I have a love-hate relationship with Wikipedia. On one hand it is a great resource just chuck full of seemingly accurate information. On the other hand it has become so prominent in Google rankings that its pages normally hold the first or second slot on any single word search. This, in my opinion runs the risk of drowning out other great sources of information as there is a certain bias towards clicking the first result returned by Google. Plus everyone automatically links to Wikipedia so it is a self fulfilling prophecy. In 10 years the web will consist of social networking sites and Wikipedia (!).

I also have an issue as some of the major pages are edited by editors/dictators who rarely let anything get added to the page and control links out like they are gold (which they are due to the PR of Wikipedia pages - but still). It is almost as if Wikipedia is becoming no better than any other Encyclopedia as for the large majority of topics it is essentially a closed system.

That is why I love sites like the Uncyclopedia. It is nice to see someone poke fun at Big Brother.

One of the most talked things on the net among those of us who like to launch websites is search engine optimization (SEO). There are armies of consultants, programmers, online marketers and bloggers who devote their time to guessing what the major search engine robots will find relevant and what they will flag as negative content. Since Google is the Gorilla in the space, much of that effort is devoted to chasing the elusive PageRank, where a difference of a few points can make the difference between web obscurity or a great Alexa ranking.

In building TravelGator we’ve had many of these conversations between ourselves and despite thinking we knew something about SEO when we started (and actually reading all the advice out there from all those consultants who have hung up an SEO shingle), we quickly learned that we were SEO idiots and sequentially made every mistake in the book.

So, while I cannot claim to have figured it out yet (and am still working hard on making Google love the sites I put up), I can share a bit of what not to do from personal experience.

• Don’t build a dynamic site without burying page level parameters in the session (as opposed to the url). TravelGator is excessively dynamic as we wanted to let users access content from many different directions. Therefore our pages change their structure rather significantly depending on where a user comes from. Good for navigation and usability, bad for SEO. We had to go back and bury all that sophistication in session objects and make our urls plain vanilla.
• While usability and great content is critical to designing a good site, don’t forget that the search engines are still rather focused on keyword density and check the density of your pages. TravelGator still gets killed by other sites that have much less content but repeat the same phrase over and over again on the site (not surprisingly including major keywords in the phrase). As much as I’d prefer Google actually learned to read, it is still pretty dumb when it comes to text and context.
• Don’t forget that you still need an old fashioned, 1995 era site map even if your site is pretty dynamic. We’ve found the robots are pretty lazy and won’t dive down into the site much below 4-5 levels or navigate long paths of clicks to get to a page. They are also terrible at following the results of internal site search engines. So on your site only for the consumption of the robots, build a site map that is a couple of levels deep.
• Don’t duplicate your content around the site. Google ascribes what seems like a huge penalty to duplicated content so if you have the same material on more than one page, give all versions of the page a seemingly static url (see point one above).
• Ignore all the chatter on XML based site maps. We did it…had no effect whatsoever on the number of pages we got indexed. Maybe at some point in the future this will matter but not right now. Do it only if you have lots of free time.
• Don’t forget to put keywords in your urls. This is a huge debate online as to whether they are required or not. After a year of looking at search results and trying to figure out why one page is ranked higher than the other my only conclusion is that they matter. We’ve caved and are adding them to the site.
• Don’t syndicate your content. See the point above about duplicated content - it doesn’t matter where it is. So if you are buying, stealing or otherwise acquiring content that is visible somewhere out there on the web, forget about it helping with your PageRank. It will only hurt it (those pages will also end up in the supplemental index…otherwise known as Google hell). The same rule applies if you sell your stuff to others. It becomes pretty useless from an SEO perspective. Unique content rules.

That’s my list so far. Everything else we’ve tried hasn’t worked but by fixing the above issues we’ve had some good success. Not fantastic mind you, but better than if we hadn’t made the changes. I’ll update this post if I come up with anything else that actually works.

There are some days I dream of the original static Yahoo Index…ahh…1994