Working with a very dynamic singer

Normal Bill

Normal Bill
Looking for advice. (always looking for advice, really)
I don't sing my own stuff. I write music for a woman who has a very powerful and dynamic voice tending toward a brighter sound. (Think Cranberries, Kate Bush, etc.)
I do a lot of automation, compression, de-essers and everything. For the last album I put a dynamic EQ up at the top of the chain to cut out the harshness before it hit any compressors or anything.

Does anyone have any tricks that work for them in this situation?
Maybe something I haven't considered?
Yeah, multi-tracking. If she's able to perform very consistently (in terms of both pitch and timing), you can stack a bunch of vocal parts on top of each other, and it will have a compressing effect on the dynamics overall. You can bring the tracks in and out as needed to adjust the arrangement dynamics throughout the song.
Harmonies can be especially useful for this, since if she's trained, she should naturally modulate her harmony deliveries to be a little less dynamic, so they'll naturally compress more.
Coaching? A good experienced vocalist knows he to work the microphone. That fixes a lot of problems right there. The quieter passages they lean in, the louder they back off.
Have them watch the meters while they’re singing as an exercise.

Just because some one can sing well, doesn’t mean they can control the mic.
Worth considering two different mics, side by side, and always recording both, so you have the best for the quiet, more intimate bits and the other mic handles the potent bits. The only thing is to be careful to treat them the same. Once you start swapping it’s easy to add something to one and forget the other.
Ha! I thought this was going to be about digital overload.
No matter, for the benefit of any newbs with that problem, record at 24bits (always) and it will be perfectly fine to crank mic gain down to an average level of -30--35dBFS thus giving massive headroom.

Many newbs, especially those with a TR background, think this will make for noise. It won't and the signal can be post limited then brought up to neg 18 or so.

Well, it kind of was @ecc83
but the advice goes in the direction it goes, I guess. (I never said anything about the person being untrained, etc. She just goes from whisper to clip in 2.2 seconds. lol)
I'm fine with it, though.
Sometimes with the gain down low like that I felt like I was doing something wrong. Nice to know that that is an acceptable way to make things happen.
On certain songs my vocal is dynamic and I agree with the previous posts about working the mic and use of compression. Also, I always sing off-axis with my LDC mic because it helps to reduce those direct hits of the P's and S's. Once the track is recorded I will actually manually go in and edit the worst offenders and adjust the volume of single words and add a fade in to hard hitters. Yeah it's grunt work but there are usually only a handful of these bad actors. I know there are ways to automate this task but the manual way gives me exactly what I want.