Eye tracking vs. mouse tracking for usability testing and site optimisation, when to use which method?

We’ve written about this before but Ian just forwarded me the below article on the ClickTale blog about eye tracking vs. mouse tracking and I think it’s a very nice comparison.
Eye tracking, as used by top enterprises such as Google, uses cameras and specialist software to track where the eyes of internet users land on a webpage. Mouse tracking follows the mouse movements of an internet user to simulate eye movement on a webpage. Over the last few years, mouse tracking has greatly matured, developing features and achieving accuracy that make it a credible alternative to eye tracking.
Mouse-or-eye-tracking

Heatmaps created using traditional eye tracking (left) and mouse tracking (right)

Research has shown that when both methods of testing are conducted simultaneously, there is an 84%-88% correlation in the results. In addition, both the eye and mouse move to relatively the same rhythm and focus in on the same page content. 
Now you could argue that the above heat maps actually don’t really look like they’re 80% the same (and I would have to agree) but this is more a case of cost vs. benefit. Eye tracking seems to be more accurate but also much more expensive compared to mouse tracking so maybe mouse tracking could be a good initial first step to inform a potential later eye tracking study.