I am still thinking about the MMM conversation I had with an agency recently. Patrick Gilbert and Nechama Teigman from AdVenture Media pulled back the curtain on how they implemented Meridian without enterprise-level overhead and with the help of AI tools. (Update! Since this conversation, Meridian has introduced interactive Agentic Skills to help you prepare data, configure priors, fit models, analyze diagnostics, run budget optimizations, and plan scenarios.) If you want to upgrade your measurement strategy but have been intimidated by the perceived barriers to entry, give this a read. 👇 https://lnkd.in/ey-6ttAy
I’m always happy to ditch my opinions if new data shows a better way. It might sting for a second, but that’s just marketing 😊
MMM is becoming far more practical when teams stop treating it as an annual presentation and use it as a decision system. Meridian’s lower barrier is valuable, but the quality of priors, spend taxonomy and experiment calibration still determines whether the recommendation is actionable. I’d be interested to see how agencies are reconciling Meridian outputs with incrementality tests and platform-level attribution when signals disagree.