This article describes a dataset providing temporal sensory descriptions and preferences for four lemon-flavoured carbonated alcoholic drinks. The recruited Japanese consumers (97 men, 96 women) corresponded to the target for this kind of drink: aged between 20 and 40 and regular consumers of flavoured alcoholic drinks. They had to consume a whole can of each drink at home, each on a different day. For sips 1, 4 and 7, they had to check from a check-all-that-apply (CATA) list of eight attributes (alcohol, bitter, carbonated, lemon, refreshing, sour, sweet aroma and sweet taste) that were applicable during three periods of perception – “in mouth before swallowing”, “immediately after swallowing” and “aftertaste”. They were separated into two panels: the consumers in panel SIM (96 consumers) had to do the task simultaneously with the tasting, while the consumers in panel RET (97 consumers) had to do it retrospectively. They also had to rate their liking and report the number of crackers they consumed during the tasting. Once the can had been fully consumed, they had to score their satisfaction level and optionally report comments about the products and the task. The data were used to compare retrospective and concurrent temporal evaluations in a methodologically oriented article entitled “Concurrent vs. immediate retrospective temporal sensory data collection: A case study on lemon-flavoured carbonated alcoholic drinks.” The data could be reused by researchers interested in understanding interactions between alcohol, carbonation, sour, sweet and bitter or to relate temporal perception and preferences for improving product formulation.