leading search engines:

http://www.SearchEngineGeek.com/graphs/dcy02.jpg
http://www.SearchEngineGeek.com/graphs/dcg02.jpg

The result is very conclusive and very surprising. Both
leading search engines rank pages that utilize javascript
higher than pages that do not utilize javascript. This is
another example of how the guesses of the SEO community
are often just that... guesses... They have a 50% chance
of being correct on any particular factor. In this case,
the majority of the SEO community guessed wrong.

Notes:

1. There was no exercise to attempt to isolate different
keywords. I merely took a random sampling of the queries
performed by myself and three associates during the month.

Conclusion:

Pages which utilize javascript rank higher than pages that
do not utilize javascript on both of the leading search
engines.

This is merely a correlation study, so it cannot be
determined from this study whether the leading search
engines purposefully entertain this factor or not. The
actual factors used may be far distant from the factor we
studied, but the end result is that these search engines
do, in fact, rank pages with javascript higher than pages
without javascript in the study.

Jon Ricerca is one of the leading researchers and authors
of the Search Engine Ranking Factor (SERF) reports at
SearchEngineGeek.com. For access to the other SERF
reports, please visit:
http://www.SearchEngineGeek.com

Does URL Length Affect Ranking?
by Jon Ricerca
http://www.SearchEngineGeek.com

I was recently perusing results on the leading search
engines, I seemed to notice that long URLs weren't very
popular on the first couple pages for several keywords.
That intrigued me, so I performed a true statistical
analysis to see if my observation was merely a coincidence
or a true correlation.

The methodology: I gathered the results of the queries
that were naturally performed last month by myself and
three associates using the two leading search engines and
analyzed them. I counted the characters in the URL
(including the http://) and tabulated the results against
the ranking of the URL in the search results. The
tabulated results were finally converted into a
normalized "ranking correlation." The results for each of
the two leading search engines were kept separate so that
we could discover any differences between the two leading
search engines for this factor.

The resulting graphs show the results for groupings of URL
lengths normalized into a number between -100 and +100
showing the likelihood of being ranked higher/lower. A
value of +100 shows that all 10 rankings were in the
proper order to show that pages of the studied value
ALWAYS rank HIGHER than pages of another value. A value
of -100 shows that all 10 rankings were in the proper
order to show that pages of the studied value ALWAYS rank
LOWER than pages of another value. Numbers in between show
the varying likelihood of rankings proportionally between -
100 and +100.

That is the number you see on the Y-axis. On the X-axis,
we have the URL lengths starting with a grouping of URLs
that were 11-20 characters long and continuing with each
group of 10 URL lengths up until 61-70. Here are the
graphs for the two leading search engines:

http://www.SearchEngineGeek.com/graphs/ddy03.jpg
http://www.SearchEngineGeek.com/graphs/ddg03.jpg

The URL lengths were grouped in this way in order to
increase the number of data points available.
Unfortunately it also reduces the precision of the
results. It is possible to tell that URL lengths from 21-
30 rank much higher than URLs of lengths 61-70, but you
are unable to see if URLs of a length of 22 rank
differently than URLs of a length of 24 (for example). I
did not list the grouping of URLs of a length of 1-10
because the number of data points were too small to
accurately calculate a ranking correlation. I did not show
any URLs longer than 70 characters for the same reason.

The result is very conclusive. Both leading search engines
rank sites having URLs between 11 and 30 characters
(inclusive) much higher than any other URL length studied!

Notes:

1. There was no exercise to attempt to isolate different
keywords. I merely took a random sampling of the queries
performed by myself and three associates during the month.

2. This is merely a correlation study, so it cannot be
determined from this study whether the leading search
engines purposefully entertain this factor or not. The
actual factors used may be far distant from the factor we
studied.

Jon Ricerca is one of the leading researchers and authors
of the Search Engine Ranking Factor (SERF) reports at
SearchEngineGeek.com. For access to the other SERF reports,
please visit: http://www.SearchEngineGeek.com

Do Static IP Sites Rank Higher?
by Jon Ricerca
http://www.SearchEngineGeek.com

This is a hotly debated topic. Some SEOs claim that sites with a static IP address rank higher while other SEOs claim that shared hosting is just fine... that it would be stupid for search engines to penalize shared hosting since we are running out of IP addresses and so many sites are currently using name based hosting.

