Duffl Campus Merchandising Moat
David Lin, CEO of Duffl, on the economics of hyperlocal ultrafast delivery
The real moat here is not just speed, it is turning each campus into its own convenience store with products that feel native to that community. Most ultrafast players compete on a standard dark store playbook, but Duffl is pairing campus density with a local assortment layer, like fruit trucks, student bakers, and taco stands, that raises order frequency, gives each store a reason to exist, and makes the app feel less interchangeable.
-
Campus density makes local products economically viable. Duffl says racers average 10 to 12 orders per hour, versus 3 to 6 for most of the industry, because they can batch multiple dorm deliveries on a single trip. That higher throughput gives room to carry small batch items that would be harder to support in a lower density city network.
-
The local layer also solves the assortment problem that defines ultrafast delivery. Broad players typically stock 1,000 to 2,000 SKUs in a 3,000 square foot dark store, so every item has to earn its slot. Duffl uses campus specific demand data and local suppliers to make each store look more like the bodega students actually want nearby, not a generic mini warehouse.
-
This is also a defense against larger apps. DoorDash, Gopuff, and similar players can copy delivery speed or subsidies more easily than they can copy student labor networks, campus real estate close to dorms, and a merchandising loop built around local social knowledge. In ultrafast, product mix and neighborhood fit are often what create habit, not just lower prices.
The next step is turning this campus by campus merchandising edge into a repeatable launch system. If Duffl can keep using demand data to identify the right local partners, rotate winning SKUs quickly, and deepen university ties, it can build a chain of highly customized micro stores that scale like software but still feel local in each market.