When following participants across multiple sessions one needs a way to link the different session records while protecting the participants’ privacy. Privacy is required by recent legislation such as the General Data Protection Regulation (GDPR). Many anonymous linking methods have been proposed, but these involve effort from the participants, involve long IDs, or require all participants to be known a priori. This study presents the BRIDGE procedure for anonymous linking of participant records with dynamically increasing samples. The procedure relies on human intervention to resolve ambiguous cases using manual recognition challenges. Simulation results show that the procedure can successfully map participant names to short and unique anonymous IDs, and that the percentage of human interventions is low. The procedure holds potential for HCI researchers who need to employ simple, flexible, and incremental sampling strategies while protecting participants’ privacy.