I have posted to several blogs suggesting a "first stage" and "second stage" definitions for "innovation." The problem is that current definition constrains innovation to a "new idea that makes a lot of money." While obviously the desired outcome, this definition constitutes a single equation with two unknowns and is unsolvable. If we split the definitions up into two parts, we may have better luck measuring innovation. The first part "new-idea-innovation" should be defined as the rate of change of knowledge with respect to time (ex; the ah-ha moment) where knowledge is defined as the rate of change of information with respect to time, and information is defined as "facts and data." So while innovation itself is difficult to measure, a proxy for innovation in the rate of change of knowledge is easier to identify and measure. Once you have identified and measured high rates of change of knowledge (derived from high rates of change of information), you can set out on the next piece, finding the ones with potential economic outcomes.