Dr Pepper has launched the ninth season of “Fansville”, its satirical college football platform, making it one of the longest-running campaigns in the sport. This year’s instalment runs under the line “Fandom Never Sleeps” and premieres on ESPN on 5 September during the LSU versus Clemson game.
The headline casting is LSU head coach Lane Kiffin, one of the most argued-about figures in college football following his move from Ole Miss. In the teaser “Illegal Lane Change”, Kiffin is pulled over by a sheriff played by Brian Bosworth and rolls up his window in silence. Keurig Dr Pepper has said the moment came from Kiffin himself rather than the script.
Derek Dabrowski, senior vice president and head of cold brand marketing at Keurig Dr Pepper, has attributed the platform’s longevity to the irrational passion of college football fans, and said the team asks each year whether there are still fresh fan truths left to find. He has described Kiffin as a natural fit for a world built on the drama and absurdity of the sport.
The part marketers should be looking at
The comedy is the visible layer. The commercially interesting part sits underneath it, and Keurig Dr Pepper has already disclosed it to investors.
Speaking at the Deutsche Bank dbAccess Global Consumer Conference on 3 June 2026, chief executive Tim Cofer said the previous autumn’s Fansville campaign ran more than 2,500 different marketing permutations. Creative was tailored to variables including a viewer’s location, their preferred retailer, their beverage and occasion preferences, and whether their favourite team had won or lost its most recent game.
Where those precision and personalisation tools were deployed, Cofer said the company saw more than double the sales lift and roughly 30% higher incremental return on ad spend compared with its national campaign average. He also indicated the same tools are now being extended to other brands in the portfolio.
That last point is the real signal. A creative platform has been converted into a repeatable operating capability.
A compounding asset, not an annual campaign
Nine seasons of the same world means the brand no longer pays to establish context. Audiences already know the town, the tone and the joke format, so each new execution starts with recognition already banked.
Dabrowski has framed the approach as protecting what makes the campaign instantly recognisable while refreshing what makes it relevant. The fixed layer is the world and the comic register. The variable layer is the storylines, which draw on conference realignment, playoff expansion, player movement and the growth of name, image and likeness deals.
Alongside the ads, the season carries branded merchandise and fan-facing activations tied to moments in the college football calendar.
The Engage Positive view
The asset is the format, not the film. Fansville works because a fixed world plus variable storylines allows near-unlimited executions without a rebrief. Most brands rebuild from scratch each year and pay the recognition cost every time.
2,500 permutations is a production system question, not a media question. Nobody hand-cuts 2,500 assets. If you want this, invest in modular creative, a naming and versioning discipline and a feed structure before you touch the media plan.
The winning variables were behavioural, not demographic. Retailer preference and whether your team won last weekend beat age and gender brackets. Ecommerce brands already hold equivalents: last category purchased, basket value, replenishment window, weather.
Measure incremental ROAS, not blended ROAS. The published claim is a 30% lift in incremental return, not platform-reported return. Any brand copying this without an incrementality measurement plan will not be able to tell whether it worked.
Persist the platform across seasons in your own category. Diwali, Ramadan, back to school, end of financial year. The compounding only starts once you commit to the same world for a third and fourth outing.
The honest counterpoint. Fansville sits on a nine-year investment, a national broadcast budget and a category where the purchase is impulsive and cheap. A mid-market brand cannot replicate the recognition base, and personalisation at this scale has a real fixed cost in tooling, data and production governance. Below a certain media weight, 2,500 permutations produce sample sizes too small to read, and the sensible version of this strategy is 20 well-chosen variants, not 2,500.
