DHK on Data and AI

Most teams think AI will improve their thinking automatically.

It doesn’t.

In this episode, I break down why most AI workflows fail to produce better thinking—and what to do instead.

The missing piece is the ideation layer: the step where you apply judgment, frame the problem, and decide what actually matters before you ever write a prompt.

I walk through a simple system:

Input → Ideation → Prompt → AI → Output → Reflection

And show how shifting where you do the thinking changes the quality of everything that follows.

If you’re working in AI, analytics, or product, this is the part that actually matters.

Here's the working article