The paper introduces the g-index as an improvement to the h-index, designed to better measure the global citation performance of a set of articles. The g-index is defined as the largest number such that the top g articles together receive at least g² citations. The author proves the unique existence of g for any set of articles and shows that g ≥ h, where h is the h-index. The g-index is further analyzed within the context of Lotkaian theory, providing a formula for g in terms of the Lotkaian exponent α and the total number of sources T. The paper also compares the g-index and h-index of the (still active) Price medallists, demonstrating that the g-index better captures the citation scores of the top articles and provides a more accurate distinction between scientists based on their visibility. The g-index is shown to be more intuitive and aligns better with the raw data compared to the h-index. The paper concludes by discussing the potential for further exploration of the g-index, including its time-dependent version and the construction of other h- or g-like indices.The paper introduces the g-index as an improvement to the h-index, designed to better measure the global citation performance of a set of articles. The g-index is defined as the largest number such that the top g articles together receive at least g² citations. The author proves the unique existence of g for any set of articles and shows that g ≥ h, where h is the h-index. The g-index is further analyzed within the context of Lotkaian theory, providing a formula for g in terms of the Lotkaian exponent α and the total number of sources T. The paper also compares the g-index and h-index of the (still active) Price medallists, demonstrating that the g-index better captures the citation scores of the top articles and provides a more accurate distinction between scientists based on their visibility. The g-index is shown to be more intuitive and aligns better with the raw data compared to the h-index. The paper concludes by discussing the potential for further exploration of the g-index, including its time-dependent version and the construction of other h- or g-like indices.