The existence of a cocrystal between curcumin (CUR) and phloroglucinol (PHL) was suspected but could not be demonstrated in a recent systematic effort to synthesize novel curcumin cocrystals. We hypothesize that the elusive CUR-PHL cocrystal is a kinetically stable form that can be prepared by trapping it using a fast solvent removal crystallization process. The polarity of crystallization solvent and relative solubility of cocrystal formers in the solvent appear to be critical parameters governing the phase purity of the resulting cocrystal. Organic solvents of higher polarity and in which the cocrystal formers exhibited congruent solubility tended to afford a purer cocrystal phase. Essentially phase-pure CUR-PHL cocrystals were successfully obtained from acetone, and their 1:1 stoichiometry was confirmed by differential scanning calorimetry and solid state nuclear magnetic resonance spectroscopy. Compared with the individual component phases, the cocrystal displayed reduced hygroscopicity and improved tabletability. However, the intrinsic dissolution rate of the cocrystal showed no significant improvement compared with the pure CUR crystal due to the instantaneous conversion of the cocrystal to CUR at its surface upon contact with the dissolution medium.