The difference in terms of detection is that the wave does this in a time-varying, periodic fashion.

For something like LIGO, we're trying to measure length changes on the order of 1e-18 meters. We're not actually measuring the lengths of LIGO's arms to that accuracy, though. What we're measuring is the difference between the times light takes to travel down those arms. And even that's hard to measure on an absolute scale, so what we really measure is how that difference changes in time.

Or put another way, the effect of Earth's gravitational well is not really distinguishable from inaccuracies in making the two legs of the interferometer equal length to start with, and is a much smaller effect than those inaccuracies. Again, if I understand this right...