It's the bandwidth, stupid!
Yes, there are other issues like strategic alignment and repeatable processes, but bandwidth is probably the most crucial. Here's why. Most people who are assigned to "lead the innovation process" start from ground zero. They have to determine what their responsibilities are, how to accomplish this poorly defined task, how to get the rest of the organization to do what they've been chartered to do, and build the processes and tools to help accomplish the task. Oh, and they need to do all of these things at the same time. What often happens is that people flit from task to task without accomplishing much because the prioritizations are difficult and everything seems to be of equal importance.
Also, most firms simply don't assign enough manpower to the job. Giving one person the task of making a firm more innovative is like asking one soldier to storm the beach at Normandy. Even if he or she could accomplish the task, how long could they hold on before they'd need reinforcements?
Another analogy I like is the Yogi Berra concept of forks. You know, If you come to a fork in the road, take it. What do you do if there are eight equally important forks? Can one person adequately pursue all of the critieria necessary for innovation to become sustainable in any reasonable timeframe? I think the answer is an easy "no".
So, to quickly doom any innovation initiative, simply assign one innovation leader with no clear direction and no one to help her. Otherwise, carefully examine the workloads and timeframes and staff the initiative accordingly.