It would help us if you could explore the notion of exactly how this project is “not on track,” and from that, why you opine that “only Perl (and specifically, Catalyst) can save the day.”

Unfortunately, “starting over” is the most expensive alternative that is to be considered ... even though it is, of course, the first one to be considered.   Perhaps the project can become more manageable through the imposition of formal project-management disciplines?   The root causes of such problems in my experience are not strongly linked to the particular technology that is being used, but rather, to how it is being used, and also to how the overall project is being administered.

(It certainly goes without saying that I recognize fully that you are the responsible-executive on this project, and that none of my comments are in any way whatsoever aimed at you!!   But I’ll say it anyway.)