The Myths and Math of SEO

Just a quick note before we get into the meat of today’s column. I wanted to mention the general response to my campaign to take the “engine” out of “search engine marketing,” from two columns ago.

There’s no doubt many people want to see “engine” disappear and stick with “search marketing.” Looking down the long tail of my responses, however, it’s not hard to see many people would like to see even “search” go and just stick with a more rounded marketing description. Interactive, online, Web, or whatever precedes it — but more so, just emphasize the marketing.

The debate is open and live, please send your comments.

Myths and Math

If you have a formal background in marketing, you’ll have a deep knowledge of the mechanisms that underlie the marketing plan based on both quantitative and qualitative research.

As business people, we can always be certain of one thing: uncertainty. This is why empirical evidence in business (and, in this case, SEO) counts for so much more than anecdotal hoo-hah.

When a little math is applied to marketing, a myriad of data become available and useful. I’m not a mathematician. I pointed this out in the second edition of my book and hid myself neatly under a wonderful quote from a mathematicians’ mathematician:
When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of science. —Lord KelvinConsidering I’m not a math guy, I surprise myself at the amount of time I spend reading research papers. They don’t count for a great deal unless you can make some sense of them. Yet when it comes to the math behind algorithms developed specifically for Web information retrieval, there seems to be a lot of BS and hearsay that prevails online. Which is a shame, because the truth is out there. If you can be bothered to find it.

This is why when pitching for business and confronted with the usual “My current/previous vendor said blah, blah, blah,” my answer is fairly simple. You should never accept the “we just make it work” routine. Because when it stops working, it’s too late to ask your SEO firm if it really knows how it works.

I’m all for getting real information about how search engines work into the public domain. And that’s why I love it when my friend and colleague Dr. Edel Garcia makes common sense out of more complicated research papers and produces his excellent tutorials, including the three new tutorials on matrices, Eigenvalues, and Eigenvectors.

Recently, I conducted a quick interview with Garcia. I spoke to him on his background and emergence in the SEO field. I also asked him about his thoughts on the level of knowledge relating to information retrieval within the SEO community when he first started observing.

“Pure nonsense,” he said. “However, in the interest of fairness I could say the same about the wrong impression many CS scientists have about SEOs and SEMs, labeling them as a bunch of spammers.”

Garcia coined a few phrases on his blog that recently put a grin on my face. “Blogonomies,” “blogorrhea,” and “linkphilis” are new terms to the industry. I asked him to explain them.

“These are social behaviors worthy of study,” Garcia responded. “I define a ‘blogonomy’ as the dissemination of false knowledge through electronic forums and ‘blogorrhea’ when the dissemination is intentionally done for a profit or commercial interests. These behaviors are prevalent in the blogosphere.” Hence, the “blogo-” stem.

He continued, “Professor Jon Klienberg has researched the concept of ‘burst’ in the blogosphere. Along that line, while not a requirement, a ‘blogorrhea’ outbreak can be observed when a blogospheric burst is the result of a blogonomy.

“Spreading a blogonomy by means of not using a ‘link condom’ leads to ‘linkphilis,'” said Garcia. “Often the spreading agents are link spammers or someone posing as a respected authority. Linkphilis is a condition involving links pointing to documents with corrupted knowledge, like blogonomies. Thus, weight transmitted through these links can be considered infected weights. Note that here we are not talking about mere link farms or off-topic content, but about the spreading of false knowledge and a special kind of burst. Most link models score weights but overlook this problem.”

Further, Garcia said, “by a ‘link condom,’ I mean any mechanism that will prevent the transmission of weight from corrupted links. Such a mechanism is not just a mere ‘no follow’ attribute. So far the only protection is human reviewers. Tagging is not an option since this has its own ‘staggers’ [spam taggers] and ‘maggers’ [malicious taggers] to deal with. The problem is pervasive: how could one use human reviewers or link condoms in large scale databases that are continuously changing?”

There’s no doubt that Garcia has very strong opinions about both sides of the SEO industry. And his Web site is a mine of useful information for those who want to know more about what really happens under the hood of search engines.

You can find the full text of my short interview with Garcia here.

Meet Mike at Search Engine Strategies in San Jose, August 7-10, 2006, at the San Jose McEnery Convention Center.

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

Related reading

Brand Top Level Domains