ITS: what is needed is the aim of the phylogenetic investigation. ITS evolves very quickly so it's good for a few closely related species and sub-species. Understanding the behaviour of a locus is as important as the volume of the locus analysed. Thus, the answer is yes.
I have a feeling these are insects/arthropods simply because of the fragmented nature of the genome: they're hard to assemble and genetically diverse so there's often no reference genome. The issue of genome vs locus was addressed by David Penny years ago: genome trees can be misleading and that would be justification for using the contigs to derive select loci. The reason why genome trees can be misleading is complex and a separate question.
Moreover, if you can map the loci to a chromosome and demonstrate the contigs are on different chromosomes you can use differences/similarities in trees between chromosomes as the first stage of disequilibrium analysis, i.e. selection acting via Mendelian exchange.
To reiterate there is good phylogenetic justification for switching contigs because the loci are more important than the size of the contig (you'll need supporting references however).