Complex interventions are not just the sum of their parts and require a more theoretical approach to reflect their complexity. While standardisation and randomised controlled trials (RCTs) are often seen as linked, this may hinder the effectiveness of community interventions. The authors propose a shift in how large-scale interventions are conceptualised, allowing them to be more responsive to local contexts and potentially more effective while still being evaluated in controlled designs. The key is to focus on complex system functions and processes rather than simple elements. The suitability of cluster RCTs for evaluating interventions directed at whole communities or organisations remains debated. Some health promotion advocates argue that RCTs are inappropriate due to the need for standardisation, which may prevent context-level adaptation. However, the Medical Research Council (MRC) suggests that RCTs can still be appropriate if the focus is on standardising the process and function of the intervention rather than its components. The authors argue that the current view of standardisation is at odds with the notion of complex systems. They propose an alternative view where standardisation allows for interventions that may look different in different sites to be more effective and evaluated in RCTs. This involves defining the steps in the change process or key functions that the elements are meant to facilitate, rather than the components themselves. The authors also discuss the importance of defining intervention integrity based on function rather than composition. They argue that context-level adaptation does not have to mean losing integrity, and that interventions can be evaluated in RCTs if their functions are standardised. The authors conclude that the shackles of simple intervention thinking may be hard to throw off. They argue that complex systems rhetoric should not become an excuse for "anything goes." More critical interrogation of intervention logic may build stronger, more effective interventions. They also highlight the need for more studies of this type to reverse the current evidence imbalance when policymakers weigh up "best buys" in health promotion.Complex interventions are not just the sum of their parts and require a more theoretical approach to reflect their complexity. While standardisation and randomised controlled trials (RCTs) are often seen as linked, this may hinder the effectiveness of community interventions. The authors propose a shift in how large-scale interventions are conceptualised, allowing them to be more responsive to local contexts and potentially more effective while still being evaluated in controlled designs. The key is to focus on complex system functions and processes rather than simple elements. The suitability of cluster RCTs for evaluating interventions directed at whole communities or organisations remains debated. Some health promotion advocates argue that RCTs are inappropriate due to the need for standardisation, which may prevent context-level adaptation. However, the Medical Research Council (MRC) suggests that RCTs can still be appropriate if the focus is on standardising the process and function of the intervention rather than its components. The authors argue that the current view of standardisation is at odds with the notion of complex systems. They propose an alternative view where standardisation allows for interventions that may look different in different sites to be more effective and evaluated in RCTs. This involves defining the steps in the change process or key functions that the elements are meant to facilitate, rather than the components themselves. The authors also discuss the importance of defining intervention integrity based on function rather than composition. They argue that context-level adaptation does not have to mean losing integrity, and that interventions can be evaluated in RCTs if their functions are standardised. The authors conclude that the shackles of simple intervention thinking may be hard to throw off. They argue that complex systems rhetoric should not become an excuse for "anything goes." More critical interrogation of intervention logic may build stronger, more effective interventions. They also highlight the need for more studies of this type to reverse the current evidence imbalance when policymakers weigh up "best buys" in health promotion.