Making it up as we go along
I think there are several explanations for that.
First, unlike virtually every other business function, innovation has no accepted dogma or methodology. If I visit my local bookstore, I can find several sales methodologies that have withstood the test of time. If I visit the purchasing and inventory teams within most businesses, many of them have been through APICS or other certification. We don't even have to define the GAAP and other requirements and methods that the finance team lives under, or the JIT/MRP processes of manufacturing or Six Sigma processes for quality. All of these business functions have well-defined and well-received business methodologies and documentation.
Innovation, on the other hand, really doesn't have a defined, codified methodology or set of business processes and practices that are published. In fact, most firms I've spoken with basically make up their process as they work on new innovations, and rarely use the same process twice in a row. This means innovation is more haphazard and subject to failure than other business processes. If you take for granted that innovation is likely to have more failures (and hopefully more big wins) than other business processes, then let's at least put some defining processes and metrics to manage the swings in the outcomes.
There's an old saying from my days as a consultant - No one ever got fired for hiring IBM. I guess now that's not quite true, but the meaning behind it was that a project could fail without significant punishment if IBM could not make it happen, but if a project failed and you picked a smaller, no-name consulting firm to work with, you were in trouble. In a situation where there's a lot of risk involved, people want some definitive guidelines. What are the established paths? What experts should we turn to? Who has been down this road before and where are their maps?
This aggregate learning should be converted into methodologies for innovation, and business processes to implement to support innovation. This would follow the normal process that we've seen before. Before Motorola and GE, there was no "Six Sigma" methodology. Someone had to invent it, define it and codify it so others could adopt and follow it. Likewise, right now there is no Holy Grail of innovation methodology. In a year or two we'll look back at some defined process and wonder why it did not exist back then.
So, it's up to us. Let's get together and define an innovation methodology and business process that we can share. This would mean that a number of consulting firms, software firms and innovation thought leaders would pool their ideas and we can define one "standard" innovation methodology and write the "Six Sigma" or Miller-Heiman for innovation. Once a methodology is codified, it will be much easier for firms to actually implement innovation initiatives.