Back on topic, compressing to a lossy format changes the waveform. The overall peak level will rise to some degree. If the file is destined for lossy compression it needs more headroom. How much extra headroom will be enough is difficult to predict.
Leaving about 1dB of headroom is usually a good idea for wave files destined for the conversion process for youtube, soundcloud spotify, streaming etc..
Itunes and Sonnox also have apps/plug-ins that will let you audition what different conversion processes will sound like. Sonnox | Fraunhofer Pro-Codec
Mark me down for another that dislikes FB and most of the other networking sites.
I pay my 17 year old son to keep up with some of that stuff for me, but I think I'm going to fire him soon. ; )
With the viewership down on many forums, I can only imagine that people have been turning more and more to social media
and things that are easy to navigate on a phone. I dislike phones as well, ..but anyway, it's probably best to keep your feet wet in the social arena, because it certainly ain't brick and mortar any more and with business, it's usually a choice of adapt or die.