Ahh, this is a shame, but not surprising. Sprigs competitors have all fallen as well (Bento, Spoonrocket) for the same reasons. I think there is a business tha

Ahh, this is a shame, but not surprising. Sprigs competitors have all fallen as well (Bento, Spoonrocket) for the same reasons. I think there is a business that can be made with this model, as restaurant delivery does work, and the margins are improved nicely if you have roving delivery units with all the required deliverables (since you cut out half of the round-trip delivery time). There was a pizza company (I've been unable to quickly google which one) that learned they could take a good deal more orders by sending out drivers with extra unordered pepperoni pizzas. When an order came in for a pepperoni pizza, they'd just relay it to a driver already on a delivery and have them make an extra stop. 2x deliveries for maybe 1.1x delivery distance. Sprig, Bento, Spoonrocket, Presto Pizza, and all these other startups were trying to capture this market, but the problem is that demand is not consistent and slight mistakes in supply are onerously expensive. Too much food, and you've got spoilage; too little and you've lost a customer because of your 'Sold Out' message. Additionally, the question of quality and variety is a difficult one to tackle. The pizza place that found success only had to carry pepperoni or cheese pizzas; Spoonrocket failed because it cheapened its product to try to increase the margin wiggle-room and lost customers; Bento failed because it couldn't offer much intraday variety. Sprig was at least hopeful, in my eyes -- they managed to keep their meals at a higher standard than Spoonrocket. I was fine paying extra for it (although I admit that is subjective). The achilles heel is that if you want startup growth, then managing to keep your supply within the margins of profitability is a very difficult task. The only solutions are to grow a lot slower (and learn your demand) or to double-down DEEP on your datascience team and hope they can divine a line of best fit. Unfortunately, that latter would likely need to be a Psychohistorian and those won't be around for another 10k years.