If you look at the current state of the art for a "money is no object" manpack size VSAT/SNG terminal for geostationary use in a random very hard to reach area, such as might be owned by CNN or the BBC, the GaN Tx RF chain BUC/SSPA is remarkably small now compared to ten, twelve, fourteen years ago.
as your geostationary-transponder-kHz using SNG terminal gets smaller you are still forced into using really rudimentary modulation so the other calculation is the $ per month for a dedicated piece of transponder space lease, or the even higher costs for occasional-use transponder space.
Why would a change in the antenna radiation pattern limit the power spectral density, is it a regulatory limit?
In general the radiation pattern envelope (RPE) gets worse when viewed on an azimuth and elevation plot chart as the circular parabolic (or elliptical offset parabola) gets smaller.
Same general idea of why you might be able to successfully coordinate a fcc part 101 fdd band plan link terrestrially in a major urban area with a 120cm antenna with tighter RPE than same theoretical link if you submit it with cheaper antennas.