I think it would be correct but cumbersome.
keep a second of silence at the beginning of the sound file. cut when ready.
both compressor settings must be sample-accurate and not add or delete samples when processed through wavelab.
when the song starts with a hard hit, the compressors might overshoot in different ways. then I would copy the start attack so that it repeats with a minimal gap. later cut out like said.
you might consider a different host. I use tracktion (for its price), with some tricks it can route the signals the right way.