Some variation of this question comes up pretty regularly.  Below is a copied/pasted response I gave to a colleague via NCURA Collaborate message boards to a similar question for departmental RAs.  The first paragraph addresses the flaws in using research $ to determine FTE needs; the rest deals with factors I use to determine workload balance. 

TLDR: I agree with Becky.

As far as I know, we do not have any institutionally used metrics; each area on campus uses the reporting tools available to measure what they think is important.  All I can really provide you with is my opinion on the subject.  Personally - and no disrespect is meant to anyone who uses this as a measure - I do not see any value whatsoever in considering dollars when determining workloads for RAs.  In my experience, the dollar amount of the award has almost no correlation with how much RA time and effort that award consumes.  At UF, we have multi-million dollar federal grants that practically run on auto pilot; the majority of the budget is salary for personnel who remain constant across years, perhaps a large equipment purchase, and then other basic travel and supply expenses.  These awards require minimal interaction from the RA, are consistent across years, and usually close with little drama.  Meanwhile, we have state funded awards that bring in a few thousand dollars, yet are incredibly labor intensive to manage.  They often involve cost share of faculty time since they are too small to cover it, some have bizarre budgetary restrictions, some of them require us to provide receipts for every charge at invoicing, as well as state-mandated forms.  Measuring workload by awarded dollars has the potential to negatively impact units and their RAs.  I have seen departmental administrators deny requests for additional FTE based on a comparison of awarded dollars (i.e. Unit A has a portfolio valued similarly to that of Unit B and therefore should be sustainable with similar FTE numbers).  The issue here, again, is that the awarded dollars is not a 1:1 translation to effort required to manage. 

Sorry for the long post, it is something that I have run into time and time again and I feel strongly about it.  All that said, here are the factors (also imperfect) that I use to measure and compare workload over a certain period of time (typically one year):

# of proposals submitted

# of current active projects (including cash based funds)

# of faculty managed

# of people on payroll (especially important if RA is also responsible for effort reporting)

# of modifications (i.e. post-award changes that require communication with the sponsor)

# of departments/units (if in a shared services environment)

# of various fiscal transactions (pcard approvals, reimbursements, purchase orders, etc. - if RA is responsible for fiscal functions)

In addition to these numbers, one must consider some subjective items:

- Skill level of RA; what takes an RAI all afternoon to complete, an RAIII may be able to do in an hour

- Skill level of other RAs in a team environment; is the RA spending time training/mentoring less experienced RAs?

- Personalities of faculty in portfolio; there are faculty who take up an hour of your time every encounter and others that you may never talk to

- Customer service in general; is the RA the main point of contact for multiple outside offices? 

- Continued education/professional development; does the RA make efforts to keep up with changing regulations?  are they involved with NCURA or on-campus training opportunities?

- Service to institution; does the RA contribute to campus efforts?  are they leading trainings, serving on focus groups/beta groups for changes affecting research administration on campus?

If you're still with me, thanks for reading.  I hope you find something in here helpful.  Good luck!