Uncensored CMO

The episode has a solid density of applied insights - reward prediction error applied to advertising, a proprietary 15% surprise landing zone, and the short O/long O framework - but is diluted by significant self-promotion, throat-clearing about their joint conference tour, and passages that are more recollection than substance.

“surprise accounts for around 11% of all emotions captured over, you know, hundreds of thousands of tests. So it has a really big role to play and a positive role to play as well in both short and long term”

“the correlation between kind of the campaigns that led to long term results, what we found is there was a landing zone of around 15% surprise”

The short O / long O framework and the application of reward prediction error to ad expectation management are genuinely fresh angles, but the episode leans heavily on pre-existing external frameworks (Kahneman, Seth Godin's purple cow, peak-end rule) and the central thesis - surprise is underused in marketing - is not a startling claim.

“two different Families of surprise, which we're calling the short O and the long O”

“reward prediction error, that actually our, uh, dopamine hits are as a result of our expectations of an experience rather than experience itself”

Adam Morgan is a legitimate marketing strategist (founder of eatbigfish, author of Eating the Big Fish) with real consulting relationships and proprietary project data, but this is ultimately two collaborators jointly presenting their own work rather than an operator who has built and scaled something speaking candidly about how they did it.

“I had a very interesting conversation with Kirsten uh, Halley, who until recently was the CMO of Aldi about what they did and how they approached it”

“her answer to what's the most important thing how a marketer uses a surprise was rigor and discipline, which is not what you expect to hear”

The episode is well-stocked with named researchers, specific System 1 data points, named brand examples with concrete details (Aldi champagne at £11.99, Chili's 25 million margaritas, 80/20 innovation split, 400% emotion multiplier from surprise), which lifts it above the average despite some vagueness around methodology.

“Rolf from Schultz, I think about 2016, produced, uh, a study, and it talked about something called reward prediction error”

“they sold 25 million margaritas A, uh, year”

The host adds genuine value by contributing his own System 1 data and examples, but the format is fundamentally two collaborators co-presenting their shared project rather than an interview - there is no challenge, pushback, or uncomfortable follow-up throughout the entire episode, and questions are often leading or purely logistical.

“Is there also a conversation to be had on the right balance of surprise? Because I think you can shock people on the one hand or you can pleasantly surprise people on the other.”

“I wonder how much of how much surprise is intentional. Because if I think about Jaguar”