For sake of simplicity in predicting the shelf life of biologically stable products, it is often the case that their thermal history is not taken into account. In many cases, however this approximation is not appropriate and the shelf life predicted can be quite different from the actual one. This paper presents three different examples of shelf life prediction of soft drinks aimed at showing the error that one can make by underestimating the importance of the bottle thermal history. In the first example, the shelf life of the beverage was estimated assuming that the storage temperature was constant and equal to room temperature for the entire storage period In the second case it was assumed that the temperature of the bottle varies during the storage period but for sake of simplicity, in performing the calculation the temperature was kept constant and equal to the average temperature of storage. In the last case the situation in which the temperature of the bottle of carbonated beverage varied during the storage period and during the day was examined. It was shown that under conditions comparable to that occurring during distribution the shelf life estimated differs significantly from that predicted by neglecting the temperature rise due to outdoor storage and sunlight exposure.