Whenever I run a new recipe, I have to run a test of the seating depth versus the powder fill.
As mentioned above, I take a representative case that is new virgin brass that is heavy for the batch, and one that has been fired but unsized. These would represent the two extremes of case volume.
I take my mandrels and find the one that will expand the necks to the point where a bullet can just barely slide. I install a spent primer in the pocket, upside down in order to plug the hole.
I then measure internal capacity in water grains. In the samples below, the virgin case is 31.52 grains of water, and the fired-unsized case is 32.44 grains of water.
I know the bullet geometry and roughly where I will stop taking data is when the bullet would get slightly into jam. I also note to myself where magazine length would be as well as when the boat tail is at the neck-shoulder junction.
I get to work dispensing powder then shaking and tapping to settle the bullet onto the powder column and I take a COAL reading.
Sometimes, you may want to lift the bullet slightly and re-settle to get several readings before you plot the point. There is some dispersion in each attempt, but it works out when you get methodical about how you do this.
In the plots, you can see inflection points when the powder starts to climb the shoulder and the boat tail gets up into the neck. Having done this drill now for many decades, I have seen the pattern dozens of times.
Here is an example of Norma 204 Ruger brass, with VV N-140 and a Hornady 40 VMax. Each point represents the COAL when the bullet is slightly compressing the powder charge.
It takes a little time, but without any other place to get the information, I have no choice but to run the test. Once I know when the bullet hits the powder column and starts to compress the load, I know what to expect in the ladder tests and when I select a seating depth. YMMV