even if you don't have returns, it was common practice to use spare normal input channels for returns - then you could EQ them, and even add more effects, or do commonly requested things like the singers wanting reverb in their monitors - send the vocal channel to the reverb, make it wet and splashy, then return that into channel 10, or whatever, and then leave the channel fader down, and use the send that goes to the singers floor monitor - you can give them reverb in huge amounts and take it away when they do the between song speak bits. The audience can have the usual reverb, they can have different. Doing it that way can be more flexible.