This study is designed to fill any explanatory gaps that have been traditionally underexplored when considering the interplay of implicit/automatic/unconscious and explicit/deliberate/conscious attitudes in the context of celebrity endorsement. The main experiment employed single target-implicit association test as a measure of implicit attitudes. The key finding of the study is that the experimental condition where the fit between the celebrity and the endorsed product was low induced favorable implicit and explicit attitudes, similar to the condition where the fit was high. The explanation for this finding was that dissonance enhanced the association strength of the attitude object through biased attention and elaboration, which provided a basis for favorable propositions. In propositional reasoning, retroactive confirmation of the favorable implicit attitudes resolved dissonance. This study seeks to go beyond the existing endorsement and sponsorship literature developed based on matching principles such as source models, match-up hypothesis, the congruity theory, and associative network models. The authors recommend that managers search for more creative and novel partners in addition to image-matching ones.