When results show up

The shape of the curve

Podcast episodes front-load their audience: a large share of an episode’s lifetime downloads happens near release, which is why marketplaces price against first-30-day downloads (the CPM mechanics). Your response follows the same shape per spot: a bump in the days after air, a tail as later listeners catch up. Across a three-spot flight the bumps stack, which is the mechanical reason frequency reads better than one-offs.

Why mid-flight numbers mislead

Halfway through a flight you have seen the primed-audience effect of zero spots: the people hearing spot one are unprimed, and the flight’s design assumes later spots convert them. Mid-flight CAC therefore runs structurally high, and teams that panic at it cancel campaigns exactly when the mechanism they paid for is starting. The discipline from the $15k playbook: weekly checks for delivery problems only, one CAC judgment at flight end.

Trails and lags

After the flight: dynamic ads stop serving, but episode pages, show notes links, and listener memory keep producing trickle conversions for weeks; baked-in spots (the longevity trade) keep working indefinitely. B2B is slower everywhere: a listener becomes a demo weeks later and revenue a quarter after that, so judge B2B tests at the sales cycle’s length (the B2B playbook uses two quarters), and keep the attribution question on your forms permanently.