The thing about being a large company in an innovative industry is that the optimal number of “poorly conceived and managed product and R&D adventures” you need to run is not zero.

Certainly it’s possible to run too many of them, but if you are overindexing on perfect vetting and execution of every idea you might not be running enough.