To quote that post:
“I believe that while you can use prediction markets to help quantify innovative ideas, using PM’s to prioritise ideas is a poor use of the tool.”

Prediction markets are great when you have a rather well-defined idea or innovation. For example, in the final stages of creating a new consumer product it would be very useful to determine what the company (and distributors) think demand of that product will be in order to manufacturing figures correct. But when you're dealing with 30+ different early-stage ideas, all of which will change as they become more and more real, prediction markets are a poor tool. There is simply too much uncertainty for a market to be of much use; other tools are more appropriate.

I think if you go and read my earlier post you'll see that I've probably addressed your concerns.

Take Google, I think they have used PMs to get quantifiable data in all phases of “innovation” and implementing ideas into practice. Intel has used them to get predictions on how products will do in markets (lots of examples from using PMs in final stages of “innovation process”).

So, I don't agree, although I think you could be right for blaming companies focusing too much on front end.

It seems that a lot of prediction market-based applications tend to focus solely on ideas, which I think is unfortunate and misses most of the innovation process. Action is so much more important than the idea when making innovation happen. I really like what InnoCentive is doing, but also think there's a potential for another whole range of applications to help companies that have trouble defining their problems.

Thanks for commenting,
Jed