Phenotypic chemogenomics studies require screening strategies that account for the complex nature of the experimental system. Unknown mechanism of action and high frequency of false positives and false negatives necessitate iterative experiments based on hypotheses formed on the basis of results from the previous step. Process-driven High Throughput Screening (HTS), aiming to "industrialize" lead finding and developed to maximize throughput, is rarely affording sufficient flexibility to design hypothesis-based experiments.In this contribution, we describe a High Throughput Cherry Picking (HTCP) system based on acoustic dispensing technology that was developed to support a new screening paradigm. We demonstrate the power of hypothesis-based screening in three chemogenomics studies that were recently conducted.