The correlation between illite/smectite (I/S) diagenesis and mean vitrinite reflectance (Ro) data is examined in mudrocks from a hydrocarbon exploration well (geothermal gradient 35°C km−1) from the Great Hungarian Plain of the Pannonian Basin System. The expandability of I/S decreases with depth and there is a change from random to ordered mixed-layering at about 2500 m depth. At this depth Ro is about 0.6%. Comparison of the correlation of expandability and Ro from this study to published data for the Vienna Basin and the Transcarpathian Basin, sub-basins of the Pannonian Basin System, shows that the correlation is systematically different for each sub-basin, according to their thermal histories. In the Vienna Basin (geothermal gradient 25°C km−1), for any given value of Ro, the expandability of I/S is less than in the Transcarpathian Basin (geothermal gradient 55°C km−1) and the sediments are older and more deeply buried. Data from the present study are intermediate. This variation is believed to be due to the effect of time on the smectite-to-illite reaction. Results of an optimization procedure to calculate the kinetics of the smectite-to-illite reaction, using as input the expandability depth profiles, and thermal histories constrained by comparison of observed and calculated Ro data, showed that I/S diagenesis in the Pannonian Basin System can be modelled by a single first order rate equation: where S = fraction of smectite layers in I/S, t = time (Ma), e = exponential function, log(A) = frequency factor = 7.5 (Ma−1), E = activation energy = 31.0 kJ mol−1, R = universal gas constant, and T is temperature in Kelvin. This result also suggests an important role for time. By combining the kinetics of the smectite-to-illite reaction with a kinetic model of vitrinite maturation it is possible to define a domain within which all ‘normal’ (burial diagenesis) correlations between Ro and I/S diagenesis should lie. Such diagrams can be used to identify different thermal histories related to different geotectonic settings and ‘anomalous’ data such as that affected by igneous intrusions.