The full quote: Ninety-plus percent of the innovation at Amazon is incremental and critical and much less risky. We know how to open new product categories. We

The full quote: Ninety-plus percent of the innovation at Amazon is incremental and critical and much less risky. We know how to open new product categories. We know how to open new geographies. That doesn’t mean that these things are guaranteed to work, but we have a lot of expertise and a lot of knowledge. We know how to open new fulfillment centers, whether to open one, where to locate it, how big to make it. All of these things based on our operating history are things that we can analyze quantitatively rather than to have to make intuitive judgments. In other words, you can measure things after they become variations on a theme, because you know what to measure, and you can measure them against past successes and failures. They've minimized the "nova" in innovation. When something is truly new, like the first time Amazon opened a new product category or a new fulfillment center, quantitative analysis isn't as helpful. "Stuff is up 12%, and crap is down 5%." "Is that good or bad?" "I dunno."