The capacitive resistivity (CR) technique is a generalization of the conventional DC resistivity method that facilitates measurements of electrical resistivity on engineered surfaces and highly resistive ground. The CR methodology allows the use of towed sensor arrays, thus enabling the rapid collection of high-resolution resistivity data. Under quasi-static conditions, CR data are equivalent to galvanic DC measurements so that CR datasets can be interpreted with conventional DC inversion algorithms. In this study, we demonstrate that the methodology and fundamental parameters of the CR technique facilitate spatial sampling at the centimeter scale for towing speeds of the order of 2.5 km/h. We argue that the dipole-dipole array is the most suitable geometry for dynamic CR measurements and present example data acquired with a prototype instrument using plate-wire combinations arranged in an equatorial geometry. The information content of raw CR data is found to be dominant over towing-induced noise and a direct comparison with DC profile data shows good agreement between both techniques. Based on these findings, we show that tomographic imaging is possible using datasets acquired with moving arrays. Closer investigation of the practical aspects of towed-array capacitive resistivity imaging (CRI) highlights the similarities with galvanic multi-electrode surveys, but the different geometric constraints and sampling regime of CRI give rise to advantages (high lateral resolution) as well as disadvantages (limitations in vertical resolution). We conclude our study with recommendations for practical CRI survey procedures and present a field example where we have successfully imaged a subsurface target in 3D. Towed-array CRI is found to provide equivalent (and in some ways superior) information about the shallow subsurface when compared to DC ERT (electrical resistivity tomography); it therefore widens the scope of electrical imaging surveys to environments where the conventional methodology would be impractical.