Notes · AI in marketing work

Using AI in everyday marketing work

Most teams don't need an AI strategy deck. They need AI doing real work on an ordinary Tuesday. Reports, briefs, segment ideas, first drafts.

What I built at YAMU Media

At YAMU Media I was Head of AI Marketing, on contract. I built an AI operating system in Claude and Grok. Specialist agents took research, briefs, reporting, and daily ops, and each one had a single job.

I also built 15+ custom AI apps. Those included social listening tools, a live launch timeline, a creative QA system, and a CRM data dictionary.

Every asset went through human QA. We shipped 300+ ads, landing pages, videos, and content assets that way.

Where AI helps most

Reporting. Pulling numbers, summarizing what changed, and flagging what looks off. That used to eat hours every week. Now a person reads the summary and checks the parts that matter.

Briefs. An agent can take a goal, an audience, and past work and turn it into a solid first draft of a brief. The team spends its time on the thinking, not the formatting.

Personalization. AI can draft variants for different segments much faster than a person can. It still needs clean data and clear rules about who gets what.

Human QA is part of the system

AI drafts fast and gets things wrong with confidence. Somebody has to check the facts, the brand voice, and the numbers before anything goes out.

I build that check into the workflow from day one. Every agent has an owner, and every output has a person who signs off.

Start small

Pick one recurring task that eats hours every week. Build the agent for that one task. Put a person on QA. Once it holds up, do the next one.

If you want help with AI marketing, CRM, or lifecycle, text (508) 322-1436 or email drew@drewpion.com.