I'm about to launch a startup which if it works well will distribute experiences, moving people away from all getting the same recommendations for the same plac

I'm about to launch a startup which if it works well will distribute experiences, moving people away from all getting the same recommendations for the same places because some algorithm created the recommendation. The problem with crowd sourced averages is that the curve is exponential. If there are 2 places with 4 reviews and one has a 3.5 rating and the other a 4.5 rating more people will go to the 4.5 rated place.The increase in traffic becomes a self-fulfilling cycle of validation. Bibimapp is a social place recommendation engine using the follower / following model. You can signup for early beta access here: bibimapp.com