Arguing that the moat was data.

Leadership's answer to "who is our primary customer?" was, in effect, "everyone", which made prioritisation incoherent.

Nobody had asked me to do this. There was no mandate; it was self-initiated.

• Defined local authorities as the primary customer, derived a North Star metric, a metrics tree and an opportunity solution tree, and proposed organising teams around opportunity areas.
• Presented it; it wasn't adopted.
• Went back to first principles on competitive strategy, and pressure-tested my thinking with a departing executive who had no reason to be diplomatic.
• A year later, presented a sharper case to the CEO that efficiency gains were copyable and the durable advantage was proprietary data on an underserved group of people.

What changed between the first version and the second wasn't volume or seniority. I'd spent a year letting the argument get shot at by someone with no reason to be kind about it, instead of just waiting to be vindicated.

The CEO adopted the data-first direction, and the company has since shifted towards building AI products. It's also where my own thinking about data as a moat comes from.