Economic productivity is one of the most important concepts in economics. Increasing economic productivity will boost a society. This study intends to compare the impact of different criteria on the economic productivity of Iranian factories with differential methods and neural networks. The model estimation results using the difference-in-difference method show that exports have a significant positive effect on productivity growth. The model estimation results using the neural network method show that export has a significant positive effect on productivity growth. The presentation of export development variables in two methods of difference in difference and neural networks showed that the level of productivity could be increased well according to the predictability of the neural network.