Two commercial FCC catalysts were investigated to explore the effect of their different accessibility on the catalyst activity, selectivity and deactivation due to deleterious feed metals (V and Ni). Feed metal Fe was not included in the present study. E-Cats (equilibrium samples from a commercial FCC unit) of both FCC catalysts and the corresponding laboratory-deactivated samples (applying the cyclic deactivation (CD) and the cyclic propylene steaming (CPS) methods) were thoroughly studied. Extensive characterization (XRD, N-2 physisorption, measurement of Akzo Accessibility Index (AAI), SEM-EDS analysis) of all samples was realized to investigate variations in their crucial properties due to metal deposition. Comparison of E-Cats, CD and CPS samples revealed a very different nickel deposition profile over the CPS samples. In all cases, V was homogenously distributed throughout the particle, as expected due to its mobility. Ni-Al mixed phases, observed on the E-Cat samples, were probably formed during ageing and are expected to be inactive. The absence of such phases on the laboratory-deactivated samples can be attributed both to the inability of the two deactivation methods to simulate metal ageing during commercial utilization of the FCC catalyst and the absence of Fe incorporation during laboratory deactivation. All catalytic samples (E-Cats and artificially deactivated FCC catalysts) were evaluated in the laboratory using two bench-scale Microactivity Test (MAT) units of different reactor configuration: fixed-bed (SCT-MAT unit) and fluid-bed (AUTOMAT unit). Similar ranking of the catalysts is achieved when using both units. However, AUTOMAT unit seems to provide a clearer diversification of catalysts with different accessibility. Both laboratory deactivation methods seem to be rather inefficient in simulating the real deactivation, since they are always exaggerating metal effects. (C) 2007 Elsevier B.V. All rights reserved.