In the present study an attempt is made to relate the structural properties of Fluid Catalytic Cracking (FCC) catalysts with the process performance. Stochastic network models are employed to represent the porous structure of two catalysts with similar catalytic activity but different pore structure based on N-2 porosimetry experiments. Accordingly, reaction and deactivation are studied by means of a transient model based on percolation theory. Extension at the reactor level is achieved by employing the mixing cell in series model. The results show that at early times the catalyst with the larger surface area and pore volume shows higher conversion and yield while, at the same time, has a lower fraction of poisoned sites in the porous network. At later stages the catalyst with the higher degree of connectivity exhibits stronger resistance to structural phase transition due to pore plugging. Comparison between experimental and simulation N-2 isotherms during catalyst deactivation shows excellent agreement between model and experiment confirming the mechanism of coke formation. Further comparison with experimental results at the reactor level shows very good agreement in terms of both conversion and yield differences between the two catalysts. The close agreement between model and experiments in both catalyst structure and reactor conversion for two entirely different structures of FCC catalysts opens up the possibility of architectural design of these catalysts for optimum process performance achievement.