Advergames are increasingly used by companies for online advertising campaigns. However, the effectiveness of this unique advertising strategy has not been investigated in the IS research literature. This study sheds light on the effectiveness of advergames by introducing three design elements of an advergame: interactivity (the extent to which a player has control over the choice of viewing the advertising message), relevancy (the extent to which the game context identifies with the advertising information), and expectancy (the extent to which the advergame form is within the expectation of the player). Two dependent variables, attitude toward advergame and attitude toward brand, are used to evaluate the effectiveness of the advergame. Based on the transportation theory, we hypothesized two-way interaction effects between each pair of the three independent variables on the attitude toward the advergame, and also main effects of the dependent variables on the attitude toward a brand. A 2*2*2 factorial design experiment in a virtual world environment was used to test our hypotheses. The results indicate that in the condition of high relevancy, a more favorable attitude toward advergame was found for both high interactivity and low expectancy. Contrary to our hypothesis, in the low interactivity condition, high expectancy generates a more positive attitude toward advergame. Interestingly, interactivity was the only independent variable found to have a significant effect on the attitude toward brand.