SEO Rick on 18 Mar 2008 08:19 am
SEO Lessons From Wikipedia
Wikipedia gets a lot of Google love. By some estimations Google drives more traffic to Wikipedia than to any other site. Just do some random searches in Google and see how many times a Wikipedia page comes up very high in the results.
So presumably we can learn something about SEO from Wikipedia. According to this article and discussion, What Wikipedia can teach us about SEO, there are a few important things that account for Wikipedia’s SEO success:
1. Wikipedia uses search engine friendly file names, title tags and page descriptions.
2. Wikipedia has a ton of text based content - and, of course, it is very keyword focused.
3. The internal linking between Wikipedia pages is very extensive and very thorough.
4. Wikipedia has more than 5 million inbound links, and many (if not most) to internal pages.
This should make Wikipedia a much-copied model for all websites looking for good Google rankings. But the fact is, many marketers - especially those running ecommerce sites - do not appreciate the importance of on-page optimization. They are content to show a picture of a product with a short description and price. But if they were to follow the Wikipedia model they would use keyword-rich titles for all product pages, along with optimized descriptions on every page. They would also inter-link internal pages - not just from a common navigation bar, but from within the text of pages by using such devices as “Similar products you might be interested in…”
However, as many of us involved in providing SEO services know, many clients are not prepared to go that far. It would take much more than just tweaking their page titles and metatags. It would take a remake of their entire site, involving writing of a lot of text, and actually providing some useful information about the products.
Most prefer to just offload the problem to someone else and spend their money on things like link building. But that is only half the battle. Without decent content it is hard to win at the SEO game.
on 19 Mar 2008 at 4:27 pm 1.Jake said …
The problem is that other sites can never generate the immense about of editors and therefore content that Wikipedia has. However, I believe you are missing the essential success of Wikipedia that is what SEOs need to focus on. If you are running a site, you need to write loads of content. That is what all of the major sites have. Good quality content. The other sites are primarily ways to find the said content.
I am personally running an experiment with my newest website (my name should be hyperlinked to it for those interested) to try putting my theories to the test. I am using only free resources (including hosting and “domain”) and no real SEO tactics (I’m not even including the meta description and keyword tags to make a new website. My focus is going to be on putting out one to two articles of quality content a day. Note that quality is relative of course, as I can not put out near as good of content as many can.
Though I may be a bit to ambitious for my own right, I am going to try to keep this up for at least six months to see if I can do well in both traffic and search engine rankings. I am indirectly following the most essential part of Wikipedia and other search engine “hogs” such as About.com: content.
I apologize if I misread your article in any way. Feel free to correct me.
on 19 Mar 2008 at 5:39 pm 2.Rick said …
Hi Jake,
Thanks for the extensive comment. Sounds like a very aggressive experiment you are planning. If you can do an article or two of good stuff every day you will be doing better than most of us.
Just to make it clear, I agree with you about the importance of content. But I’m not sure why you conclude that the “essential success of Wikipedia” does not include the very obvious SEO elements that I mentioned. Especially since these are part and parcel of the Wikipedia approach to the presentation of their content. If you just publish “quality content” without making it keyword focused, without interlinking it, without having lots of inbound links, I suspect you will find it will be ignored by Google.
That’s because it is not at all obvious what is meant by “quality content”. The search engines do not have some kind of magical “quality” filters that let them distinguish between my political commentary (for instance) and that of some world famous professor or author.
The SEs simply look for patterns and concentrations of words, and inter-relationships (primarily links) with other content having similar concentrations. As far as I know, that is what SEO is about - making those patterns obvious.
on 19 Mar 2008 at 6:52 pm 3.Alice said …
Thanks for the valuable information. I think you covered a very good point about the internal linking.
on 20 Mar 2008 at 6:49 am 4.Jake said …
@Rick: I know one web browser of a successful site (whose name I will not mention here unless I am later able to get his permission) who does not care in the slightest about SEO. He instead considers it as stupid as, say, witchcraft. However, his highly successful computer site now comes out near the top of Google SERPs for many searches, particularly Vista problem related ones. A good percentage of his traffic originates from Google. In fact, the number is over eighty percent I believe.
