Wikipedia Has The Answer To Life, The Universe, And Everything

Filed under: Search Engines by TheMadHat |
8/16/07 at 2:47 pm

This Wikipedia phenomenon is getting way out of hand. I know it’s been said before many times. Graywolf had an interesting post some time ago that stuck in my mind about Google and their love affair with Wikipedia that I decided to expand upon. I set my ~~scrapers~~ interns to work. Here is the process I went through (the data was pulled on August 12th):

From Wordtracker I pulled the top 1000 long term keywords (last 90 days) and the top 1000 short term keywords (last 48 hours) with offensive terms removed. I downloaded the buzz list from the SEOmoz popular searches tool. I then de-duped and cleaned the list and it came to a final count of 1,792 keywords. I needed to compare this to some other sites with a high level of “domain authority” so I also had my ~~scrapers~~ neighbors kids run rankings for About.com, Amazon, Craigslist, eBay, and MySpace. First Conclusion: Wikipedia has secret agents inside Google. Not only are they having a love affair with Google but MSN as well. They’re just married to Yahoo as they get the least amount of action from them. See the final results for yourself:

Google

Wikipedia Google Rankings

Yahoo

Wikipedia Yahoo Rankings

MSN

Wikipedia MSN Rankings

Holy Shit Captain Kirk! Wikipedia is the most or nearly the most relevant for almost half of all my searches! I bet Google can even search for Spock and find him faster than you! Give me a break Google. And Yahoo. And MSN. Wikipedia is full of crap all over the place. Where is Encyclopedia Britannica? Wouldn’t they also be a little bit relevant for something? Let’s look at some terms Wikipedia ranks for in the top 3:

* [search engine optimization] - Currently #1 on every engine and the most relevant source on SEO that exists. Sorry SearchEngineWatch. I guess you and even Danny Sullivan are much less relevant than Wikipedia on SEO.

* [SEO] - See above.

* [text messages] - okay, I suppose in case I don’t know what a text message is.

* [boys] - That’s creepy.

* [girls] - “A girl is a female child, as opposed to a boy, a male child.” Thanks. That deserves a #2 spot.

* [firefox] - (MSN) obviously the Wikipedia entry is more relevant than the actual Firefox page.

* [cheerleaders] - You really think I want to know what a cheerleader is?

* [lap dance] - See above

* [internet] - what is this thing?

* [art] - I’d like to buy some art. Or look at some art. Nope, I get to find out that “Art is a (product of) human activity, made with the intention of stimulating the human senses”

* [booty] - what?

Obviously this could go on and on. I think it’s time for all the search engines to take a hard look at the value of Wikipedia and what a trusted domain really is.

** If you would like to download all the data (terms, ranks, etc) from this experiment, you may grab the rar file here.

You Might Also Enjoy:

Do Not Click Here
Yahoo Screwing Up Titles And Results
In Loving Memory Of The Two Greatest Dogs That Ever Lived

8 Comments »

the reason for wikipedia’s high rankings is simple - whenever anyone wants to explain an unfamiliar term (either in a forum or on a blog post), they make the term a link to it’s wikipedia article. As a result of unintentional Google bombing by bloggers everywhere, wikipedia’s search ranking goes out of control. (they’re the number one result under “google bombing”, too, in case you were curious)

Comment by fin — August 20, 2007 @ 3:12 am

Wrong. It’s because in Google’s eyes they’re the most trusted domain anywhere. I understand the fact that wikipedia is going to rank for a lot of terms due to inbound linking, but these results should be relevant and probably should be showing up. Let me show you some examples that takes the guts out of that statement:

[BOYS]:
Rank #1 - Inlinks: 395 - (Wikiepedia)
Rank #8 - Inlinks: 1,986
Rank #10 - Inlinks: 2,145

In this case I scraped the backlinks of #1 wikipedia page and the term [boys] was in the anchor text of 18 links and [boy] in 119. and some iteration with boy 163 times.
I scraped the backlinks of #10 walmart and the term [boys] was in the anchor text of 348.

SEO:
Rank #1 - Inlinks: 14,127 (wikipedia)
Rank #2 - Inlinks: 469,464
Rank #6 - Inlinks: 163,164
Rank #10 - Inlinks: 209,141

It’s all about domain trust and Wikipedia has too much of it.

Comment by TheMadHat' — August 20, 2007 @ 11:15 am

Wow! Its not like we in the SEO community haven’t been saying this for a while. The saddest thing is , to see young kids thinking they can do book reports based on Wiki. Or people who lazily link to it to define something.

Comment by Steaprok — August 20, 2007 @ 9:49 pm

Thanks for the comment Steaprok. What’s worse is mainstream media referencing and quoting Wiki. These people need to get it through their skulls that anyone can edit and manipulate these pages almost at will. The fact that Wiki ranks in the top 10 for 70% of the most popular terms on the web almost gave me a stroke.

Comment by TheMadHat — August 20, 2007 @ 11:46 pm

This post is water on my mill (as we say in Denmark).

I just wrote a post on how companies, governments and even the Vatican edits their own pages on Wiki.

Usually they remove the negative stuff or add rumours to their competitors pages.

The recent release of a Wiki-scanner helped reveal some of this.

http://www.thebetanews.com/maj.....a-scanner/

Comment by Dennis Bjørn Petersen — August 21, 2007 @ 10:39 am

[…] Aaron from The Mad Hat explains why Wikipedia has the answers to Life, the Universe and Everything. My previous post about the Wiki Scanner is a great supplement to Aaron’s post. […]

Pingback by Linking by The Beta News | The Beta News — August 22, 2007 @ 10:29 am