Online SEO Information: Trick or Treat?

When I started in this industry back in the day, successful SEO (define) techniques were proven with anecdotal evidence. Although the engines are getting better at their job in many cases, much online info still relies on anecdotal evidence.

Don’t get me wrong. There’s some great info out there provided by true professionals. There’s a whole lot of badly researched and speculative nonsense, too.

I’m coming to the research cut-off point for the third edition of my SEM (define) book. The interesting change I can see for the next edition is it’s more about “the application of marketing communications to information retrieval (IR) on the web” and less about age-old technical SEO issues.

The above quote isn’t a snappy tagline, but it’s what I actually do for a living: apply my background in marketing communications to the science of IR on the Web. And what I’ve learned about IR has helped me dispel so many myths, many of which still abound in the industry.

I have to admit, I was sucked into some of those myths during the ’90s. Then, a contact at a (long-gone) search engine recommended I read “Introduction to Modern Information Retrieval” by Gerard Salton. Though it’s been out of print for a long time, you may gasp at the price people are still willing to pay for it (often over $400). It was also recommended I read “Modern Information Retrieval.” Fortunately, this one’s still in print and sells for around $43 new. I began to get a better grasp of the concepts, fundamental principles, and drivers behind IR. Since then, I’ve been fortunate to make acquaintances with many scientists and researchers in the field.

People come to SEO from other disciplines, such as development and design. Many think there are some magic tags you stick in pages to make them fly up search engines’ results pages. They ask me about coding: can search engines read it, and where’s the best place to put this or that tag for better ranking? They appear at a loss when I explain coding’s just not that important, as long as a crawler can get to the pages and parse the text out of them.

Then there’s the old “keyword density” chestnut.

I’m amazed at the number of potential clients who harp on about their previous or existing vendor who has “magic” software that downloads thousands of pages and analyzes them to reverse-engineer the algorithm. It’s not possible to reverse-engineer anything, unless you know what all the components are. And with search engine algorithms, so much is not accessible.

Do a little research into IR methods, such as the vector space model and methods of term weighting. You’ll quickly figure out why measuring so-called keyword density is nonsense. My good friend Dr. Edel Garcia wrote just that in an article on the subject.

The great thing about Garcia is he passionately wants to bring search marketers closer to the real world of IR and away from the many myths. He’s got a new tutorial on term vector theory, and his site has a ton of information to help bring search marketers up to speed on need-to-know stuff.

If you’re new to IR, some of it may seem a little daunting. Yet Garcia writes in a very clear, simple manner. You’ll certainly be able to grasp the concepts, if not the math.

More than anything else, perhaps, you’ll realize there’s more to text analysis than the number of times a word appears on a page. And that’s basically all a crude keyword-density analyzer shows you.

As Garcia explained, “Neither term weight nor keyword density scores are measures of term importance, which can only be assessed through semantics, contextuality, and on-topic analysis.”

When I studied conventional marketing in college, I was taken into the realms of economics, psychology, strategy, creativity, and other disciplines that come together to form marketing. The same applies online. We don’t all need to become IR scientists. Yet IR is extremely important among the many disciplines that form our own branch of marketing. The better our knowledge of the underlying principles search engine algorithms are based on, the better prepared we are to act on behalf of our clients.

Keyword-density software or scientific evidence? Give me scientific evidence any day. I’d rather have the treat, not the trick.

Join us at Search Engine Strategies in Chicago, December 5-8, 2005.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.

Related reading

Brand Top Level Domains