ProfitWell always subtracts discounts and credits before calculating MRR.
For example, a user that's on a $120 annual plan who is given a 10% discount, will have MRR of $9. ($108/12 months)
Likewise, for a user that's on a $10 monthly plan who is discounted 10%, will have MRR of $9 for the length of the discount, and increase to $10 thereafter.
We believe that having them affect your MRR is the right thing, and here are some of the reasons why that is:
- Presumably, a 12-month discount shouldn't be considered on your MRR. How about a six-month discount? A Two-month one? But say you'll draw the line at one-month, and anything over that goes into MRR. What do you do now when a forever-discount (that you already marked as a downgrade on your MRR) gets cancelled after just one month? And before answering, "well, I'd just go back re-write my historical MRR", try to imagine doing that at scale, when you're making 10, 20, 100M a year (as many ProfitWell customers do).
- A similar problem arises when you consider account credits which, for all practical purposes, are basically discounts applied retroactively. What do you do when the credit applied to an account is less than the amount corresponding to one month of service? Would you treat that discount differently than one for an amount equal to 6 months worth of service?
Because we don't have good answers to those questions, we take the position that "all credits and discounts ALWAYS go into your MRR".
On ProfitWell, we calculate the MRR of thousands of companies with MRRs ranging from a few thousand USDs/year to a couple hundred million (yes, we've got some unicorns :). To do so consistently and reliably at the scale at which we do it we need to follow a few principles, some of which apply here:
- Consistency (and predictability) is mandatory.
- Simplify as much as you can (as we need to be able to explain our rules to folks, which can get quite complicated).
- Try really, really hard to never re-state previous MRR numbers (just imagine the looks on your investors' faces when you tell them last month's growth numbers were wrong). In other words, we (almost never) re-write history (the almost being related mostly to annual subscriptions that get cancelled and refunded within a certain period of time).
We think —and talk with SaaS folks— about how to calculate metrics perhaps more than about any other topic. And when it comes to one-time discounts and its close cousin, account credits, we find out there's little consensus amongst the initiated.