First, let's define what we are talking about when we say "static IP" vs "name based" hosting. Here are some
synonyms:

For "static hosting", the following all mean the same
thing: static IP, dedicated IP, http/1.0, non-shared hosting. For "dynamic hosting", the following all mean the same thing: dynamic IP, name based hosting, http/1.1.

Let's take a brief history of the Internet to put things more in focus. Once upon a time, every host on the Internet had a unique IP address. It is usually expressed as 4 numbers from 0-255 separated by dots. An example would be 207.44.161.131. There are billions of such IP addresses possible... however, there is a finite number.
An organization called ARIN hands out IP addresses in the western hemisphere while another organization handles Europe and another Asia. Those organizations noticed several years ago that we were going to eventually run out of those IP addresses because of the proliferation of web-sites. At the time, every website had it's own dedicated IP address that was associated with it's domain name.

The solution? A new protocol was developed called HTTP 1.1 (to replace HTTP 1.0). The new protocol allowed more than one domain/website to share the same IP address. In fact, hundreds of websites can now share the same IP address.
The new type of hosting is called "name based", "shared IP", "http 1.1", etc.

Some SEOs theorize that your choice of dedicated hosting vs. shared hosting might affect your rankings. Some others claim that is ridiculous because all hosting will eventually be shared in order to preserve IP addresses.
Which are correct?

I decided to run it through our statistical analysis engine to get the facts. Here is the methodology I used to answer this question. I gathered the results of the queries naturally performed last month by myself and three associates using Yahoo and Google. I then pinged each site to get it's IP address. I then tried to visit the site using the IP address. With shared hosting, this isn't possible. You get some kind of generic page instead of the specific site you want. I tallied my results for each of the first eight rankings.

On the Y-axis, you will see the number of sites found that use a static IP (do not use shared/name based hosting). On the X-axis, we have rankings from 1 to 8. Here is the graph showing Yahoo and Google results:

http://www.SearchEngineGeek.com/graphs/de05.gif

(Note to webmasters: You are welcome to hot link to the above graph or even copy it to your own site. You are also welcome to delete this note.)

First, it is interesting to note that the number of sites using shared vs. dedicated hosting is just about half and half. We expect that as time goes on, more and more sites will be using shared hosting.

The second thing to note is that there is no trend for either Yahoo or Google that would indicate any preference for either type of site. Google might show a slight preference for static IP. Yahoo might show a slight preference for shared hosting. The net effect is null. The correlations for both are very close to zero (on a scale of -100 to +100, Google's correlation was a +35 and Yahoo's was a -21). I generally consider anything from -35 to +35 to be statistically insignificant.

Do you use dedicated hosting in order to improve your rankings? Don't bother. There isn't any advantage in this factor. Often dedicated hosting is more expensive than static IP hosting... so save your money and invest it in other factors that do affect your ranking!

Notes:

1. Over 1,000 queries and over 10,000 sites were examined for this study.

2. There was no exercise to attempt to isolate different keywords. I merely took a random sampling of the queries performed by myself and three associates during the prior month.

Conclusion:

Sites using static hosting do not rank significantly higher or lower than sites using shared hosting on both Yahoo and Google.

This is merely a correlation study, so it cannot be determined from this study whether the leading search engines purposefully entertain this factor or not. The actual factors used may be far distant from the factor we studied, but the end result is that both of these search engines do, in fact, rank pages with a "window.open"
command higher on average.

Jon Ricerca is one of the leading researchers and authors of the Search Engine Ranking Factor (SERF) reports at SearchEngineGeek.com. For access to the other SERF reports, please visit:
http://www.SearchEngineGeek.com

Does Keyword Location On Page Affect Ranking?
by Jon Ricerca
http://www.SearchEngineGeek.com

One of our members asked for a study of keyword location on a page. Is it more effective to have your keyword mentioned in the top third, middle third or bottom third of a page?

Here is the methodology I used to answer this question. I gathered the results of the queries naturally performed last month by myself and three associates using Yahoo and Google. I then fetched the pages and divided the body section into three equal parts for each page. I tallied the results for the first 8 rankings on both Yahoo and Google (keeping the results separate) and then converted them into a percentage of the total results for each search engine.

