Coffee is one of the most consumed beverages in the world and its consumption sweetened with sucrose is common in the brazilian diet. The aim of this study was to determine the ideal concentration of sucrose in espresso coffee samples with and without caffeine. Espresso coffee (ECO) and decaffeinated espresso coffee (DEC) samples were prepared and sweetened with five different concentrations of sucrose, for each coffee. The samples were presented in monadic form in complete balanced blocks. For the sweetness considered ideal for espresso, studies were conducted with 120 consumers using 9cm unstructured line JAR scale (- 4.5 = extremely less sweet than ideal, 4.5 = extremely sweeter than ideal) using FIZZ Sensory Software, Biosystemes version 2.47B. The obtained data were analyzed statistically by analysis of distribution histogram of sensory responses and simple linear regression between the hedonic values and the concentration of sucrose. The ideal concentrations were 9.17% and 7.47% for ECO and DEC, respectively. The absence of caffeine in the DEC sample contributed to the lower sucrose ideal concentration. It is important to set boundaries for sucrose consumption, since excessive consumption is related with some diseases.