>, but you can also go too far in that direction of analysis. If you let the empirical gain in traction to sneak in as a factor in your analysis, you effectivel

>, but you can also go too far in that direction of analysis. If you let the empirical gain in traction to sneak in as a factor in your analysis, you effectively have a tautology, and you lose all predictive power That's a good point. I'd have to give it more thought on how much predictive power we're supposed to ascribe to multiple-dimensions. Maybe multiple dimensions is a better thinking tool for experimentation than prediction. This frees one from being paralyzed by the status quo. For example, a 1990s entrepreneur sees that the internet is becoming popular and maybe he thinks he can sell books on it. But then he goes to Barnes & Noble and sees that people are pulling real books off the shelves, taking them to the lounge chairs, and then leisurely sitting down and flipping through the pages before buying. If the dimension of "in-store experience" constrains the worldview, it will not occur to the entrepreneur that other dimensions like near-infinite selection in a "virtual bookshelf" and convenient delivery to the home will be prioritized over the store's lounge chairs. One doesn't know if those other dimensions for "book buying" matter more to buyers until he tries it. Another well-known example of constrained single-dimensional thinking is analyzing the Blackberry vs the iPhone in 2007. Both the CEO of Research In Motion and Steve Ballmer of Microsoft were focused on the physical tactile keyboard. They saw that the iPhone obviously lacked real QWERTY buttons, so they both predicted it would flop in the marketplace. Steve Jobs' Apple team ignored the dimension that RIM & Microsoft used for criticism. Instead, they looked at the dimension of "screen size" and made it larger. The also bet on the dimension of "capacitive touch screen" which enables finger gestures. These were the dimensions that a billion smartphone buyers ultimately preferred. The Blackberry was cheaper and buyers still preferred the iPhone touchscreens.