The size (cross sectional diameter) that the cone of the beam makes on the surface of the sample affects 1) the resolution of the image and 2) the number of electrons generated (therefore the graininess of the image). At low magnifications we use a larger spot size than at higher magnifications.
In actual fact the number of dwell points (spots in the row in the diagram) is constant for any one magnification so that too small a small spot size will result in gaps where signal is not generated and too large a spot size will result in overlap and average the signal.
Images taken at the same magnification, kV, and working distance but using different spot sizes show these affects. The difference in blurriness (resolution) is easily seen across the series
The spot size on the far left side is 60 and the spot size on the far right is 17 (the series is 60, 50, 40, 30 and 17). The way spot size is reported depends on the make of the machine used. The image is of a cross section through carbon fibres (note the round profiles) embedded in a polymer. Some extraneous material is present on the surface.
Spot size changes as a number of machine parameters are modified. For example the spot size is larger at long WD (working distance) than at short WD. A smaller objective lens aperture produces a smaller spot size. Also, a higher current though the condenser lens (labelled spot size on the control panel or software) creates a smaller spot on the sample no matter what WD is used. Therefore when the WD is small, the condenser lens setting high, and the aperture small, we see the smallest possible spot size. These three parameters interact and need to be considered carefully to achieve the best image because they also affect other parameters such as field of focus and strength of the electron signal.