My Perspective On How The Correlation and Causation Concepts Are Correlated to SEO
Recently as per my observation and experience, I think more and more people are understanding the true meaning of optimizing a site for search engines and are preparing themselves to accept the fact that SEO is a long term proposition and an ongoing process .
The search engines will constantly upgrade and improve their algorithms and keep shifting the priorities of the ranking factors for maintaining the quality of search results. Hence keeping an eye on the shifting priorities and the algo updates is a constant R & D effort which has to be put in unconditionally by every SEO.
The correlation and the causation aspect is being discussed and talked about a lot in the SEO blogosphere and conferences but, before we start correlating these two terms with the SEO world let us understand a few definitions:
What is Correlation:
Correlation is a statistical measurement of the relationship between two variables. Possible correlations range from +1 to –1. A zero correlation indicates that there is no relationship between the variables. A correlation of –1 indicates a perfect negative correlation, meaning that as one variable goes up, the other goes down. A correlation of +1 indicates a perfect positive correlation, meaning that both variables move in the same direction together.
What is Coeffecient of Corelation and A Rank Corelation:
A coefficient is a multiplicative factor in some term of an expression (or of a series) it is a measure of the strength of the linear relationship between two variables that is defined in terms of the (sample) covariance of the variables divided by their (sample) standard deviations. An increasing rank correlation coefficient implies increasing agreement between rankings. The coefficient is inside the interval [−1, 1] and assumes the value:
- −1 if the disagreement between the two rankings is perfect; one ranking is the reverse of the other.
- 0 if the rankings are completely independent.
- 1 if the agreement between the two rankings is perfect; the two rankings are the same.
In statistics, a rank correlation is the relationship between different rankings of the same set of items. A rank correlation coefficient measures the degree of similarity between two rankings, and can be used to assess its significance. A ranking is a relationship between a set of items such that, for any two items, the first is either 'ranked higher than', 'ranked lower than' or 'ranked equal to' the second
What is Causation:
It is the act or fact of causing; the production of an effect by a cause . Basically the relationship of cause and effect, the act or process of causing, the act or agency which produces an effect.
Let us come back to the SEO world and try to correlate the importance of the above mentioned definitions for SEO.
The actual task of SEO is very difficult to define.
It is a whole package of :
- on-page optimization
- off-page optimization
- working on good quality inbound links
- creating a potential for the site so that it starts getting natural links over a period of time
- brand building
- helping the site to become more social media friendly and Social Media ready so that the site gets the links and the social acceptance with a quality buzz around it and
- last but not the least the regular monitoring and measuring of the analytics to ensure that everything is being geared in the right direction and there is constant improvement in the SEO metrics.
This regular monitoring is the stage where actually an SEO gets the idea about what is working for the site and what is going against it as every site is different and has different objectives to achieve. This monitoring also helps you analyze and study data from the R&D perspective as this regular monitoring of the relevant data gives you an idea about the correlations between what changes you made on the site and what results or effects were visible in the data set. Google analytics can give you a good idea about these correlations with the help of the custom reports feature.
Google says there is a combination of more than 200 ranking factors which determine the rankings of the search results and the priorities keep on changing hence if you make some changes on-page and off-page and you see a direct effect on the results you can know what factors are being given preference to by Google. In other words you can correlate the SEO changes and modifications made by you to the changes you see in the results as per the SEO metrics.
1) If you make the changes in the title tags of some of the pages and you see that page being ranked higher than before and the Google Analytics shows that page as the landing page for the keywords it was targeted and shows an increase in the no. of new visits and shows the source as search engines then you can be sure that there is a causation and this improvement can be correlated to the title tag changes made and one can conclude that this correlation is positive .
2) If you add a canonical tag and the those pages get deindexed and you lose the rankings then you know that this causation is due to the negative correlation between the two variables i.e the canonical tag and the loss of rankings .
The above two are very simple examples but if there are atleast 200 ranking factors then there are atleast 200 variables which can be assessed over a period of time for the various causations in the SEO metrics you want to measure.
When we say that the constant monitoring of the site is one of the major SEO deliverables, we mean that this correlation analysis of various factors on the causation of results is the main focus.
Data in itself does not have any value unless it is analyzed and understood so that corrective actions can be taken, otherwise it is like a car without a driver. You can use the car efficiently only if you have a good driver. Similarly you can make use of the Google Analytics data if you have a good analyst who can correlate between the metrics and understand the effect on the result oriented metrics.
Some points to ponder on this concept of correlation and causation are:
- Almost all SEO is correlative rather than causative.
- Moreover , a correlation need not be causative.
- You need a sizable chunk of data over a period of time (atleast 4-6 months) in order to establish any connect between the correlation and causation to come to any conclusion.
- This correlation analysis requires a lot of time, effort and an eye for detail by an expert
- Jumping to quick conclusions will just cause more confusion
When we say SEO is an ongoing process this is one of the major SEO tasks which need to be undertaken regularly by every SEO for every site as every site is different and needs to be analyzed and studied as per what metrics matter the most for that site.
SEO is becoming more and more specialized and the whole SEO task needs to be conducted by various experts to get the most out of every SEO campaign. Ideally if a SEO company needs to take more no. of SEO projects and do justice to the SEO campaign then the SEO team should comprise of experts in the following fields:
- Keyword Research
- On-Page SEO
- Off-Page SEO
- Natural Link Building Engineers
- Social Media Strategists
- Research and Development Analysts whose job is to keep an eye on what changes are being reported by Google on their various blogs and other authority sites
- Coordinators who can convey the SEO process and progress to the clients keeping the training aspect in mind so that every meeting with the client enriches the client with the SEO relevant knowledge in a very simple way so that he is convinced to renew the contract year after year.