CEPA is a foreign-policy institute publishing on NATO, Ukraine and EU tech policy. One keyword search returned 349 results, with people, events, articles and PDFs run together in a single list.

I ran the content audit before there was a contract, and it framed the proposal.

What it found was that staff were not tagging carelessly. Tagging cost more than it visibly returned, every single time. Treat that as a staffing problem and it gets a policy, a training session, and stays broken. So it got a shorter vocabulary instead: one taxonomy across all four content types, nineteen shared issue tags.

Load times fell 70%. Three years on, the relationship is still running.