I usually don't gate anything to get rid of noises between phrases, I much prefer to use clip gain automation or cutting the gaps out manually.
Then I'll do a final run of compression, usually something with a bit of character, and a final EQ to make it fit into the track better. If necessary, I'll de-ess again if it needs it after all of that stuff (usually not necessary). That then gets recombined back to the one path. On the grit path, I'll usually EQ it as well to not crowd or muddy up the main vocal sound. That gives it hair but still retains the original clarity. Then I like to split my vocal into 2 paths: one clean, and one where I can dial in a bit of grit underneath that clean vocal. Then usually a fairly hot compressor (although I might leave this out if I know I'm adding a fair bit of grit later, or I want the vocal to be more dynamic). A pre-shaping EQ to highpass the vocal, and add anything I think is missing from the raw mic sound
