Background: Alcohol misuse is highly prevalent in the United States and results in a huge financial and public health burden. Current alcohol reduction treatments are underused, and there is a critical need for innovation in the field. Transdermal alcohol biosensors measure alcohol use passively and continuously and may be helpful tools in alcohol interventions. To date, however, alcohol biosensors have not been widely used to directly intervene on alcohol use. There is a new wrist-worn biosensor that could be used to help people reduce their drinking, although it is unclear how best to incorporate such a device into an alcohol intervention. Objective: We aimed to identify desired features that would be acceptable and helpful in a wrist-worn biosensor–based alcohol intervention for adults who drink heavily. Methods: Participants were recruited through an alcohol contingency management study, a contact registry, and participant referral. To qualify, participants had to be aged at least 40 years, report drinking at least twice per week, and indicate interest in reducing their drinking. We conducted a semistructured interview with each participant via Zoom (Zoom Video Communications, Inc). The interview guide addressed general thoughts on the wrist-worn biosensor, how participants thought a wrist-worn biosensor could be used to help people quit or reduce drinking, types of information that participants would want to receive from the biosensor, how they would want to receive this information, and how they thought this information could be used to change their behavior. Interviews were transcribed verbatim and analyzed using thematic analysis. Results: The sample comprised 20 adults (mean age 55.1, SD 6.1 years; 11/20, 55%, women; and 17/20, 85%, Black). Of the 20 participants, 9 (45%) had previous experience with the Secure Continuous Remote Alcohol Monitor continuous alcohol monitoring ankle biosensor from participating in an alcohol contingency management study. The desirable features could be grouped into 5 main themes: features that would influence willingness to use the biosensor (it should look attractive and be both comfortable to wear and accessible), personalized messaging (personalized biosensor-based prompts and feedback could be helpful), preference for time wearing the biosensor (for some, just wearing the biosensor could have an intervention effect), sharing data with others (this was appealing to many but not to all), and mental health support (many felt that mental health support could be incorporated into the biosensor). Conclusions: Five main themes that would maximize interest in using a wrist-worn biosensor for alcohol intervention were identified. Taken together, the identified themes could inform the development of a just-in-time adaptive intervention that uses a wrist-worn biosensor to help adults who drink heavily reduce their alcohol use.