How do you prove anything?  Very,very difficult from an engineering perspective.  Instead, how about “verifying a hypothesis”.  In other words… create an H0 hypothesis about your sales & marketing strategy… then; test that hypothesis, and most importantly measure the results closely.

If you can say that your hypothesis works, then do the hard part: try to scale it up!  (wow, that can be tough).

So many companies think: well I’ll get X customers, and then I’ll be at a critical mass and then my model will be Y.  The problems… how do you get to X customers?  How do you know what will happen when you get there?

Instead, I focus on getting to x customers (note the smaller x), and prove that the model works at x.. (e.g. the business is profitable at x).  Then, I slowly scale up towards X.  If Y happens, great… but my X is still my x and that’s good enough for a profitable business.

Go x yourself.