1.Traditionally, trait‐based studies have explored single‐trait‐fitness relationships. However, this approximation in the study of fitness components is often too simplistic, given that fitness is determined by the interplay of multiple traits, which could even lead to multiple functional strategies with comparable fitness (i.e. alternative designs). 2.Here we suggest that an analytical framework using boosted regression trees (BRT) can prove more informative to test hypotheses on trait combinations compared to standard linear models. We use two published datasets for comparisons: a botanical garden dataset with 557 plant species (Herben et al., 2012) and an observational dataset with 83 plant species (Adler et al., 2014). 3.Using the observational dataset, we found that BRTs predict the role of traits on the relative importance of survival, growth, and reproduction for population growth rate better than linear models do. Moreover, we split species cultivated in different habitats within the botanical garden and observed that seed and vegetative reproduction depended on trait combinations in most habitats. Our analyses suggest that, while not all traits impact fitness components to the same degree, it is crucial to consider traits that represent different ecological dimensions. 4.Synthesis:The analysis of trait combinations, and corresponding alternative designs via BRTs, represent a promising approach for understanding and managing functional changes in vegetation composition through measurement of suites of relatively easily measurable traits.