Crowding of the subcellular environment by macromolecules is thought to promote protein aggregation and phase separation. A challenge is how to parameterize the degree of crowding of the cell interior or artificial solutions that is relevant to these reactions. Here I review colloid osmotic pressure as a crowding metric. This pressure is generated by solutions of macromolecules in contact with pores that are permeable to water and ions but not macromolecules. It generates depletion forces that push macromolecules together in crowded solutions and thus promotes aggregation and phase separation. I discuss measurements of colloid osmotic pressure inside cells using the nucleus, the cytoplasmic gel, and fluorescence resonant energy transfer (FRET) biosensors as osmometers, which return a range of values from 1 to 20 kPa. I argue for a low value, 1-2 kPa, in frog eggs and perhaps more generally. This value is close to the linear range on concentration-pressure curves and is thus not crowded from an osmotic perspective. I discuss the implications of a low crowding pressure inside cells for phase separation biology, buffer design, and proteome evolution. I also discuss a pressure-tension model for nuclear shape, where colloid osmotic pressure generated by nuclear protein import inflates the nucleus.