I'm the 'dean of innovation' and I know that nobody can agree what 'innovation' actually means

Inside a lot of companies, “innovation” has become a word people use when they don’t yet agree on the work in front of them. One executive sees AI as growth, another sees copilots, another sees automation, and the meeting ends with everyone nodding at a different problem.

That is the core argument here: innovation is not one thing. It can mean adopting a technology, improving a process, launching a product, changing a business model, transforming an organization, or reacting to a crisis. Those are not interchangeable. Treat them like they are, and resources scatter, the metrics clash, and the project gets judged by the wrong standard.

The source leans hard on a simple test: innovative compared with what? A practice that feels bold in a legacy company can be routine in an AI-native startup. That is why the article pushes leaders to ask what norm is being broken and what value is actually being created, rather than hiding behind the word itself.

It also warns against the data trap. The more novel the idea, the less useful old data becomes. Generative AI in late 2022 is the example: no one could build a trustworthy five-year ROI model because the technology, costs, regulations, and use cases were moving too fast. In that setting, waiting for certainty mostly means learning later than everyone else.

The fix is not another grand innovation framework. It is a more honest language: what kind of change is this, how far is it from what the company already knows, how fast does it need to move, and what is the next experiment? A cyberattack, a software rollout, and a new business model do not belong in the same management bucket. Yet companies keep shoving them there anyway.