Why your SEO science lacks credibility

I often type in random queries when I’m looking for good topics to write about. Today one of the queries I tested was “seo science”. A few people do search for “seo science”, although I am not sure of who or why. Perhaps only the people hoping to rank for “seo science” are searching for the term.

Nearly every time someone takes a swipe at search engine optimization the SEO community comes out of the woodwork and argues defiantly that it’s a real profession with real clients solving real problems and meeting real needs.

If nothing else, the SEO community has certainly nailed down the science of emphasis because we seem to emphasize our value more than anything else. NOTE: Emphasis is mine.

I find it odd that people frequently attack momentarily “hot” SEO buzz expressions with all the grace and skill of a drunk farmer blundering through his neighbor’s field at midnight on a cloud-covered night (no offense intended to drunk farmers).

I mean, there are Web sites trying to rank for expressions like “SEO theory”, “SEO science”, “SEO art”, and other SEO-something expressions with rather weak, pathetic-looking copy and a dearth of links. I’m not sure what these experts are trying to prove, but they seem to prove only that they cannot spell “google” (as opposed to, say, “goolge”) and that they cannot obtain any links to their articles.

Several people seem to be promoting their businesses as “SEO science” but I’m not sure if anyone can actually claim a trademark for that name (it is rather generic). Certainly if you’re going to try to claim a trademark for “SEO” you should not be using documents that mention performing SEO services for domains at least 7 months prior to those domains being registered for the first time.

However, the whole “is SEO scientific” debate has been waged across the Web in a multitude of articles and forums. There have been no clear winners. One person insulted SEO by saying it’s not rocket science (obviously implying that anyone can do it). Danny Sullivan responded by saying that SEO IS rocket science to anyone who doesn’t know what it is (so is driving a car, by that broad measure of complexity, as well as making a colored candle).

There is plenty of layman’s science behind the search engine optimization industry’s work. NOTE: The term “layman’s science” cannot be trademarked — nor shall I permit you to trademark “layman’s SEO science” because it’s a very useful term. The SEO industry is largely built on layman’s SEO science.

Layman’s science is less scientific than the random musings of Diophanes but it is more scientific than, say, a drunken farmer stumbling blindly through his garage at night, discovering that motor oil might kill insects if they get stuck in it.

That is, there is a scientific method to layman’s science (and, hence, to layman’s SEO science) that is crude by formal scientific standards.

In other words, we share, we discuss, we analyze, and we critique. Scientists do this all the time. The chief difference between real scientists and SEO scientists, however, is that real scientists agree to be bound to a standard of quality that the SEO industry eschews.

Real scientists publish their data. They own up to their mistakes. They ask other people to probe their arguments and conclusions, looking for weakness.

In the SEO industry if you publish a report you don’t need to provide any data — and woe to the person who disagrees with you. All your toadies will leap upon that poor fool and call him names and tell him to shut up.

Frankly, that kind of unprofessional behavior doesn’t even meet the very lax standard required for layman’s science.

In other words, a fair number of people in our industry still just don’t get it. There is no proprietary data in science. Research firms protect their work and their inventions but when it comes time to be recognized in the scientific world they MUST publish their data.

If you don’t publish your data, you’re not being scientific. You have no excuses, no reasonable protections, and absolutely no right to talk about being scientific.

Now, data consists of many things but it’s not the only way to prove something. For example, suppose you want to write a proof for Fermat’s Last Theorem (which argues that, for all counting numbers or positive integers, there is no number 3rd power or greater that is the sum of two similar powers of other counting numbers). You don’t need any “data” such as meticulously observed facts to prove FLT. You just need logic.

But in order to be credited with solving FLT (and believe it or not there is still room to claim fame in that field) you must publish your logical argument. You have to show people the steps to follow in order to start with the resolution and work their way through its proof.

This has been done once so far in the more than 300 years since Pierre de Fermat wrote a cryptic comment in the margin of a book about Diophantine equations (equations in the form of X^m + Y^m = Z^m). Andrew Wiles achieved world-wide recognition for connecting all the dots in a very complex and esoteric branch of mathematics.

Maybe one day someone will figure out a way to show that Fermat was right so that the rest of us can understand how.

As it is in Mathematics and Economics and many other disciplines, much of the science behind search engine optimization can be articulated through models and proofs. You don’t need to publish data because any specific data is irrelevant. That’s a mathematical concept many students struggle to appreciate.

In other words, in formal logic you don’t “plug in the numbers”. You work with axioms, laws, rules, principles, and stay within the boundaries of the system you’re describing. Let A be the set of all scientific principles of SEO. Let p be any member of that set. If p is scientific, then it follows that there is a set of logical statements that explain what p is and how p works.

That is science, plain and simple. It’s just not all there is to science.

Theory in itself is not science. Theory is a formal articulation of what science has learned. Even today, in January 2010, there are people who seek to discredit SEO theory by suggesting that SEO is more an art or that SEO theory is somehow inspecific, inconclusive, or inaccurate.

