We develop a new method for analysing fault slip-rate data sets. The method converts long- and short-term rate estimates derived from geological and geodetic data into estimates for temporally localized rate averages. These averages provide the closest approximation to the instantaneous slip-rate history afforded by the given data. The method provides a means of quantitatively comparing and contrasting results obtained by different measurement techniques that fully accounts for the possibility of temporal variations in slip rate. We demonstrate the utility of the method using numerical simulations representing rate variations over late Pleistocene, and Holocene timescales. We find that a typical slip-rate data set comprising geodetic, palaeoseismologic, geomorphologic and/or structural geological constraints allows for reconstruction of instantaneous slip-rate history with resolution that decreases with time before present. Resolution for recent time periods (≤100 ka) is fairly high. The method should be useful for studies of the earthquake cycle and earthquake recurrence, repartitioning of slip rates within broad plate boundary zone fault systems over long-time intervals, and related problems, and could also be useful for identifying possible discrepancies arising from invalid assumptions and other model errors associated with the estimation of slip rates from basic geological and geodetic observables.