A fundamental assumption of hypothesis-driven decay experiments is that, during decay, the loss of anatomy follows a sequence broadly controlled by the intrinsic compositional properties of tissues. Recent work investigating the succession of postmortem endogenous microbial communities (thanatomicrobiome) challenges this assumption. These studies suggest the thanatomicrobiome exhibits a predictable, clock-like succession in response to physical and chemical environmental changes within a carcass. Therefore, it is possible that reproducible sequences of character loss during decay are controlled by thanatomicrobiome succession dynamics. If so, exceptionally preserved fossil anatomy would reflect a succession of ancient contemporaneous microbial communities, about which we know nothing, rendering decay experiments uninformative. Here, we investigate two questions: (1) what is the role of exogenous and endogenous bacteria during formation of the thanatomicrobiome and (2) do thanatomicrobiome successions control the sequence of anatomical character loss within a decaying carcass? Our analysis shows that the thanatomicrobiome is dominated by endogenous bacteria and that, even in the presence of inoculum, exogenous bacteria do not invade the carcass and replace native bacteria (while the carcass is intact). This confirms that the use of environmental inoculum in decay experiments introduces an inadvisable confounding variable. Secondly, we find no correlation between thanatomicrobiome successions and the sequence of anatomical character loss, supporting that fossil non-biomineralised characters correlate with their propensity to decay in extant relatives. These findings indicate that the inability to model ancient bacteria does not invalidate decay experiments. We also present a synthesis of the role of bacteria in non-biomineralised fossilisation.