Gravatar
Friday, March 27th, 2009

Web Position Gold and the Keyword Visibility Index

 

As many of our readers know, Web Position Gold (WPG) is a tool used to check search engine rankings. It does other things, but its reporting feature is all we use. It does a good job of managing multiple accounts and reporting rankings compared to many other choices on the market.

Some SEOs believe that position reports are useless, and while I understand their arguments, there is no denying that clients want to see this information. I am not going to argue the pros and cons about position reporting; this post is not about that. Here I am going to explain a specific chart that appears on the WPG reports, and my advice to WPG on how to improve on it.

One of the most common questions clients ask about position reports is the purpose of the “visibility index” graph that WPG displays, and how it is calculated.

How the Keyword Visibility Index Works
Essentially each ranking position is given a score, with the values increasing as you approach the #1 position. For instance, a #1 ranking gets a score of 30, with a #2 ranking getting a score of 29, and so on down to a score of 1 for the 30th position. If you have two phrases being checked in one search engine, one with a ranking in the #1 position and the other ranking in the #10 position, the score would be 30 + 21 = 51. (See Figure 1.0 below)

WPG calculates the maximum possible score (all phrases ranking #1 in all engines) divided by the actual score, to get a percentage. 100% would represent all phrases ranking #1. In the example above, two phrases in one engine would have a maximum possible score of 60, so the score would be 51/60, or 85%.

Web Position Gold fig 1

Generally, the closer your score is to 100%, the better your rankings. The problem with this is that there is one major hole that renders this visibility index nearly useless, and that is not being able to factor in the individual importance of each specific target phrase.

Fundamental Error with the Keyword Visibility Index
(This is where things get a little more complicated)
Let’s look at Figures 1.0 & 1.1 above. In Figure 1.0 the score is 85%. Now assume that PhraseA is very important, and PhraseB is of significantly lesser importance. If later we run a new report and see that the ranking for PhraseA has dropped down to #9, and PhraseB has jumped up to #1. The end result is a higher visibility score, up now to 87%. The end score reflects a positive change, but if PhraseA is significantly more important, these new rankings are actually worse.

The same can happen in reverse. Good phrases can increase in ranking while bad ones drop, resulting in a reduced score, where in fact the rankings actually are more favorable.

While an increase in this score is generally a good thing, not enough information is used to determine at a glance, if the campaign results have actually improved.

Potential Fix
There are likely numerous ways to correct this, but one idea stands out in my mind. By introducing a variable into the algorithm that determines the importance of each individual keyword, this could solve the problem. End users could enter in a value between 1 and 10, 10 being most important, to indicate how critical each keyword phrase is. By dividing the user entered score by 10, a multiplier between 0 and 1 could be generated behind the scenes.

If a user gives a value of 10 for a specific phrase, then that phrase will retain its full value as calculated by WPG. If they enter a 5/10, then the end score for that phrase will be multiplied by 0.5 (or halved). Phrases with a value of 0, would not contribute to the end score at all.

By doing this, the end result would be much more reflective of the actual rankings. As you can see in Figures 2.0 and 2.1 below, when the important PhraseA drops, and the less important PhraseB increases, the total score also drops, indicating that overall campaign is not doing as well. With this change users could, at a glance, have a better impression of how their campaign is doing with the comfort of knowing there is an extra level of accuracy.

Web Position Gold fig 2

I want to point out that I am not a programmer, or a mathematician. I would not be surprised if there were holes in my potential fix; however, I am certain, that some kind of variable to indicate importance would make WPG’s Keyword Visibility Index much more useful. Another, much more complicated method could be to have WPG automatically grab the search frequencies for each phrase and sort according, then apply importance values behind the scenes, or perhaps even a combination of this and my user entered idea. There may be even other better ways. This post is merely a suggestion to WPG on one way to improve their product.

I welcome the feedback of any WPG staff or users to chime in on what they think, so be sure to leave a comment. I will also happily accept any royalty checks should WPG successfully use my idea. :-)


4 Responses to “Web Position Gold and the Keyword Visibility Index”

  1. aimClear

    I gave up on WP reacting to personalized search a while ago. The only true measure of organic prominence is percentage of market share WW inventory (monthly searches) / traffic.

    http://www.aimclearblog.com/2007/06/11/google-makes-it-official-webposition-gold-is-dead/

    I appreciate the post and bless you it was great when we could actually measure organic prominence by this method. Those days are over :)

  2. Scott Van Achte

    I agree with you that the usefulness of these position reports has declined, however, even with personalized search, most rankings will be very similar, especially if the logged in user has not done a that same search before. These reports are still a good general guide to see how an optimization campaign is coming along.

    On that same note, I would love to see what percentage of searchers are actually logged in, I am willing to bet that they are still in the minority.

  3. Alissa M.

    I understand that in a perfect world I should be shooting for 100% every month, but realistically, where should I expect this number to be?

  4. Ross Dunn

    Honestly Alissa, I wouldn’t pay attention to it at all. Focus more on your analytics reports and watch how your traffic ebbs and flows and how well visitors convert on your website. If you focused more on the visitors and how to convert the ones you received already you would probably be a lot more successful. This is especially true since today’s web ranking reporting does not do a good job of providing ranking results from other locales; an important statistic since your ranking is likely different depending on where someone searches from and their browsing history (which is impossible to duplicate).

Leave a Reply