3DMarketVision is market research tool that uses a variety of search engines (including but not limited to Google, MSN, and Yahoo) to find keywords relevant to a word or phrase typed in by the user. The keywords, and websites they come from, are processed using Latent Semantic Analysis (LSA) and displayed in a spreadsheet view and a 3D graphics view.
The Keyword Tab holds the spreadsheet view for keywords, as shown above, while the Sites Tab holds the spreadsheet view for competing websites. For each keyword, information includes the phrase match and exact match CPC (Cost/Click), CPD (Clicks/Day), SPM (Searches/Month), and CTR (Click Through Rate) numbers. Keyword "goodness" measures such as R/S (Results/Searches) and KEI (Keyword Efficiency Index) are included. Proprietary columns include KMV (Keyword Market Value), KMV/R (KMV/Results), TR (Theme Relevance), MSize (Market Size), and Grade (0-4 scale).
The center of the screen is devoted to the 3D graphics view. Keywords are represented by colored boxes and websites are represented by light blue pyramids. This view is a "semantic map" of the market being displayed. That is, keywords that tend to appear in the same websites are mapped near each other, and websites that tend to have the same keywords are mapped near each other. This ability lets you divide up keywords into clusters or silos, depending on how similar those keywords are to each other. In the example above, the keywords have been divided up into 8 silos, shown by the color of each keyword. This division into silos is guided by a Latent Semantic Analysis (LSA) algorithm that also determines the value for the Theme Relevance (TR) column.
The TR (Theme Relevance) column is a more detailed look into whether two keywords tend to appear in the same context or not. For example, in the above screenshot "gourmet coffee" is selected and its TR is 100% since it is exactly similar to itself, but "coffee blends" also has a TR of 99.9% meaning that "gourmet coffee" and "coffee blends" tend to appear in the same context.
Knowing which keywords are similar to each other from a LSA perspective is very useful for creating websites that rank highly for similar or "themed" keywords. These similar keywords reinforce each other and cause the webpage to rank higher in search engines. The similar keywords are also useful in developing PPC campaigns that target specific themes rather than syntactic words.
Knowing which sites are similar to each other lets you see which sites are competing with each other and what keywords they are fighting over. The strength of these websites can be assessed and a judgement made about whether to compete with them, or move to a less crowded area when building a new website or web page.