Yes, that could definitely work if you have the skills for it.
There are several cases here:
A) The obvious "go into management" option. This isn't any different than proceeding into management in any other software, with one exception: going into project management, that requires some particular business skills. If you can get business familiarity with some particular business type (e.g. realestate, lending, restaurants, etc.), that's a strong advantage. I'm familiar with people who went both ways (technical to business or business to technical), that is doable and profitable.
B) There's the "do user training" option - probably pays less than what you get now, and you could still be stuck with SFDC in a way. You have to insist on doing more than that.
C) A friend got involved as an external consultant to a project for a client. What she did was to look out that the actual implemtators met milestones and help them translate business requirements into design.
The problem was that her loyalties were divided: she was hired by the client, but she needed the project to continue to get paid. You need some political juggling and business sense to be an external consultant (Pro-tip: lean to $CLIENT's side in that conflict), I'm sure that's way too early for you.
D) I'd like to note for anyone reading this why IMHO being a third-party freelancer isn't a good idea for OP:
ERP/CRM software for mid-large businesses usually has complex licensing, and the customization can also be complex. Both of these push businesses to work with consultancies that do the work (if they're really large or very tech oriented they may do in house). These consultancies have an existing relationship with the customer, and do not appreciate having a third party look and touch their code. I'm guessing that this isn't any different in SFDC-land.
So the 3rd party is limited to stuff done externally that the consultancy is too expensive for or doesn't want to take - maybe some BI, reporting, or even a billing system.
One can set up a business on that, but OP's SFDC knowledge is unlikely to grant him any advantage vs the many 3rd parties that can do these things cheaply. Interfacing with the ERP software is by far the lesser part of these projects, less important than being cheap, knowledge of some common technologies and being very familiar with the BI or cloud du jour.