This paper analyses monetary transmission mechanism in Tunisia based on two approaches, an aggregate data analysis by using a Structural Vector Auto regressive (SVAR) model to assess the impact and the delay of transmission of monetary policy decisions and to identify through which of the interest rate channel or credit channel, monetary policy stances’ changes could affect the economy; and, a bank panel data analysis by employing an ARDL model to measure the reaction of the banks’ pricing policy to monetary policy changes. For the SVAR model, a “recursive” system was used to uncover the dynamic effects of monetary policy shocks. The empirical results show that the interest rate channel was more effective than the credit channel and that’s from the 8th quarter. For the ARDL model, the empirical results show that, taken into consideration of the heterogeneity of the banking system landscape, the banks pricing’s policy are highly dependent upon money market rate’s changes. In other words, the transmission to lending rates applied to households as well as to firms is almost complete.