In this talk I will summarise the findings presented in a series of four papers dedicated to the study of early type galaxies (ETGs) with integral field spectroscopy (IFU) from the MaNGA survey. The formation channels and mass assembly of ETGs is still a matter of debate in current galaxy evolution models. The combined analysis of galaxy kinematics and stellar population gradients (age, metallicity, alpha-enhancement, initial mass function -IMF-) is a powerful tool to disentangle between in situ star formation or merger events. We divide our sample into elliptical fast rotators (E-FR), elliptical slow rotators (E-SR) and lenticular (S0). To achieve enough signal to noise to reliably measure IMF variations, we need to stack the galaxies in bins of luminosity and central velocity dispersion. We observe significant differences between the three sub-samples — e.g., E-FR are much younger and metal rich than E-SR — , suggesting that kinematics and morphology have a strong impact on galaxy content. In addition, I will show that there are three mass-scales where scaling relations change slope and for which there is a dramatic change in the number counts and stellar population gradients of ETGs. I will also discuss the implications of IMF variations within galaxies and how this helps reconciling stellar mass with dynamical mass estimates. Finally, I will show that the thickness of the Fundamental Plane depends strongly on morphology. If the sample only includes E-SRs, then the observed scatter is significantly reduced.