I believe that with Wikipedia most of the SEO benefits come with how it works. Internal linking? typical of a wiki. Tons of text based content? thousands to tens of thousands of users tend to create that. Internal links? it’s quality content that is often the clearest and most informative place to link to.
Meta tags and such, on the other hand, are a fairly recent addition if I recall correctly. I’ll try to look into that soon.
on 20 Mar 2008 at 7:41 am 5.Rick said …
Jake,
Yes, I agree that wikis, like blogs to a lesser degree, are structured to create internal links more or less automatically. However, the majority of sites do not have this feature. So a web designer/programmer not using one of these formats has to go out of his or her way to add systematic internal linking, if they think it is important.
I also agree it is possible to make a search engine friendly site without consciously engaging in SEO “witchcraft”. It is also possible to be a very good golfer or baseball player without analyzing your swing or practicing on a regular basis. But I don’t think you should conclude from that that practice is a waste of time.
With regard to metatags, most “experts” agree that the keyword tag is fairly useless as far as search engine optimization is concerned. As I understand it metatags have been with us, in one form or another, since the beginning of the web - or at least since the beginning of search engines. Before Google came along metatags were the primary means that search engines used to rank content. But Google shifted the emphasis to backlinks and content analysis, and the other SEs have followed suit.
Most “experts” (at least the ones I have read) also agree that the page title tag is still very important for Google, and the description tag is also sometimes used for descriptions in SERPs, and is therefore important if you want people looking at SERPs to know what your pages are about.
Because the SEs change things around on a regular basis, the only way any of us can be sure of the relative importance of different tags (or other optimization techniques) is by testing and analysing actual results pages. I assume that is one of the purposes of your experiment.
on 25 Mar 2008 at 11:36 am 6.Daniel Boddington said …
Good article,
I think you are quite right in saying that the techniques applied in the design of Wikipedia are textbook.
However, it is important to understand that the size of Wikipedia’s content base is what provides its almost unbeatable rankings for such a broad range of subjects, the individual content on that page may not be any better optimised than the next, but the rankings linked to it from the rest of the site give it an almost unbeatable ranking in uncompetive subjects, despite it not being an ‘authority’ site.
If anything, this has more to do with Google giving extra value to unrelated linkages within the site. But hey, if it workfor them I dont see them complaining, I wouldnt!
Ive written more about this in one of my own articles:
http://www.justwebdesigners.co.uk/blog/2008/02/05/wikipedia-the-font-of-all-knowledge/
we all love a shameless plug
Dan
on 25 Mar 2008 at 1:07 pm 7.Rick said …
Hi Dan,
Thanks for the comment. I’ve actually read and commented on your article a few weeks ago.
I’m not sure why you don’t consider Wikipedia an “authority” site. Judging from the article you reference it has something to do with the high probability that Wikipedia’s articles are not as trustworthy or accurate as those from other more reliable sites.
Correct me if I’m wrong, but Google does not rank sites on the basis of accuracy, reliability or trustworthiness. As far as I know there is no way they can get at these things just by looking at numbers and words. They rank them on the basis of things that can be quantified by their algorithms - links, prevalence of specific words, etc., and then make a leap to the conclusion that these things imply accuracy, reliability and trustworthiness.
The job of the SEO strategist (IMHO) is to address those quantifiable variables. It is assumed that the quality of the content will take care of itself - for example, that the best, most accurate content will get more inbound links. This is another very questionable Google assumption, but it is the way Google has set the game up.
Of course I may have missed your point. Perhaps the concern is that Wikipedia is generating artificial links by structure alone and that this skews the results unfairly.
I agree with this entirely. It is just another reason why Google’s results should be looked upon with scepticism. It is also another reason why using strategic SEO techniques is a good and acceptable way to level the playing field.
on 13 Apr 2008 at 9:25 pm 8.Josip Barbaric said …
Very interesting post! It made me think about a lot of things.
I really never paid too much attention to inter linking. Guess I’ll have to start!
Thanks for the post!
Cheers!