Traditional eye tracking is costly, takes awhile, and is pretty resource intensive. To fill that gap many companies have started offering little snippets of code that you can insert into your web site, to track your users mouse tracking. While valuable to a certain degree in tracking where people click, it is not a resource in terms of analyzing true Information Architecture. You end up getting heat map eye candy, which decision makers, management can drool over and proceed to “make decisions”. Here is where the problem lies.
Each of these mouse tracking software providers push metrics in front of you about eye and mouse correlation. Correlating mouse tracking to eye tracking is like farting. It is unpleasant and messy. You may think you discovered the source of the problem, but end up blaming the wrong thing.
When someone farts – a cloud of stink emanates out from its point of origin in lesser density (think heat map). You know around where it came from; yet if there is someone sitting close by, they may be blamed for the fart! Uh oh, you just blamed the wrong person for the fart.
This is exactly what happens when you start thinking that there is an important correlation between the mouse and eye tracking. Nielsen Norman Group even found that people will use their mouse to select some text to make it more readable because the IA is terrible on a page. What is a mouse tracker going to tell you about that? That a giant block of text is where people's cursor is going to loiter?
Mouse tracking software to track where people click – pure gold. Mouse tracking software trying to figure out what people look at, total shit. Sometimes it even smells that way.