video1.0<iframe src="https://www.loom.com/embed/a403afa445454e9fb7d3dbe49c0d3372" frameborder="0" width="1670" height="1252" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12521670Loomhttps://www.loom.com12521670https://cdn.loom.com/sessions/thumbnails/a403afa445454e9fb7d3dbe49c0d3372-a521813a0364d023.gif973.141This Loom focuses on how the author uses automated dashboards to track Meta ad performance against actual sales data. They update a Meta versus Books dashboard weekly by reconciling Meta MCP results with daily transactions and leads stored in Airtable, using Claude for a five minute accuracy check and flagging assumptions or dirty data. For the last 30 days, two campaigns show different outcomes: Beyond the Scroll challenge paid for itself on cash with $7,900 spent and $12.7k made (ROAS positive), while Coffee with Claude was not break even and may require checking tracking. They also review longer term cohorts since February 2025, reporting nearly 14,000 people, about $89,000 spend, roughly $180,000 gross revenue ROAS of 2.02, and rising average spend per lead over time.