Environmental DNA (eDNA) analysis is a promising tool for non-disruptive and cost-efficient estimation of species abundance. However, its practical applicability in natural environments is limited owing to a potential gap between eDNA concentration and species abundance in the field. Although the importance of accounting for eDNA dynamics, such as transport and degradation, has been discussed, the influence of eDNA characteristics, including production source and cellular/molecular state, on the accuracy of eDNA-based abundance estimation was entirely overlooked. We conducted meta-analyses using 44 of previous eDNA studies and investigated the relationships between the accuracy (R²) of eDNA-based abundance estimation and eDNA characteristics. First, we found that estimated R² values were significantly lower for crustaceans and mussels than fish. This finding suggests that less frequent eDNA production of these taxa owing to their external morphology and physiology may impede accurate estimation of their abundance via eDNA. Moreover, linear mixed modeling showed that, despite high variances, R² values were positively correlated with filter pore size, indicating that selective collection of larger-sized eDNA, which is typically fresher, could improve the estimation accuracy of species abundance. Although our collected dataset was somewhat biased to the studies targeting specific taxa, our findings shed a new light on the importance of what characteristics of eDNA should be targeted for more accurate estimation of species abundance. Further empirical studies are required to validate our findings and fully elucidate the relationship between eDNA characteristics and eDNA-based abundance estimation.