Biostimulant testing is often marred by low sample size, low number of plant parameters and manually taken endpoint measurements which make the discovery pipeline error prone. Early or small effects might be missed, leading to false negatives and removal of potentially effective product candidates from the pipeline. In addition, biostimulant effects on plant growth might not be captured by the few plant parameters measured traditionally.

Recent progress in sensor technology has allowed a shift in plant phenotyping from manual, laborious plant measurements to non-destructive, automated phenotyping systems. However, digital phenotyping comes with its own challenges from the implementation of digital plant phenotyping, infrastructure, experimental setup, environmental effects, data quality and quantity, image analysis, parameter calculation and validation to data integration and more. We will discuss challenges, potential solutions and what researchers need to consider when switching to digital phenotyping to fully use the advantages these technologies provide. E17As an example, high quality data and automated time series in open greenhouses, comparable to field like conditions, complemented by a large number of automatically calculated plant parameters and tools to easily integrate data into customer data pipelines can improve candidate selection and accelerate product development.