To identify natural and original kinase inhibitors from plant extracts, we have developed and compared a heterogeneous enzyme-linked immunosorbent assay (ELISA) and a homogeneous time-resolved fluorescence (HTRF, Cisbio International, Bagnols/Cèze, France) assay. Kinase affinity for the ATP substrate was determined in both assays, and the same [ATP]/ATP Km ratio was used in each case to enable the identification of ATP competitive and noncompetitive inhibitors. Assays were then used to screen the same collection of chemical compounds and plant extracts. The intra-assay correlation analysis of each technology showed a very good screening precision in HTRF and an acceptable one in ELISA. When the two methods were compared, a poor correlation was obtained with a higher hit rate in the ELISA. We then performed a detailed study of the ELISA hits and showed that they also presented a strong antioxidant activity, associated with high adsorption into microplate wells, which interfered with the horseradish peroxidase-based detection system. These hits were then flagged as false-positives. We also showed that many plant extracts presented this kind of activity and that this interference could explain the lack of correlation between the assays. These findings suggest that assay design should be carefully adapted to the substances to be screened and that interferences should be extensively considered before any assay development process and comparison studies. In spite of a few interferences, our results showed that a homogeneous-phase assay like the HTRF assay could be more efficiently used for plant extract screening than a heterogeneous-phase assay like ELISA.