The primary goal of optical microscopy is to visualise and thereby understand microscopic structure and dynamics. Dramatic developments over the past decades have enabled routine studies down to the single molecule level and structural observations far beyond the limits defined by the diffraction limit through the use of fluorescence as a contrast mechanism. Despite its many advantages, one of the fundamental limitations of fluorescence detection is the frequency with which photons can be emitted and thus detected. As a consequence, although images and even movies of single molecules have become commonplace, imaging speed remains limited to few to tens of frames per second by the quantum nature of single emitters. The result is a considerable gap between the speed at which dynamics can be recorded and the underlying speed of motion on the nanoscale. I will introduce an alternative approach to fluorescence microscopy that relies on the ultra-efficient detection of light scattering called interferometric scattering microscopy (iSCAT). I will show that iSCAT is capable of following the motion of nanoscopic labels comparable in size to semiconductor quantum dots with true nm accuracy down to the microsecond regime, the relevant timescale for a majority of nanoscopic dynamics. Thereby, we are able to address a surprising variety of fundamental questions in molecular biophysics including the mechanical properties of DNA, the existence and formation of lipid rafts in membranes and the mechanistic details of the processive steps taken by molecular motors. I will close with examples of the potential use of iSCAT as a bio-sensor with sensitivies down to the single protein level.