Astronomers usually just think in terms of redshift and might never convert into "X light years from Earth" until they write a press release. Red-shift is measured directly, because we have all kinds of ways of knowing about the spectrum expecte from various things, and we just see the expected pattern shifted.
The calibrarion from red-shift to distance is done by considering the brightness of objects. But here astronomers can just pick and choose those objects for which they have good models.
An example is the type 1a supernova (https://en.wikipedia.org/wiki/Cosmic_distance_ladder#Superno...). These are stars that were actually growing until the tripped over a very specific mass threshold and went nova. Because they all had the same initial mass, they all have the same brightness.
You're right in that there is sometimes debate about whether the "standard candles" as they're referred to are as bright as we expect, see: https://en.wikipedia.org/wiki/Cosmic_distance_ladder#Problem...