My recommendation would be to outsource for as long as possible and focus your time and energy elsewhere. Build the simplest, least expensive solution possible and test whether there is sufficient marketplace demand for the concept before investing one dime more than necessary. Cheers, Frank