The article deals with the issue of phenomenal data mining, the essence of data mining with commonsense knowledge. Data mining involves computer programs that infer relations among different kinds of data in large databases. The goal of this practice is to infer useful relations that might not have been noticed or at least could not have been confirmed among the available data. Information usually admits a more compact description in terms of the underlying phenomenon than in terms of the data. Developing systems employing phenomenal data mining initially requires building knowledge of general phenomenon into the data miners. This will permit data miners to use databases of commonsense facts about the world. The article demonstrates the methodology to build such a data mining system with the help of a supermarket model. The most useful information to supermarket companies doing data mining will involve probabilities of response of different kinds of customers to different stimuli. It presents some ideas about the process to develop such a system