Aggregation metrics in reputation systems are important for overcoming information overload. When usingthese metrics, technical aggregation functions such as the arithmetic mean are implemented to measure thevalence of product ratings. However, it is unclear whether the implemented aggregation functions match theinherent aggregation patterns of customers. In our experiment, we elicit customers aggregation heuristicsand contrast these with reference functions. Our findings indicate that, overall, the arithmetic mean performsbest in comparison with other aggregation functions. However, our analysis on an individual level revealsheterogeneous aggregation patterns. Major clusters exhibit a binary bias (i.e., an over-weighting of moderateratings and under-weighting of extreme ratings) in combination with the arithmetic mean. Minor clustersfocus on 1-star ratings or negative (i.e., 1-star and 2-star) ratings. Thereby, inherent aggregation patternsare neither affected by variation of provided information nor by individual characteristics such as experience,risk attitudes, or demographics.