In order to remove flat spots from case necks fired in a rifle with a strong ejector spring, it is usually necessary to use a much larger diameter mandrel than you would ever need for routine brass prep. For example, I might use a 0.3105" mandrel to reduce flat spots on a .308 Win case, because the case necks have already been expanded by the firing process, and it requires that large of a mandrel to have a noticeable effect. Even that large a mandrel will not completely remove the flat spot, but it does reduce it sufficiently that a bushing die can then do its job and size the neck properly.
Likewise with virgin brass that have necks damaged during shipment, it may be necessary to use a mandrel seemingly larger than you'd want, given that you're about to size them down with a bushing die. However, using a mandrel large enough to [mostly] remove the flat spot will allow the bushing die to do its job more effectively and completely. Even if you have to use a mandrel as large as bullet diameter, its not going to hurt the brass. If you have a caliber-specific set of mandrels, you can start with the smallest and determine empirically which one is the smallest mandrel you can use that will effectively remove [most] of the flat spot. If you only have one or two sizes, start with the smallest and see if it effectively removes most of the flat spot. If not, move on to the next larger mandrel.
If the brass you were working with was fire-formed, and the neck damage (flat spots) were coming from a overly strong ejector spring, I'd tell you to open the flat-spotted necks up first with the over-sized mandrel, then use the bushing die, then finish with the appropriate mandrel for the specific neck tension (interference fit) you're after. In other words, you'd be adding a mandrel step up front solely to unsquash any flattened necks. Then you'd process the brass as normal, using a bushing die first, then a mandrel step to open the necks up and set the desired neck tension. This would address the desired effect of pushing inconsistency in neck wall thickness toward the outside.
However, you're talking about virgin brass, so it may not be necessary for you to go through the entire process the first time through. You certainly can if you want to:
1) open up any flattened necks with a mandrel large enough to do the job
2) size necks with a bushing die set so that it does not bump the shoulder back on the virgin brass
3) set final neck tension with a mandrel
However, that's a fair bit of work for virgin brass and you may find that simply using a mandrel to open any flattened necks followed by the bushing die may be sufficient the first time through. Just be aware that in this approach you'd be using the bushing die to set neck tension, and it will likely require a different sized bushing (i.e. .001" to .002" larger) than you woulkd need if you were goiung to size them down with the bushing die first, and then open the necks back up with a mandrel to set neck tension as the final sizing step. Remember, you really want to move the brass a thousandth or two during any given sizing step so that some work is actually done on the brass.
Finally, you may even find it possible to use a single mandrel step to process virgin brass prior to its frist firing. However, this depends on the state of the brass as it comes out of the box. I have occasionally had Lot#s of brass that didn't have any flat necks and pretty much all the necks appeared to be only a tick tighter than the .002" neck tension I'm usually after. So I used a mandrel that was .0015" under bullet diameter to open up all the necks, leaving very close to .002" neck tension, then trimmed/chamfered the cases and called it a day. In other words, you never really know what you're going to get, or what you'll have to do until the virgin brass has been opened, inspected, measured, etc. Even then, there are still various options for the first prep that largely depend on your specific needs/wants. The good news is that whatever method you decide to use is not going to hurt the brass. You will still be able to process it however you want after the first firing. So a little experimentation with various approaches now isn't really going to hurt the brass, as long as you don't do anything crazy with it.
