How to explain rescaled recommendation scores to end-user?

I’ve constructed a recommendation system based on both boolean and linear features that presents products to potential buyers. The raw scores are not easily human digestible (i.e. the highest score might be 0.41 on a 0 to 1 scale). What is the best way to communicate the contribution of underlying features that comprise the original score if the final score is re-scaled to a more human readable range (0 – 100%)?