To regularly improve the composition of the lettuce collection of the Centre for Genetic Resources, the Netherlands (CGN) with modern varieties, feedback from crop experts is used to select approximately 10% of the new material for incorporation in the collection. In the present study, assessments of six experts were compared to microsatellite data of 414 new varieties and 1408 existing accessions. Based on the microsatellite data, the extent to which the genetic diversity of the collection would be enriched (added value) was calculated for specific sets of new varieties. When individual assessments of experts were evaluated, the total added value of expert-based selections was not significantly higher compared to randomly chosen groups, except for a single expert. Unfamiliarity with new varieties was shown to be a crucial factor in the assessment of crop experts. According to the current acquisition protocol that seeks for consensus among experts, varieties are selected based on recommendations from at least three experts. This protocol also did not perform better than randomly chosen groups of new varieties. However, significantly better results were obtained with alternative protocols. It was concluded that breeding value was a more decisive criterion in the current acquisition protocol than maximal extension of the genetic diversity within the collection. A modified protocol addressing both commercial and diversity aspects was suggested in order to meet the demands of plant breeders as well as conservationists.