Biostimulants are a very diverse family of products with a lot of different formula. Its current definition by the EU regulation (2019) is “a product that stimulates plant nutrition processes independently of the product's nutrient content, with the sole aim of improving one or more of the following characteristics of the plant or the plant rhizosphere : (a) nutrient use efficiency; (b)tolerance to abiotic stress; (c) quality traits; or (d) availability of nutrients confined in the soil orrhizosphere”. Although climatic conditions can be controlled in some greenhouses, this is not thecase in all companies, and there are some growing incidents. Biostimulants could be tools used to enhance crop performance or to reduce climatic stresses.Facing a huge number of commercial products, farmers are claiming for independent effectiveness tests and guidelines to use biostimulants. ASTREDHOR Sud-Ouest is working on this purpose for a long time in its greenhouses, testing different products on cyclamen, poinsettia or tomato jointly with UMR 1332 BFP. The aim of the trials led since 2016 were to assess the effectiveness of the products according to company claims in comfortable crop conditions at first. This work enables to highlight good results certain years, but it is difficult to generalize or to observe stable effects. To address this issue, we carried out multi-year factor analysis to identify a recurring effect of certain biostimulants or certain families of products. Secondly, we complemented 2022 phenotypic data with Near Infrared Spectroscopy (NIRS) measurements. Then, Crop performance was modelled using Machine Learning tools (GLMNET).Finally, the multi-year statistical analyses showed a major effect of the climatic conditions on thePoinsettia development and did not highlight the effectiveness of a particular biostimulant category. However, machine learning tools made it possible to predict the biostimulant effectiveness using plant canopy surface as proxy, thanks to NIRS data. These results should be confirmed with an independent dataset. This approach will be developed on cyclamen and tomato trials to exploit the wealth of data collected. It would permit to confirm Poinsettia results and the interest of NIRS fingerprinting to assess the effectiveness of biostimulants.Read less <