Developing a Scheme for Keyword Prioritization

This entry is part of an ongo­ing series on how to iden­tify the best key­words to help your tar­get cus­tomers find your com­pany online.

How to define a keyword prioritization scheme

The prioritization scheme can be the most contentious step in the keyword research process, as it has far-reaching consequences for keyword output and adoption. Therefore, be sure to work closely with the project sponsor to develop the scheme, and seek out direct input from other team members as well.

The first step is to understand more deeply the intended use of the output. Understanding how a keyword will be used will help you see what makes a keyword potentially “better” for that use, or what makes a keyword “worse off” for the same use. For example:

• If the keywords are meant to be used in an upcoming pay-per-click campaign with the goals of lowering conversion cost while maintaining total conversions, then average cost-per-click is a direct indicator of the conversion cost of the keyword. The number of searches done will matter as well, as keywords with very few searchers are not likely to contribute to ensuring that the total number of conversions do not drop.

• If the keywords are meant to be used in a website SEO makeover, then the current website’s Google organic search ranking position for that keyword will become a factor.

When considering a keyword, evaluate its potential by examining three main categories:

1. Search Volume, or the number of searches being done on a particular keyword over a given period, which indicates the potential volume of search-originated traffic for that keyword.
2. Competition, a measure of the difficulty to generate a unit of search-originated traffic for that particular keyword.
3. Audience Relevance, a measure of the quality of the traffic generated by that particular keyword.

Typically, a keyword ranks higher if it has greater search volume, lowercompetition, and highaudience relevance.

A prioritization scheme takes into account these factors in one form or another, but this combination can also take several forms. For example, many SEO specialists use the Keyword Effectiveness Index (KEI) to prioritize keywords. The definition of the KEI is as follows:

KEI = Number of Monthly Keyword Searches/Number of Pages in Google Results

Here, the ranking formula employs two types of factors: Search Volume and Competition.

The logic behind this is simple: If a keyword is searched more often (higher number of monthly searches), then there is more chance of getting searched and found through Google. On the other hand, the more pages in Google Results, the more “competition” there is for a web page to stand out and be found among those. This inverted relationship can translate directly into the formula above.

It is clearly not a magic formula that can be used in any situation. It is not meant to give an absolute measure of “goodness” of the keyword. In fact, the KEI only helps with ranking the keywords relatively to each other, and it also considers them equal in attracting visitors of the same relevance, while in reality, this is not the case.

Other prioritization schemes might not be based on a formula at all, but on a set of conditions and ranking criteria. These are more appropriate when non-numerical values are considered, such as when the Keyword Type and Source are taken into account when prioritizing.

The following is a list of typical metrics and factor types used most commonly in keyword prioritization (note that the last two metrics might be indicative of two factors at the same time, and should be used with caution):

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Additional resources on keyword metrics for prioritization:

How to take the guesswork out of keyword selection (Web Pro News)

Selecting the best keywords: A poisson regression model (Journal of Interactive Advertising)

Factors to consider when selecting keywords (Helium)

Stay tuned for more infor­ma­tion on find­ing the best key­words for your tar­get audi­ence. Next week I will look at how to collect your data and apply it to the prioritization scheme.