The formation of oil-water emulsions presents a major challenge for the petroleum industry. Emulsions in wells and flowlines cause higher viscosities and lead to larger pressure drops, reducing production performance. The empirical modelling of emulsion viscosity as a function of water cut is presented. Measurement techniques for emulsion viscosity in a laboratory are explored and their use in a multiphase flow simulator to predict well and flowline pressure drops is explained. The reverse of this methodology can also be used with field data to back calculate the ostensible emulsion viscosity curve. Using the BHP Billiton operated Pyrenees oil development as a case study, this paper contrasts the laboratory derived and reverse engineered viscosity curves generated across a number of wells that have produced to the point of natural emulsion inversion. A possible mechanism for the differences between the two is proposed. In the latter part of 2014, a campaign of downhole demulsifier dosing across the Pyrenees fields was intiated with considerable success. The impact of this chemical injection on emulsion viscosity modelling is also explored in detail.