So I have been supporting it by consulting, but when I'm consulting I'm not selling Appointment Reminder accounts, and consulting is always more distracting than I expect it to be. (You think I would realize this after two years of it, but I persistently underestimate how much time it takes to e.g. prospect for new clients, negotiate/schedule engagements, travel to them, and deal with follow-up stuff like invoices.)
The traditional option for scaling a software business with a working product and actual customers is to take investment, but that's another ball of wax. There's fairly attractive options in angel investing these days, but taking any outside money would have an outsized effect on the character of my business at the moment, and I'm not ready to pull that trigger yet. Also, like consulting, raising an investment round is a very distracting event -- it basically commits you to months of doing near no productive work. (And unlike consulting, taking investment causes time debt which is impossible to discharge until you exit your business: you're now committed to keeping those investors happy and in the loop for, well, forever in business terms.)
But, having seen a lot of smart software companies do side projects (37signals was a big fan of it, and Amy Hoy and company have quite a bit of success) I decided to try my hand at it once and see how it went. So I started packaging one of my most-commonly-a-win consulting offerings as an online course, and decided to experiment with it once. (And, like Brennan says and like everybody in the space will tell you, I started by making an email list and sending them lots of free stuff they enjoyed which further burnished my credibility on the subject.)
The experiment was very successful: my customers seem to have liked what I produced, many of them have actually used it to positive effect in their business, and it raised a significant amount of money. (I'll probably blog in detail about it later).
An unanticipated side effect was, because of the topic I picked for my video course, I actually sat down and took all my own advice for a change (AR had an email strategy which I would have never let fly at a consulting client simply because I shipped the minimum necessary and never revisited that decision), and that ended up working out very well.
So, basically, yes: confirmation from someone else, this does work and it is perfectly compatible with running a software business.