The light microscope has transformed our ability to observe microbial systems, but observation alone is no longer enough. Across microbiology, there is a growing need for quantitative optical methods that do more than reveal structure: they must extract measurements that are reproducible, biologically meaningful, and applicable to complex samples. In this talk, I will present our recent work aimed at advancing microscopy as a measurement science for microbiology. I will begin by discussing quantitative approaches enabled by large-volume, high-resolution imaging, where new computational tools now allow us to measure spatial organisation, population distributions, and structural heterogeneity across intact specimens. I will then highlight how super-resolution methods can be adapted to quantify sub-cellular dynamics in live microbial cells, providing precise measurements of processes that were previously accessible only qualitatively. Finally, I will describe how label-free computational imaging techniques can generate quantitative contrast without fluorescence, with machine learning supporting objective assessment of host–pathogen interactions. Together, these developments illustrate a unifying theme, namely optical imaging as a rigorous quantitative toolkit. Rather than focusing solely on producing high-quality images, I propose that the future of microscopy in microbiology lies in measuring architectures, dynamics, and physiological signatures with accuracy and scale. By integrating instrument design with computational analysis, we can turn complex microscopy datasets into reliable metrics that deepen our understanding of microbiological structure, behaviour, and interaction.