The degree to which phenotypic heterogeneity of cellular populations contains biologically or clinically important information is not well understood. Here, we investigated whether patterns of pre-existing signaling heterogeneity in cancer cell populations could reveal information about drug sensitivity. We modelled cellular heterogeneity as a mixture of stereotyped signaling states, identified based on colocalization patterns of activated signaling molecules from microscopy images. We found that patterns of heterogeneity could be used to separate the most sensitive and resistant populations to paclitaxel within a set of H460 lung cancer clones and within the NCI-60 panel of cancer cell lines, but not for a set of less heterogeneous, immortalized non-cancer human bronchial epithelial cell (HBEC) clones. Thus, tractable and useful information can be obtained from analysis of heterogeneity.