The discovery of cholestane in animal fossils from the Ediacaran (571-541 million years ago) has generated much excitement, but it is not the only interesting biomarker recovered. Coprostane, a geologically stable form of coprostanol, has also been found in Ediacaran rocks. This is surprising, since coprostanol is typically used in modern settings as an environmental biomarker for humans and other mammals, who produce the compound with help from bacteria in their gut. The prevailing hypothesis is that an abundance of coprostane in some Ediacaran fossils-particularly Dickinsonia-represents the degradation of the organism's cholesterol by bacteria in the microbial mat, comparable to what is seen in modern vertebrate corpses as they decompose. However, this hypothesis assumes coprostanol-producing bacteria were absent in the guts of Ediacaran organisms, and to date no one has tested whether such bacteria exist in modern invertebrates. In this study, we assembled 115 metagenomes to look for evidence of coprostanol-producing enzymes in invertebrate microbiomes. Ultimately, we did not find any evidence for the enzyme in any invertebrate microbiomes, supporting the hypothesis that coprostane is not a gut biomarker for Ediacaran animals. However, a reassessment of coprostane/cholestane ratios shows Dickinsonia was unique in coprostanol enrichment, with ratio levels comparable to waste polluted marine waters and modern vertebrate feces. While we cannot rule out the possibility of contamination, we prefer a novel interpretation of the coprostane signature in dickinsoniomorph fossils, where the elevated level of coprostanol comes from digestion of the microbial mat and concentration of the biologically inert compound. If correct, the elevated coprostanol signal provides new insights into the feeding strategy of these enigmatic animals.