How do I make vocals sit in the mix without making them harsh?
Vocals sit in the mix when level, timing, low-mid space, sibilance and ambience support the same focus. Harshness appears when you force presence before cleaning the source and the instruments around it. Edit first, ride level, remove masking, then add EQ, compression and space.
"But the vocal needs more presence"
Presence is not a high shelf by default. If guitars, synths or cymbals own the same upper mids, a vocal boost only makes pain. If breaths and esses are wild, compression drags them forward. The vocal needs a lane before it needs more shine.
What to check first
Listen with drums, bass and the main harmonic instrument. Pull masking away before boosting the vocal. Control obvious breaths, esses and timing. Then use smaller EQ moves and one believable ambience path. If the words get clearer at the same level, you are moving in the right direction.
Where this fits with YOUR MIX SUCKS
The book gives vocals their own lane because the vocal carries trust. A vocal problem is usually a chain of source, arrangement, level and space decisions.
FAQ
Should I EQ or compress vocals first?
Fix the thing that makes the next processor behave badly. Often that means cleanup before heavy compression.
Why does my vocal get sharp?
It gets sharp when presence is boosted before masking and sibilance are controlled.
Are doubles the answer?
Doubles help only when timing and level are controlled. Loose doubles often make the lead smaller.
CTA
The complete method is in YOUR MIX SUCKS by Marc Mozart. Get the book: YOUR MIX SUCKS