For years, plant breeders have looked at ways to improve the growth and yield of crops under different environmental conditions. Despite advances in genotyping (genetic characterisation) and phenotyping (observable trait characterisation) technologies, breeding companies are still looking for efficient phenotyping methods and ways to link phenotypic traits to genetic information. An EU-funded project called ‘Smart tools for prediction and improvement of crop yield’ (SPICY) aimed at contributing a set of tools to address the issue. The project used pepper as a model crop, and yield as an example of a complex trait. Project partners included individuals from research institutions and the breeding industry. A main aim of the project was to provide a set of molecular markers to be used in phenotype prediction in the gene-to-phenotype model. Researchers were able to produce a nearly complete library of expressed genes as well as a list of candidate genes for plant and fruit traits in pepper and other crops. A fluorescence sensor and image analysis tool also proved to be highly effective in plant phenotyping. Researchers integrated the latter into a type of imaging robot. It uses colour, range and near-infrared cameras to automatically capture images of growing plants in greenhouses without disturbing them. The team was furthermore able to develop a crop growth model capable of predicting yield for many genotypes while accounting for environmental differences. They used genetic markers to estimate the genotype-specific model parameters. Tools developed during the project will improve the selection efficiency of genotypes related to complex traits in breeding. The ultimate goal is to enhance the breeding of crop plants in support of sustainable agriculture.