We use a combination of experiments and numerical modelling to study the spreading of micro-droplets on 'patterned surfaces', with applications to the manufacture of POLED displays via inkjet printing. The term patterned surfaces is used to refer to surfaces with either micron-sized topographical features or variations in the wettability or a combination of both. We demonstrate that these patterned surfaces provide an effective means to control the spread of fluid deposited using an inkjet method so that liquid films are printed in a pixel shape. It is shown that the spreading of the fluid deposited via the inkjet method can be either locally enhanced or hindered depending on whether the topography gradient ahead of the contact line is positive or negative. Locally enhanced spreading occurs via the formation of thin rivulets through a mechanism similar to the capillary rise in sharp corners. A key feature of the liquid-substrate interaction studied in this work is the presence of significant contact angle hysteresis, which enables the persistence of non-circular fluid morphologies. However, we find that the efficacy of topographical features in restraining the spreading of deposited liquid within the boundaries of a pixel is critically dependent on the accurate positioning of droplets in a pixel. We demonstrate that this dependence on the positioning of droplets is reduced on substrates with both topographical and wettability patterns. Excellent predictions of the evolution of the liquid morphologies on patterned surfaces are provided by a simple model combining quasi-static surface tension effects within the framework of a thin-film approximation, combined with an experimentally measured dynamic spreading law, which relates the speed of the contact line to the contact angle. We also show that if an experimental spreading law is not available, the spreading of a droplet in a pixel can be adequately predicted by the Cox-Voinov spreading law. This model does not include viscous effects in the bulk of a droplet and hence a given morphology is attained more rapidly in the numerical computations compared to the experiments in which bulk viscous resistance retards the fluid motion. Nonetheless, the model can be used very effectively to predict the areas covered by the liquid and may serve as a useful design tool in the inkjet printing industry.