Dynamic signatures are a digitalized form of handwritten signatures. Their use has seen a steep increase for important transactions, such as life insurance and telecommunication contracts, sales and banking operations. A dynamic signature can be disputed and a forensic handwriting examiner may be hired to help determine whether it is genuine or not. Such a conflict may emerge years after the questioned signature has been affixed. Since reference material contemporary to the relevant period may not be made available to the expert, it can be questioned whether time influences dynamic signature data, which could affect the expert's results. This study was designed to explore this possible influence. Dynamic signatures of three participants were collected over a duration of 18 months, during 44 acquisition sessions. Based on this sample, the goals of describing variation of dynamic features over short and longer time periods, defining adequate sample collection strategies and sampling time frames, as well as laying down the foundation for using the time information for comparative analysis of dynamic signatures, were pursued. Both the relative stability and the slow 'drift' of signatures over time were illustrated by our results. The findings of this study lead to recommendations for sampling in casework, validate statements previously made by forensic scientist through an empirical investigation on dynamic signatures and strengthen the statistical basis for forensic signature comparison.