To report the design and baseline results of a rewards-based incentive to promote purchase of fruit and vegetables by lower-income households.

A four-phase randomized trial with wait-listed controls. In a pilot study, despite inadequate study coupon use, purchases of fresh fruit (but not vegetables) increased, but with little maintenance. In the present study, credits on the study store gift card replace paper coupons and a tapering phase is added. The primary outcome is the number of servings of fresh and frozen fruit and vegetables purchased per week.

A large full-service supermarket located in a predominantly minority community in Philadelphia, Pennsylvania, USA.

Fifty-eight households, with at least one child living in the home.

During the baseline period, households purchased an average of 3·7 servings of fresh vegetables and an average of less than 1 serving of frozen vegetables per week. Households purchased an average of 1·9 servings of fresh fruit per week, with little to no frozen fruit purchases. Overall, the range of fresh and frozen produce purchased during this pre-intervention period was limited.

At baseline, produce purchases were small and of limited variety. The study will contribute to understanding the impact of financial incentives on increasing the purchases of healthier foods by lower-income populations.