ABSTRACTResponse‐adaptive randomization (RAR) procedures have been extensively studied in the literature, but most of the procedures rely on updating the randomization after each response, which is impractical in many clinical trials. In this article, we propose a new family of RAR procedures that dynamically update based on the responses of a group of individuals, either when available or at fixed time intervals (weekly or biweekly). We show that the proposed design retains the essential theoretical properties of Hu and Zhang's doubly adaptive biased coin designs (DBCD), and performs well in scenarios involving delayed and randomly missing responses. Numerical studies have been conducted to demonstrate that the new proposed group doubly adaptive biased coin design has similar properties to the Hu and Zhang's DBCDs in different situations. We also apply the new design to a real clinical trial, highlighting its advantages and practicality. Our findings open the door to studying the properties of other group response adaptive designs, such as urn models, and facilitate the application of response‐adaptive randomized clinical trials in practice.