The article discusses the considerations in choosing between model-based designs and space-filling designs for an experiment. Topics covered the benefits of each type of design, the examples of the response surfaces suitable for model-based and space-filling designs and what to do if the assumed underlying model form is wrong.