During bone remodelling, bone cells are thought to add and remove tissue at sites with high and low loading, respectively. To predict remodelling, it was proposed that bone is removed below and added above certain thresholds of tissue loading and within these thresholds, called a 'lazy zone', no net change in bone mass occurs. Animal experiments linking mechanical loading with changes in bone density or microstructure support load-adaptive bone remodelling, while in humans the evidence for this relationship at the micro-scale is still lacking. Using new high-resolution CT imaging techniques and computational methods, we quantify microstructural changes and physiological tissue loading in humans. Here, we show that bone remodelling sites in healthy postmenopausal women strongly correlate with tissue loading following a linear relationship without a 'lazy zone' providing unbiased evidence for load-driven remodelling in humans. This suggests that human and animal bones both react to loading induced remodelling in a similar fashion.