You can write obscurely or you can pose a completely invalid hypothesis. Science is filled with obscure statements and invalid hypotheses. That doesn’t make science less scientific. It just means that we still have more to learn.

Science is the organized body of human knowledge, together with the principles and methods for observing and understanding the natural and unnatural phenomena we are capable of perceiving. Our science is in its infancy and it has so much more growth ahead of it.

But there are people who simply don’t understand the science behind the work they do, and they seek to discredit the science as if it is put forth by “the wrong people”. There are no right or wrong people. The only way to not participate in the scientific discussion behind search engine optimization is to say nothing.

Clearly, many people in our industry say much, though much of what is said seems rather pointless and worthless. And yet even the pointless, worthless comments contribute something to our knowledge of the searchable Web ecosystem.

SEO science does not limit itself to what publishers do. Search engineers and search scientists also seek to optimize search, and their work is generally assumed to be more scientific than the publishers’ work. Searchers, too, can be scientific in their approaches (in fact, there is a fair amount of scientific literature that discusses search from the searchers’ perspective).

The truth about SEO science is that it is neither art nor rocket science. It simply is what we know and how we study and learn about search engine optimization as publishers, indexers, and searchers.

Attacking the credibility of SEO science because you’re not publishing it on your blog is self-defeating. You cannot make yourself look better by implying that someone else looks bad.

Either put forth a proposition that is supported by logic or publish the data upon which you are basing your conclusions. Telling people how you conduct a test is insufficient to prove anything (although disclosure of testing methods can certainly help us improve them).

If you’re going to make an assertion, you need to back it up with clear proofs, not claims that you have proprietary proofs. There are no such things as proprietary proofs.

But you also need to make sensible assertions. For example, the assertion that “PageRank sculpting works” is complete nonsense. Why? Because by itself it doesn’t mean anything. What should it mean? Are there PageRank wheels and gears that whir and buzz, thus showing that work is being done? That’s more of a Talmudic issue than a scientific one.

My point is that if you want to make some assertions about specific SEO practices your statements need to be more robust and complete. Too many claims put forth by the SEO community rely upon a communal understanding of the context.

You and I, as SEO practitioners, understand that “PageRank sculpting works” is intended to mean that by restricting the flow of PageRank throughout a Website some sort of desired benefit is obtained. But these assertions fail to articulate what the desired benefit may be.

In some cases we’ve seen clarifying statements, such as, “By sculpting PageRank we were able to get more pages indexed.” That is an assertion which can be substantiated only by quantification (publishing of data), not by logic.

Another clarifying statement we’ve seen is “PageRank sculpting directs the flow of PageRank toward the most important pages of a site.” That is an assertion which can be substantiated by both quantification and logic. It can be substantiated by logic in which an argument accurately describes the flow of PageRank through a Website and the rules by which that PageRank flows.

To date, no one in the SEO industry has published a demonstrably accurate description of how PageRank flows through a Website and the rules by which that PageRank flows.

Yes, there have been quite a few really popular articles that explain how PageRank works (more-or-less). But popularity in no way establishes accuracy. Only Google can verify any particular description, and I’m not aware of their having authenticated any SEO depictions of the inner workings of one of their most secret algorithmic factors.

It is impossible for people in the SEO community to deduce the rules by which PageRank flows through a Website. You can use 10,000 Websites and track 100,000 queries and do no better than someone who uses 1 Website and tracks 1 query. All claims to the contrary are unsubstantiated.

In theory we could do it if we had an accurate method for tracking and measuring PageRank — an internal value that is not published. Remember, even if there is a direct correlation between Toolbar PR and internal PageRank, two very separate internal values might be adjusted to a TB PR 4. Hence, there is no way to use Toolbar PR to track and measure internal PageRank.

Recent attempts to belittle SEO theory and approximate SEO science have left their mark on the field. But the science moves forward regardless of who is trying to cast doubt upon it. The pace is incredibly slow but the sharing and the discussion continue on a broad scale.

Some people have claimed to be purposely distributing misinformation in forums and blog comments in some misguided belief that they will protect their competitive advantage by doing so. It’s true that I and others have argued in favor of the “STFU” philosophy, but that is a discussion about methods (and in some cases ethics), not about the science.

The science is not based in whether you build links or write copy.

The science is not based in whether you violate guidelines or honor all expectations.

The science is based solely and completely upon what we do share that is confirmable and reliable. Everything else is just talk.

You don’t have to have a Ph.D. in anything to practice layman’s science. Formal science grew out of layman’s science. Layman’s science is what our ancestors used to plant the first domesticated fig trees in the dirt. Layman’s science is what our children will use to understand the next communications technology that they adopt.

Layman’s SEO science is no more sophisticated than that. It remains rocket science to anyone who doesn’t know what it is, but it’s not beyond the reach of the average person — or the average SEO.

We cannot bottle up the science behind SEO; we cannot own it; we cannot prevent other people from sharing it or learning from it. Science doesn’t guarantee that only right conclusions and methods will be used; it merely provides us with the means of finding the right conclusions and methods.

We may have to test a few bad ideas before we get to the next good one, but that’s okay.

That is what really makes science credible.