Marketing
Reviews beat content, agent found
Dan Guiux is a real estate photographer in Rowlett, Texas. He runs four businesses with no staff. His team is a set of AI agents with three roles: a front door that takes in requests, a writer, and a grader that checks work. Anything that sends, posts, spends or deletes reaches him first as a draft. He shares the setup as a first-week plan an owner can copy day by day. For his own photography business, he ran the agent on his local market.
How it wentThe agent's finding for his market: reviews beat content. The top Dallas page for his service was short, with a wall of five-star reviews, while longer content pages ranked below it. He reports no revenue or time figure in the post. The value he claims is the setup itself: three named roles, the files each agent reads, and the rule that money and sending always wait for him.
Try it · based on the source's workflow
Set up three agent roles for [business name]: a front door that reads new requests in [inbox] and sorts them; a writer that drafts replies and posts in my voice from these examples: [paste]; and a grader that checks every draft against [rules] before it reaches me. Anything that sends, posts, spends or deletes comes to me first as a draft. Then run one market check: list the top [5] pages for [service] in [city] from public sources, and tell me what the top page has that mine lacks: reviews, length, offers or proof. Give me a day-by-day plan for week one.
A starting template based on the workflow the source describes. Review it before you use it; you are responsible for what it does.