You are asked to restart an enterprise architecture function that was disbanded two years ago after being seen as a blocker. How do you approach the first six months?

Understand what it did that got it disbanded

The pattern is consistent: it reviewed everything, decided slowly, produced documents nobody read, and became a queue teams routed around. Restarting the same function with a new name gets the same result on a slower timeline.

Find out specifically, from the teams rather than from the leadership that disbanded it. The complaints will be concrete and they are your design brief.

Lead with something useful before anything governing

Six months of governance before any value earns no goodwill. Start with an artifact teams want:

• An application inventory with ownership, because "who owns this?" is currently an archaeology exercise.
• The integration landscape, because nobody can answer what breaks when a system moves.
• A duplication view — a capability map with system count overlaid — because it produces an immediate cost conversation.

Each is useful to the teams and to leadership, and each establishes the function as a source of answers rather than of approvals.

Then govern narrowly

Publish what the function reviews and, more importantly, what it does not. It reviews decisions expensive to reverse, crossing more than one team, or introducing a new capability. It does not review conformance to standards — automate that — nor implementation choices within an agreed boundary.

Commit to a turnaround time and meet it. A review that takes three weeks will be avoided, rationally.

Measure the right thing

Not the count of reviews. Time from request to decision, and adoption of whatever the function provides. If teams are not using the inventory, it is not useful, and that is information about the artifact rather than about the teams.

The framing to leadership

Enterprise architecture succeeds by making good decisions easy, not by preventing bad ones. Ask to be measured on duplication removed, delivery lead time and decisions made per week — outcomes rather than activities. That framing is also what stops the function drifting back into a review queue.