
What Will the Most Talented People Make With AI?
Judging AI output without asking who used it, how, and with which model is like judging photography by the worst phone camera.
Ironwood insights
Ideas for agency owners and teams on better workflows, creative practice, and building AI skills that last.
Agency strategy · Featured insight
When a new technology arrives that touches everything, the worst thing you can do is treat it like one thing.
Perspectives on running an agency as the work changes.

Judging AI output without asking who used it, how, and with which model is like judging photography by the worst phone camera.

Most agencies are stuck asking how AI does old work faster. The better question is what you can build now that was never possible before.

SERIES CONTEXT: post 12 of 14 in “The Six-Step Professional Services Marketing System.”
(Background for the whole series. The subject and instructions for THIS post follow below the line.)
The series explains a professional services marketing system built around six connected stages: Step Zero posi

In a recent episode of 2Bobs, Blair Enns and David C. Baker were talking about strategy, a word agencies use constantly but rarely define very carefully. Blair made a point that really stuck with me: some of the best definitions of strategy have less to do with deciding what you’re going to do and m

Work for me over the last six months has fundamentally changed. Part of my work life used to be building websites for museums, and I still do that, though with AI, it’s completely changed. I frankly enjoyed the productive days of building, putting on some nice music, and just going to work. Quite pl

LinkedIn has become increasingly polluted by automated posts, generic comments, mass-produced articles, and robotic direct messages. None of us wants more of that. But LinkedIn is still one of the most important channels available to professionals who need to build relationships, demonstrate experti

Marketing and creative firms are experimenting with AI everywhere. One person is building a proposal tool. Another is refining a research workflow. Someone else has figured out how to accelerate production, automate reporting, or improve prospecting. Many of these experiments are genuinely valuable-

The debate over AI training often collapses two very different actions into a single accusation: unlawfully acquiring content and learning from content. Large language and image models do not generally operate as enormous databases containing copies of everything they were trained on. Training expos

The Marketing AI Institute survey finding – that most professionals now call AI essential but their organizations are lagging – is getting read as a training problem or a budget problem, and both of those framings let the owner off the hook too easily. In most small agencies, the gap between ‘we bel