In order to improve the efficiency of the last-mile delivery system when customers are possibly absent for deliveries, we propose the idea of employing the crowd to work as keepers and to provide storage services for their neighbors. Crowd-keepers have more flexibility, larger availability, and lower costs than fixed-storages such as automated lockers, and this leads to a more efficient and a more profitable system for last-mile deliveries. We present a bilevel program that jointly determines the assignment, routing, and pricing decisions while considering customer preferences, keeper behaviors, and platform operations. We develop an equivalent single-level program, which takes the form of a quadratic mixed-integer program with subtour elimination constraints, and it can be solved to optimality using a row generation algorithm. To improve the efficiency of the solution procedure, we further derive the exact best response sets for both customers and keepers, and approximate the optimal travel time using linear regression. We present a numerical study involving a real-world dataset. Both the fixed-storage and the no-storage systems are set as benchmarks to evaluate the performances of crowd-keepers. The results show that the crowdkeeping delivery system has the potential to yield more profits and produce less pollution due to its higher capability of consolidating deliveries and eliminating failed deliveries.