Wikipedia is page one for 99% of Google UK searches, according to a study of 1,000 searches using randomly generated nouns. 

The research, carried out by Intelligent Positioning, raises a number of questions. 

Is Google giving too much prominence to Wikipedia, or are these rankings justified? 

Intelligent Positioning’s Sam Silverwood Cope searched Google in incognito mode using 1,000 keywords from a random noun generator. 

The results? 

  • Wikipedia is on page one of Google for 99% of noun searches (of nouns).
  • 96% of searches had Wikipedia in the top five positions. 
  • Wikipedia takes the number one spot for 56% of searches.
  • The online encyclopedia is number two for 24% of searches

As you would expect, Wikipedia does very well for ‘informational’ searches. This seems fair enough, as Wikipedia will often provide a quick, and well structured answer. Great for pub quiz cheats too. 

Sam questions the prominence of Wikipedia for pages with little real content. He quotes the example of a search for the term ‘air’ which returns Wikipedia at number two with a disambiguation page.  

This point about pages with little content ranking well has also been made by Barry Schwartz and others.

I asked Bigmouthmedia’s Andrew Girdwood if there was anything in this: 

It’s hard to deny Wikipedia its position as king of search. Google would certainly be able to see if Wikipedia failed to satisfy searchers as the engine would be able to detect the repeat searches and additional clicks. As long as Wikipedia provides satisfactory answers to searches then it is hard to argue the site does not deserve to rank.

That’s not to say there are instances when a particular Wikipedia page does not seem to be the best possible fit for one particular search. Individual pages do benefit from the authority of the site as a whole.

Girdwood suggests another way to look at the issue: 

The 1,000 search study was good and interesting. It certainly helps us investigate Google and algorithmic trends. However, not every word is equal and Google has talked for a long while about navigational, informational and commercial search terms. Wikipedia tends to rank strongly in the informational bracket and most words tend to be informational.  

A similar experiment that takes 1,000 brand terms that might be mistaken as informational rather than navigation would be interesting. For example, with a selection of words like  [apple], [virgin], etc and see how well Wikipedia does for those would likely have a very different result curve.

I haven’t had time to conduct an experiment using 1,000 brand terms, but I tried a few in incognito mode and Wikipedia ranks well for these:

  • Apple: number three on Google UK.
  • Argos: number two.
  • Hoover: third.
  • Orange: third.
  • Virgin: eighth.
  • Currys: third.
  • Boots: sixth.
  • Game: third. 

Sam did manage to find a few words for which Wikipedia didn’t rank on page one of Google. There were just eight of them:

  • Mail
  • news
  • trainers
  • national
  • sweets
  • wardrobe
  • phone
  • flight

What do you think? Is Wikipedia unfairly favoured by Google, or this this just a case of a content-rich site which deserves its lofty position?