I disagree on this specific point. Yes, subjectivity is an issue, and the person submitting the ticket generally has an incentive to overstate priority. More generally though, inventing such practices from scratch without experience is probably not the best approach. People do degree courses in computer support these days, getting a hefty dose of ITIL best practices stuff. Which is to say that a lot has been learned and written about already.
Going straight to the ITIL docs would be very much overkill for where you are at, but I would strongly recommend that you look for existing guidelines at a level of complexity and formality that’s suited to your needs. Also consider that if you grab an existing system that’sdesigned around the needs of a service desk, then the initial default settings will likely get you started towards good practices in a way that you won’t get from trying to extend a more general purpose system like discourse into doing ticket management.
For your needs Cawas, something like Redmine (self hosted) or JIRA (hosted) might be an easy option for you to get started with, and then review your needs once you’re a little more familiar with the ground, and in particular once you have enough tickets that you start to understand what reporting you need to keep everything on track. It may be that you could get what you need out of discourse, but I would not recommend that you design such a system without experience when there are such easy off the shelf tools available that will help you learn what you need faster.
In terms of the discourse project, I think not having a proper ticketing system is a significant shortcoming, which will hinder the operations of the discourse team and of anyone external to that who wants to get involved. Currently this is being done within meta.discourse.org, but without good tools. I think there needs to be a decision to actually develop good support for use as a ticketing system, as this topic suggests, or a separate ticketing system wants to be put in place, perhaps with some integration with meta.discourse.org.