In the last tutorial, we filtered keywords based on Latent Semantic Analysis (LSA) and on the root words that the keywords contain. In this tutorial we will filter keywords based on data that we collect about each keyword.
Now let’s gather more data about our remaining keywords. To do this click, on the Select/UnSelect All button in the keywords tab area, or simply use the control-K shortcut key. This will select all the keywords, and also change keywords in the 3D view area to yellow to show that they are selected. The pyramids in the 3D view area will not turn yellow, because they represent websites, not keywords.
Once all the keywords are selected, click the Get More Data button in the keywords tab area. If you have not already entered your Google Adwords e-mail and password you will be requested to do so and, after that, click the Get More Data button again. While 3DMarketVision is gathering the keyword information, it notifies you of its progress in the status bar at the bottom of the program. When 3DMarketVision has finished, click the Select/UnSelect All button again to make sure that all keywords are unselected.
Now click the keyword market value (KMV) column header and sort by KMV in ascending order, so that keywords with the smallest KMV values are on top. The KMV column measures how much it would cost to monopolize all the organic searches for that keyword for a year, based on an equivalent pay per click campaign. It is based on the phrase match searches per month (PhSPM column) and the phrase match cost per click (PhCPC column). The higher the numbers in these two columns, the higher the keyword market value will be.
Select the first keyword which should have the lowest KMV value (usually this is $0). Scroll down the keyword list until you find the last keyword that has a KMV value $0. Then shift-click the last keyword with a KMV of $0, so that all the keywords with a KMV of $0 are now selected. Finally, click the Delete/UnDelete button so that those keywords disappear.
In the keywords tab area, click the Select/Unselect All button to select all the 67 remaining keywords. Then click the Get Final Data button, also in the keywords tab area. 3DMarketVision will now go out to the web, and for each keyword, find the number of sites with that keyword in the title. For example if the keyword is “ipod speakers” then 3DMarketVision will find the number of websites whose title includes “ipod speakers” and display the number in the Sites column.
After the Sites column has been populated for all keywords, additional columns are calculated, including the three keyword efficiency columns (R/S, KEI, and KMV/R), the market size (MSize) column, and the Grade column. Of these columns, we will be concentrating on the KMV/R column. This column measures the keyword market value (KMV column) divided by the number of sites (Sites column) optimized for that keyword. The number of sites optimized for that keyword is approximated by the number of websites with that keyword in the title, since one of the first steps in optimizing a site for a keyword is to put the keyword in the title.
While in the keywords tab area, click on the KMV/R column to sort the column in ascending order, so low values appear first. Now select the first keyword, which has a $0 KMV/R number, so that a checkmark appears in front of the keyword. Scroll down the keywords until you see the last keyword that has a $0 KMV/R number. Hold down the shift key and click this last keyword, and all the keywords from the first keyword down to this one should be selected.
Now click the Delete/UnDelete Keywords button and all the selected keywords, i.e. those keywords with a $0 KMV/R value, will be deleted. Only 46 keywords remain after this filtering step.
During this part of the tutorial, we have reduced the number of keywords from 137 down to 46 by gathering more information about the keywords and using our proprietary keyword columns KMV and KMV/R. In the next tutorial we shall see how to divide these keywords into tightly clustered silos, using the theme relevance or TR column.
