Potent phytochemicals can attract optimization effort before the properties that determine whether their activity can be translated into reproducible exposure, stable handling, tractable disposition, and sustainable material access are known. This article develops a proposed pre-optimization decision model for ranking such candidates across solubility, permeability, chemical stability, metabolic liability, and supply constraints without collapsing these dimensions into a single drug-likeness score. The model treats potency as necessary but insufficient evidence and distinguishes measured liabilities from predictions, intrinsic properties from context-dependent behavior, and remediable problems from liabilities for which no credible corrective route is evident. Particular emphasis is placed on interactions: increasing apparent solubility may not increase absorption if thermodynamic activity or permeability falls; structural flexibility can change exposed polarity; chemical degradation can make nominal assay concentration misleading; metabolism can disconnect parent-compound potency from the species that actually circulate; and formulation, dose, physiology, or production technology can reverse candidate order. The proposed architecture therefore ranks candidates conditionally, using evidence confidence, liability severity, interaction structure, and plausible compensability rather than universal numerical weights. Its purpose is to identify which phytochemicals justify optimization, which require targeted de-risking first, and which should be deprioritized pending new evidence. It does not predict clinical success, establish universal thresholds, or replace experimental validation, and it requires prospective testing across independent natural-product series.
