This article evaluates the impact of two nudges on stimulating the use of doggy bags in restaurants. We run a field experiment at 14 restaurants in the province of Turin (Italy). In the first group of restaurants, we manipulated the descriptive social norm. In a second group we changed the default option - whereby customers asked for the doggy bag - by instead directing the waiter to automatically deliver the doggy bag unless told not to. A third group of restaurants was used as a control with no intervention. We find that the social norm intervention led to a sizeable increase in the use of doggy bags, while the default manipulation had a non-statistically significant impact.