Exploration of a germplasm collection for a particular trait is viewed as a search within a given distribution. An optimal strategy would be to search and collect additional accessions for traits as long as expected benefit is greater or equal to the cost of collecting, conserving and testing it. The probability of finding a desirable trait depends on the number of accessions that are screened for the trait, and the distribution of that trait in the collection or in certain subcollections. This study will estimate the expected net return from an additional search opportunity in regard to several pest resistance traits for soybeans both when the additional accession is from the existing collection, and when it is newly acquired.