Nobody has a tape measure a billion light-years long. Distances to galaxies rest on a ladder of overlapping tricks — each rung calibrating the next — and right now, the ladder's top is producing the biggest argument in cosmology.
Rung one: the parallax trick
Hold up a thumb and blink each eye — the thumb jumps against the background. That jump is parallax, and it's the only direct distance measurement astronomy has. As Earth swings around the Sun, nearby stars shift minutely against distant ones; the size of the shift gives the distance by pure geometry. The unit astronomers actually use, the parsec (3.26 light-years), is defined by this triangle. The shifts are absurdly small — well under an arcsecond — which is why the Gaia spacecraft was built: it has now pinned down positions and distances for nearly two billion stars, turning the Milky Way's neighborhood into a measured map.
Rung two: lighthouses with labels
Parallax dies beyond a few thousand light-years, so the ladder needs 'standard candles' — objects whose true brightness is knowable, so their apparent faintness reveals their distance. The first was found in 1912 by Henrietta Leavitt: Cepheid variable stars pulse with a rhythm set by their true luminosity — read the rhythm, know the wattage, compute the distance. Cepheids measured by parallax calibrate Cepheids in other galaxies; it was a Cepheid that let Edwin Hubble prove Andromeda lies far beyond the Milky Way, detonating the universe from one galaxy into billions. For truly vast distances, astronomers use Type Ia supernovae — exploding white dwarfs of near-uniform peak brightness, visible across most of the observable universe.
The ladder's wobbling top
Stack the rungs and you can measure the universe's expansion rate — the Hubble constant — and here modern cosmology hits its most honest embarrassment. The ladder method (parallax → Cepheids → supernovae) gives about 73 km/s per megaparsec. A completely independent method, reading the universe's baby picture in the cosmic microwave background, gives about 67. The error bars no longer overlap, and nobody knows why. Either a subtle mistake hides in one method, or the universe contains physics the standard model lacks. It's called the Hubble tension, it is unresolved as of this writing, and it's worth savoring: a measurement disagreement, published openly, may be the crack where the next revolution gets in.