Configuring products or environments for the size of their human users requires the consideration of several characteristics of the target user population, including body dimensions (anthropometry) and preferred interaction. Users are both adaptable and imperfect observers, which often makes it difficult for them to distinguish between candidate designs. This insensitivity is described by a concept called ‘just noticeable difference’, or JND. This paper presents an implementation of JND modelling and demonstrates how its use in the sizing of products or environments for target user populations can improve expected performance. Two facets of this problem are explored: (1) how experimental measures of JND for dimensional optimisation tasks may be obtained, and (2) how measures of JND may be included in models of user-device interaction for both adjustable and discretely sized products and the assumptions required. A case study demonstrating the collection and modelling of JND for a simple univariate problem is also presented. Practitioner Summary: Since people are adaptable and imperfect observers, there exists a ‘just noticeable difference’ that can be considered when designing products and environments. When JND is modelled for a target population, less variability in design dimensions due to physical user requirements may be necessary. This paper considers JND in quantitative simulations of population accommodation.