The usual notions of (in)direct effects, based on nested counterfactuals, run into interpretational difficulties in the context of survival analyses when the mediatior is also measured over time. I will propose an alternative approach following an idea by Robins and Richardson. Here, mechanisms need to be specified allowing a separation of treatment components, formalized using an augmented directed acyclic graph (DAG). It can be shown that under specific assumptions identification of such separanle effects is possible, resulting in the familiar mediation formula. In continuous time, it can further be shown that for the particular case of combining a linear model for the mediator with an additive hazard model, the familiar path-tracing formula can be recovered. For illustration, I consider an application to blood-pressure treatment effect on time to kidney failure. Moreover, it will be discussed how separable effects can further be extended to competing risks. The proposed new approach is founded in decision theory, avoids genuine counterfactual assumptions and - even in non-survival contexts - constitutes an interesting alternative to the prevailing structural equation models.