Suggestion: make something that will let you charge on a recurring basis. (I once thought BCC could never sustain that. It probably could, if I had a mind to implement it, but the pain involved isn't worth it to me.)
Starting from $0 revenue on the 1st day of every month sucks. Starting from "I'll always have at least 90% of the revenue I did last month" is, on the other hand, a wonderful thing for the business in every possible way.
This is, for better or worse, exactly the opposite of how customers actually think.
This goes back to the concept of reducing risk. Customers are very risk-averse, so things like charging a lower monthly fee or offering a money-back guarantee will often give you a significant increase in sales.
Do you mind if I ask why not? It seems like something that should make a whole load more money, and if it puts people off, changing the code to say is_subscription_valid() {return TRUE;} and switching back should be trivial.
Implementing in the first place shouldn't be too painful if you can steal some code from Appointment Reminder.
I know I'm underestimating the total amount of work here, but is it really going to take more than a couple of days? Unless there's crazy invoice integration work required, in which case I understand - that's a painful world!
Edit: although I appreciate that consulting rates way outstrip revenues from SaaS a lot of the time. Still, it's nice to have some income whilst you're on holiday ;)
Call it, oh, two weeks of work. Two weeks at my consulting rate is... umm... a more reliable method of getting a fairly sizable chunk of money up-front, if I am interested in getting more money right away. Two weeks of work on AR, by comparison, gets me fairly little money right now but gets me much closer to where I eventually want to be than the same marginal work on BCC. (e.g. I did two weeks of work on AR in the last two months, and increased paying customers by 50%. Not like rate of acquisition, like total. Feel free to guesstimate how much that is worth to either a) my income in 2013 or b) the implied value of owning 100% of the shares in AR. It's a heck of a lot more than either a) how much money BCC could reasonably make in 2013 or b) the delta in implicit sale values of BCC. There are other projects I could spend two weeks on, with a variety of professional and personal goals attached, which all sound more fun to me.
As of September 2012, BCC is a fun laboratory I get to play in (and draw stories from) more than it is a core line of business for me.
Additionally, I think a recurring fee would scare customers off in this market, and I go out of my way to tell them it's a one time fee and that they get access forever. If you're wondering, "Do some of them feel a sense of entitlement for essentially... forever?" Oh boy yes, they do!!! But that's OK, I treat them real nicely anyhow.
Ideally, you want your customers' usage to align with more money to you over time, not less.
If you need money now, offer your customers a discount for quarterly/annual prepayments. You're more able to predict your costs out 12 months, so you're less likely to make a fatal mistake here.
Charging a one-time fee is evil for the customer with common sense. If your service stops growing, what about the customers that already paid for it?
You don't pay for updates in a SaaS app as it's in the cloud, so... I can only truly see a recurring model working. UNLESS you know very well your userbase and you know very deeply it's the only model that would work.
Annuity tables (or the annuity formula) may be helpful here.