Here are the graphs showing Google and Yahoo results:

http://www.SearchEngineGeek.com/graphs/dfg02.gif
http://www.SearchEngineGeek.com/graphs/dfy02.gif

(Note to webmasters/ezine editors: You are welcome to hot link to these graphs or even copy them to your own site.
You are also welcome to remove this note.)

The X-axis shows the ranking (from #1 through # of the search engine results in the study. The Y-axis shows the percentage of domains that contained the keyword in the top (red line), middle (blue line) and bottom (purple
line) thirds of the body section of the page.

It is interesting to note that pages containing the keyword in the top and bottom third of the body section ranked much better on Google. The top section had a normalized correlation of +42 on a scale of -100 to +100.
The bottom third also showed a remarkable positive correlation of +46 on the same scale. Having the keyword in the middle third had no significant effect (no correlation whatsoever... neither positive nor negative).

The Yahoo results were even more interesting. I generally ignore any correlations between -35 and +35 as being generally insignificant. On Yahoo, none of the three sections showed any remarkable correlations. The scores were a +17 for the top, -3 for the middle and -17 for the bottom third of the body section. Does this mean that Yahoo doesn't even look for the keyword in the body section?

Advice: Mention your keywords near the top and/or bottom of a page for Google ranking.

Jon Ricerca is one of the leading researchers and authors of the Search Engine Ranking Factor (SERF) reports at SearchEngineGeek.com. For access to the other SERF reports, please visit:
http://www.SearchEngineGeek.com

Do Commas In Your Keywords Metatag Affect Ranking?

One of our members recently reported that Yahoo was recommending separating the keywords in your meta-tag using commas. As we know from experience, the official representatives of the search engines don't always give us the best advice as it pertains to ranking. We decided to check out this claim using a statistical analysis to find out if using commas in your keywords meta-tag had any affect on ranking on Yahoo. We also studied Google to see if there was any affect on that search engine.

Here is the methodology I used to answer this question. I gathered the results of the queries naturally performed last month by myself and four associates using Yahoo and Google. I then fetched the pages and looks at any keywords meta-tags on the listed pages. I tallied the results for the first 8 rankings on both Yahoo and Google (keeping the results separate) and then converted them into a percentage of the total results for each search engine.

Here is the graph showing Google and Yahoo results:

http://searchenginegeek.com/graphs/df03.gif

(Note to webmasters: Feel free to hot-link to the above graph or even copy it to your own site. Also feel free to remove this note.)

The X-axis shows the ranking (from #1 through # of the search engine results in the study. The Y-axis shows the percentage of domains that contained commas in the keywords meta-tag.

The first thing to notice is that roughly 40%-50% of all pages in the first eight rankings have a keywords meta-tag with commas separating the keywords. That really doesn't tell us much about the ranking of such pages though, only the general distribution of pages with commas in a keywords meta-tag in the top 8 results. Still, I was surprised that the percentages were so high. For several years, the general consensus was that no significant search engine even utilized the keywords meta-tag (this was before Yahoo switched from service Google results to their own). With that general consensus, I expected that a vast majority of sites had dropped using this meta-tag.
However, over 40% of top ranked sites continue to persist in their use. I find that interesting.

The next thing to notice is that Google showed absolutely no ranking difference between sites that use commas in the keywords meta-tag vs. sites that do not use commas in the keywords meta-tag or have no keywords meta-tag at all. The ranking correlation was an exact zero on a scale of -100 to +100.

Yahoo's ranking correlation was a -28 on a scale of -100 to +100 for pages having commas in the keywords meta-tags.
I generally treat results between -35 and +35 as insignificant (don't not affect ranking or the affect on ranking is small to insignificant). However, since Yahoo recommended using commas, I did find the -28 result interesting. Once again, it appears that the official representatives of a search engine are steering us wrong (at least for purposes of ranking higher on their search engine). This result indicates that using commas is either insignificant or if there is any significant affect... it is NEGATIVE!

Advice: Don't use commas in your keywords meta-tag. Pages that use commas in the keywords meta-tag do NOT rank higher and there may be a slight negative affect on Yahoo.

Jon Ricerca is one of the leading researchers and authors of the Search Engine Ranking Factor (SERF) reports at SearchEngineGeek.com. For access to the other SERF reports, please visit:
http://www.SearchEngineGeek.com