This is where the "lifetime value of a customer" idea comes in.

Basically, you estimate how much average profit an average customer represents over the typical course of doing business with you. For example, if you know that the average customer makes you $2000 profit over the course of your business with them, then you can count backwards to determine how much you should spend to acquire that customer. Anything less than $2000 will put you in profitable territory. So, if it costs $100 to acquire an average customer, then you can expect an average of $1500 profit from each customer. However, you must also consider time: if the average term of doing business with you is, say, 3 years, well, you would probably need a larger volume of customers to keep afloat.

Also, when thinking along these lines, its probably a good idea to leave a little margin for error, especially if you don't have a lot of data or if your data contains a lot of variance.

Hope that helps.

-Phil

P.S. - This line of thinking helps explain why cold calling, especially the high-probability prospecting way, is so effective. If you are cold calling a local area, each call costs less than ten cents -- much less than a postage stamp -- and then you only visit or send information to people who have stated that they want what you are offering. So you end up spending more on your most likely candidates and less on your least likely candidates (as it should be).