To reduce the level of redundancy in a collection of cultivated lettuce, data from 160 amplified fragment length polymorphism (AFLP) fragments and 10 polymorphic microsatellites were used in combination with passport data and morphological data, the latter obtained from an experimental field trial performed for verification purposes. Based on the observed distribution of the number of marker differences between and within accessions, a minimum of three AFLP differences and two microsatellite differences were regarded as levels warranting distinction between accessions in the redundancy analysis. The strategy followed in the redundancy analysis was mainly based on the confirmation of duplication by each of two independently generated data sources. The molecular data were used for the validation as well as the identification of potential duplicates, revealing a total number of 198 redundancies, corresponding to 12.9% of the total collection. Trueness to type, number of characterization and evaluation data, and collection management considerations, such as available seed quantities and germination percentages, were used as primary, secondary and tertiary criteria to decide which accession from duplication groups to maintain in the collection. Removal of accessions showed negligible effects on total collection diversity, as quantified for AFLPs and microsatellites, characterization and evaluation traits and resistance profiles against downy mildew pathotypes, indicating that the applied strategy was effective.