Statistical ysis for preference test?

We conducted a preference for two flavors or soda: blackberry and goji berry. We had 40 participants that tried each kind. 23 people picked the blackberry and 17 picked goji berry. We know that more people preferred the blackberry but is it statistically preferred? We don't know the right ysis to conduct in excel to know if the blackberry was statistically preferred over the goji. Any ideas??