Abstract The initial Bonga field development was based on a "green field" integrated saturation model using Gulf of Mexico (GoM) data as analogue. In an earlier revision of the Field Development Plan, the relative permeability and capillary pressure models were based on steady-state and centrifuge special core analysis (SCAL) experiments from a few Bonga Main wells and also integrating GoM data. The saturation modeling was dependent on permeability classes (intended to be, but not explicitly representative of the facies). As production continued over the years, several attempts have been made at history-matching the observed performance history of the reservoirs that constitute the Bonga field. The results indicate that a review of some of the input parameters including the relative permeability and capillary pressure model are required to obtain a reasonable history match of the reservoir performance. The predominant facies types in Bonga field include Channel Storey Axis (CSA), Channel Storey Margin (CSM) and Inter-channel Thin Beds (ICTB). A relationship was observed between the raw SCAL data and the different facies classes from detailed core description. Current production and recent drilling/logging information coupled with 4D seismic monitor indicate preferential sweep from some of the facies. This may have a significant impact on the ability to maximize recovery from the various facies in the reservoirs. This study focuses on developing a facies-dependent saturation model that honors the current understanding of the waterflood performance of Bonga reservoirs. Good history-matched, robust and fit-for-purpose models were obtained using this methodology. This has led to better evaluation of the recovery efficiency and flood pattern of the field which has been used to support further field development options and optimize production from existing wells.