This paper presents results generated using high-resolution 3D monitoring of rock slopes and focuses upon the development of slope surface strain prior to failure. The data suggest that given sufficient measurement precision, precursory behavior, including both the rates of small rock fall activity and creep in the rock mass prior to large slope failure can be detected, measured and monitored. Combined with models of the temporal nature of slope failure evolution, for example first time failure due to brittle crack nucleation and coalescence, these data have the potential to be used to predict both the time of failure and the geometry of the mobile mass. Results are presented from monthly monitoring using 3D laser scanning of 100,000 m2 of near-vertical rock face over a three year period. During the monitoring period a dataset of in excess of 500,000 rock falls has been derived, ranging in volume from 0.0001 m3 to over 1,000 m3. The temporal patterns in the rockfall dataset suggest that the cumulative volumes of small rockfall can be treated as precursory indicators of larger failures. Environmental conditions have a perhaps surprisingly limited impact on triggering rockfall. Precursory behavior also appears scale dependant both in time and space. The behavior of the rock mass prior to the largest events recorded appears consistent, following a hyperbolic acceleration through time prior to ultimate failure. This has implications for understanding of both the conditions within the rock mass that lead to rockfalls and those environmental factors which bring the rock mass closer to failure.