Robot sound influences aspects of human-robot interaction (HRI) from the way robots are perceived socially to the adoptability and even monetary value of these systems. Accordingly, our past work (and the work of others in the HRI community) has sought to empirically investigate robot sound and propose design tactics for successfully designing and incorporating new sound use in robotic systems. This paper presents a case study of using these collective past works to design and evaluate robot sound as used on a real delivery robot (Daxbot’s Dax robot) in its day-to-day commercial operations. We present the steps of our design thinking process for incorporating new and beneficial sound into the Dax robot, followed by evaluations of these designs in the wild and through a broad online survey. In-the-wild users ( \(n=6\) ) found the newly designed sounds to be agreeable, and more participants preferred the new sounds compared to past built-in sounds of the robot. The online survey-based study ( \(n=75\) ) supported these initial findings and showed that the added sounds also yielded higher perceived robot competence and purchasing interest compared to a base no-sound condition. Overall, this work offers evidence that our team’s past robot sound work (often conducted in controlled environments with research robots) has the potential to transfer to real-world use settings and commercial robots.