As Old Mycroft pointed out, basing the IT budget on future projected earnings is a sketchy proposition; as it includes too many variables to reasonably accommodate for the sake of planning.
What you can do, though, is base the IT budget on a per user basis. Figure out what it costs to sufficiently support one user (combination of equipment, licenses, training and avg. # of support hours x avg. $$$ of support personnel on hourly basis). Then speak with HR to get projections on head count (any HR department worth their salt will have an approximate idea of how many people they intend to hire/have on board by year end). From there, you can get an idea of what the minimum amount of $$$ you'd need would be.
Add in realistic numbers for capital improvements (replacement/addition of servers, network equipment, A/V equipment, telephony) and some 'just in case' padding, and you should be able to come up with a pretty good financial number.
From my experience, the two key benefits from this are that you're using fairly firm logistics to justify your budget (bean counters like solid math) and that if you are experiencing exponential growth, the economy-of-scale can kick in, and you can start decreasing the $$$ per user across the board (lowering costs via bulk purchasing), meaning you may end up with a budget surplus at the end of the year (barring unforeseen circumstance and assuming your head count #'s were relatively close to actual).
There are downsides, as well (especially if they don't allow you to add headcount to your staff or equipment necessary to support such rapid expansion); so the real trick is figuring out a good dollar number you can work with, and sell to your execs and accounting team.
Best of luck!