Standard wound healing assays often deal with just one initial wound shape, and it is unclear whether varying the initial wound shape might impact how we interpret results from these experiments. Here, we use a new kind of wound healing assay, called a sticker assay, to explore the role of initial wound shape. We examine wound healing with square, circular and triangular wounds. Measurements of wound area as a function of time show that the rate of wound closure depends on the initial wound shape. This is interesting because the only aspect of the assay we change is the initial wound shape, and the reason for different closure rates is unclear. To provide more insight we calibrate a typical mathematical model to match the data. This shows that the rates of cell motility and proliferation associated with different initial wound shapes are approximately the same, implying that the differences in the closure rates are consistent with a typical mathematical model of wound healing. Therefore, parameter estimates from an experiment with a particular wound shape can be used to predict an experiment with a different initial wound shape. (C) 2018 Elsevier Ltd. All rights reserved.