Wireless communication has an indispensable role in today’s world. We investigate multi-destination RIS-assisted wireless communication in the context of URLLC requirements. A source node aims to communicate with multiple destinations within stringent latency and reliability constraints. It utilizes a reconfigurable intelligent surface to assist in the downlink transmission, both via the RIS-reflected signals and direct paths. Due to the geometric structure of the RIS, spatial channel correlation arises which can impact performance. We also evaluate several scheduling strategies for selecting a single destination during each time slot, including random, absolute channel power-based, and normalized channel power-based scheduling. It derives analytical expressions for the block error rate under each approach and validates the analysis via Monte- Carlo simulations. The effects of key parameters are also explored, such as the number of channel uses available, the count of destinations, the information bit quantity, and the degree of spatial correlation. Ultimately, the effectiveness of the different scheduling schemes is assessed and compared in order to provide insights into their suitability under varying channel conditions and system constraints imposed by URLLC.