Murphy and Kaplan have reviewed the current status of radio transient studies, focussing on image domain (or ‘slow’) transients, on timescales of seconds to years, and summarise the developments enabled by SKA pathfinders and precursors. The plot above shows the relevant timescales of different classes of radio transients. Approximate limits of variability timescales are shown for different sources and different mechanisms. Highly-polarised largely coherent transients are separated in the top red box from the synchrotron afterglow in the bottom blue box. The timescales are roughly delineated for traditional transient searches that find sources individually in each epoch and associate them across epochs (>∼ hours) and those that use image-subtraction or related techniques to find shorter-timescale variability at a reduced computational cost (<∼ hours)