>Has someone calculated this out? Or is it more of a "well we need an exceedingly sensitive instrument to detect some of the most energetic events in the universe from half-a galaxy away" gut-feel?

I was thinking about the energy required to transmit the gravitational waves, not receive them. Being able to move objects massive enough (stellar mass or more) to create detectable gravitational waves in a quick and precise enough manner to allow for communication would require mind-boggling amounts of energy.

Right, sorry I wasn't clear. If you were more interested in "local" communications, and less interested in broadcasting to the rest of the cosmos, do you still need gigantic amounts of energy at the transmitter? And how much power would you need, even if the energy is relatively high? For low power transmissions, what is the limiting factor on the receiving end? Or why can't you detect really low power transmissions? Can't get your receiver close enough to absolute zero, so thermal fluctuations kill receiver sensitivity? Background gravitation noise floor is too high across the band? Quantum fluctuations are a limiting factor? Can't make an X-ray/gamma-ray interferometer? "Antenna" size scales with length rather than area? Other?

I suppose the ratio of Coulomb's constant to the gravitation constant (or something similar) govern the relative difficulty in using gravitational vs. EM? But that's not obvious to me that it would make gravitational wave communications inefficient in absolute terms.

Oh, yeah sorry, I was thinking more along the lines of inter-galaxy communications!

I definitely do not know enough about the topic to approach answering your questions, but I'd certainly be interested in knowing the answers. I really hadn't thought about it in that context.

Maybe you don’t need to generate the waves, if you can collect them and redirect them into the shape you want.