Why SEO collective wisdom lacks credibility

We have no authoritative voices in the SEO community. We all lack proper credentials, formal training, and credible validation for our expertise. Search engine optimization itself is still chaotically defined according to whims and personal preferences. What is the meaing of “search engine marketing”, for example? I’ve always preferred to associate it with pay-per-click advertising but too many people still associate it with so-called organic SEO for that association to be authoritative.

It’s perfectly possible and normal in the English language (and many others, I suppose) for an expression to have more than one meaning or association, so we’re not violating any rules of linguistics by confusing our definitions. The lack of consensus on what search engine optimization and search engine marketing should mean, do mean, may mean, cannot mean, and will eventually be used for really doesn’t matter to the average person.

As long as you can explain what you’re referring to when someone asks, you should be okay. But if you’re going to claim to be an expert or a leader in the field (something I do from time to time) you’re walking on pretty thin ice. There are thought leaders in the field but are there credible knowledge leaders in the field? I would say we only have thought leaders. We haev no credible knowledge leaders.

Search engine optimization is not a science, and if ever a formal discipline grows up around this topic the odds that the theorems and definitions I have proposed on this blog will carry much weight in the academic literature are pretty long. Academics do, desipte my occasional criticisms, scour the Web for grass-roots data, raw social expression of emerging concepts. My science fiction and fantasy Web activities have been mentioned in several academic studies, for example, and I know many other similar “hobbyist” sites have been profiled, surveyed, documented, critiqued, and analyzed through formal analysis.

Google’s Web Authoring Stats study is one of the most profound and comprehensive reports on how people design Web sites. It’s a very useful tool for people who want to understand why Web authoring standards should not govern search engine algorithms. It’s also a rare street-level glimpse into how search engineers think.

But there are no formal studies that open the doors on search engine optimization. There are probably 200-300 academic papers (at this time) that attempt to define Web spam and prescribe methods for combatting it. Perhaps more than I have found. But most of these attempts sadly lack an adequate knowledge or understanding of what Web spam is or how it evolves in tandem with search engine results and algorithms.

Web spam is usually the pioneering tool that shapes search engine optimization theory and practice. “Best practices” SEO and “darkside SEO” often differ only by a matter of excess. In fact, so-called White Hat SEO practices usually develop as search engines react to new methods of influencing their results. If search engines provide a favorable response, the White Hats adopt the practice; otherwise, the White Hats avoid it.

SEO Industry attempts to document practices usually fall short of the mark by a wide margin. Surveys are naturally limited by the surveyers’ preferences, goals, and knowledge. They also tend to focus more on economics (how much do you make, how many people do you employ, how much do you charge, how large an advertising budget do you handle per client). The vast majority of SEO practices are left unstudied, undocumented, and the data collection processes most preferred in these types of studies are inadequate anyway.

SEOs like to play with polls. They’ll put up polls to make their sites more sticky, to infuse ther readers with new interest, to look authoritative to prospects and clients, and to encourage their online friends to link to their sites. Polls and surveys really don’t tell us anything useful because their respondents are random or closely-knit elite groups.

But even if someone wrote up a survey that asked a mix of structured questions (Do you place links to sites in directories? Do you use software to place links in forums, blogs, and social media sites?) and open-ended questions (How do you place links? How much do you rely on links for your optimization?), and even if that person could entice 1,000-2,000 responses from bona fide real SEOs (White Hats and Black Hats), such a survey would be inadequate unless it was coupled with a selection of case studies that tracked Web sites through search results.

The real science behind search optimization — when it is developed — will focus not on people’s practices and opinions but rather on the methods and means of influencing search results. Asking someone if they use link placing software isn’t the same as watching people do this in the field. Asking someone what they think is the most important factor in search results scoring isn’t the same as analyzing their actual optimization work.

Do people who claim that the title tag is the most important element in optimization always inject keywords into theri title tags? If so, how do you compensate for the inherent bias in their activity in order to understand if their opinion is based on their results or if their results are based on their opinion?

That is, the only way to objectively determine if the opinion is driving the optimization is to find out if the optimizers do everything else that is possible. If they ignore other opportunities for optimization then their optimization is based on their opinion, rather than their opinion being based on their optimization.

But that kind of distinction alone only looks at one aspect of search engine optimization: practice, not the cause-and-effect mechanisms that SEOs are constantly seeking. Proving cause-and-effect in a complex system is not as easy to do as today’s Web marketers would have us believe.

It’s not enough to say you optimized 100, 1000, or 10,000 clients’ sites using your proprietary methods. You have to look at all the factors involved and people who have served 10,000 clients do NOT look at all the factors involved (it would be humanly impossible to do so).

Data aggregation shows us patterns that we cannot see in the minute details. That is why I have often — when people reported strange happenings in search engines — asked people to show me their Web sites and queries in private. My own network and query base is quite large, but looking at portions of the Web I don’t normally analyze helps me see patterns more quickly. Also, I practice a fairly uniform optimization strategy and there is more than one way to optimize for search.

In fact, that we can say “there is more than one way to optimize for search” underscores just how much we underutilize the full range of potential optimization elements. It could be argued that some people have apparently experienced penalties for “over-optimization” but in my (limited) experience such sites which I have examined engaged in what would be more aptly described as “an over abundance of hyper optimization”. They opened the flood gates and produced thousands, sometimes hundreds of thousands of pages of “over-optimized” content. The average site is less likely to trip such filters.

Volume itself is a factor in both search indexing and search optimization, and it’s a factor that is rarely taken into consideration. Our confusion over standards, practices, methods, and metrics ensures that we not only optimize with blinders on, we analyze with blinders on. The only way we can improve the quality of our collective knowledge is to share it openly and uniformly.

However, our industry is not set up to share information either openly or uniformly. People share linking tips and basic HTML optimization tips on a daily basis but that covers only about 20% of what a robust search optimization campaign deals with. I rarely find find acceptable tutorials on query analysis and keyword research, and all the good ones I have read are way out of date.

Non-disclosure and competitive pressure make it virtually impossible for us to review each other’s work. Hence, only random chance provides opportunities for people to hear feedback or learn about deficiencies in their execution of methods. In order to bring out the hidden data, the undisclosed ideas, the experimental measurements that may give rise to useful metrics, we need to build accountability into our communal processes.

There is no accountability today. Even professional organizations are only accountable to their memberships. The SEO industry needs a benevolent representative body that doesn’t rely on memberships. Furthermore, we need a truly professional journal that accepts only peer-reviewed documents. But who should be the peers? Who should be the editors? Who can afford to take the step to commit time and resources to do this?

We lack the economic foundation to provide for our own objective internal review structures, without which we remain saddled with a mishmash of anecdotes, opinions, hogwash, and nonsense.

So where do we go from here?