What do SEO’s Know?

Niels Bohr said, “An expert is a person who has made all the mistakes that can be made in a very narrow field.” I feel that this also applies to the field of search engine optimisation also. It is not an exact science. At best, it can be compared to economics or sociology. Data is noisy, there are lots of theories and everyone has their own opinion. It may be worse, since unlike economics or sociology, how search engines work can be changed without notice.

While you cannot argue with results, there are many roads to Mecca. So how confident can you be that the methods you apply are actually having the effect you expect? You must care because the next time you apply changes to a website you will want them to be successful.

How is Knowledge Built Up?

Common sense points to things which are likely to be correct and rejects many more things which are obviously not correct. I am reminded of an episode of the Young Ones. They have a video recorder they cannot get to work:

MIKE: Must have gone wrong somewhere.

VYVYAN: Oh God!

MIKE: Maybe you shouldn’t have poured all of that washing-up liquid in it.

VYVYAN: But it says here, Michael look, “Ensure machine is clean, and free from dust”!

MIKE: Yeah, but it don’t say, “Ensure the machine is full of washing-up liquid”!

VYVYAN: No, but it doesn’t say, “Ensure the machine isn’t full of washing-up liquid”!

MIKE: Well, it wouldn’t would it! I mean, it doesn’t say, “Ensure you don’t chop up your video machine with an axe, put all the bits in a plastic bag, and bung ‘em down the lavatory”!

VYVYAN: Doesn’t it? Well maybe that’s what’s going wrong!

In this vain, there are parts of search engine optimisation which can be guessed as right because if you were designing a search engine it is what you might choose. You would not design a search to rank untitled pages, that were poorly related to queries, badly coded and did not link to other pages and a special rule which ranks your page at the top if you add 39 e’s in a row in red. Things such as add a title to the page, text content, links to other pages and so on.

Papers and Patents

This gives information about what is possible and may be implemented in information retrieval and computer science. PageRank for example is known about because of a paper published when Google was a research project. Today we would not get to see such insights into the working of the search giant. Even when you know how something works you are still don’t know how important it is in ranking the search results. There are many factors and these can be modified by parameters to add weight or take it away.
Patents can provide information on what company’s are thinking about introducing but just because a company has a patent, it doesn’t mean it will use it.

SEO Experiments

Very difficult to get results that say anything about the algorithms because there are so many factors too much noise and the results are not reproducible.

Dissemination of Information on the Web

A good source of basic information comes directly from the sources the Google webmaster blog posts basic information which will make it clear to anyone publishing a web site what will make it easier for their search engine to visit. Keeping track of what is said in the blogs when Google employees such as Matt Cuts and Marissa Meyer are interviewed is also a good idea as they often clear up confusion on SEO matters. The problem with other sources is that unless they are reporting on scientific techniques or first hand information and even with experiments they are guessing.

Pigeons and SEO

Apart from Google using Pigeons in their search algorithm, they can also be made to perform superstitious rituals in order to gain food. An experiment by B. F. Skinner in 1948 showed how reinforcement can be used to generate superstition. At five minute intervals food would be given to a pigeon in a cage regardless of its behaviour. The pigeon would repeat the actions it was doing at the time it received the food in order to trigger the food tray. Where it was successful, i.e. happened to coincide with the food it reinforced the pigeon’s behaviour and they carried on performing the same ritual. I think that this also happens in SEO. People swear that some things will give you a better ranking without the evidence, even if the results support their idea. It worked for me before so it will work again. This is good if it works but you never know if it really is working and so you might put a lot of effort into something which is have no real effect. It becomes a superstition.

Conclusion

None of the methods is ideal and the best we can do is to look at current research in information retreival so we may have an understanding of what is possible, cautously test, measure and experiment to temper your common sense and read what other people are saying then cross your fingers and hope for the best.