
How do we know it's done?
Match the method to the determinism of the work. When the output is probabilistic, “done” is a measured band, not a fixed state.
Perspectives on transformation, delivery, and the human side of change, written from inside transformations we are driving with organisations. Learning from doing, not from the sidelines. The current series is colour-coded by the four disciplines of our AI adoption work.

Match the method to the determinism of the work. When the output is probabilistic, “done” is a measured band, not a fixed state.

A crosswalk from traditional governance to agentic delivery. You don't give up control; you move it from the plan to the gate.

You don't need to know how AI works to govern it. Eight testable concepts, small enough to learn in an hour, solid enough to carry a funding decision.

The role does not go obsolete. It moves from moving rows to shaping what an agent can find, trust, and act on, and becomes the seat that decides whether the whole system can be trusted.

After a decade building and exiting a technology consultancy, here's why we're doing it all again, differently.

It's not a technology problem. It's an alignment problem.