Pd/In2O3 has been shown to be a promising catalyst used for hydrogenation of CO2 to methanol. Characterisation of the Pd/In2O3 pre- and post-reaction combined with theoretical modelling, pointed to the proper anchoring of palladium clusters to In2O3, permitting particularly improved H2 dissociation. In contrast, in situ XAFS data has shown that In and Pd go onto form a bimetallic nanoparticulate species during reaction. Some of our recent work, has shown that it is possible to create supported Pd/In (on Al2O3) with a similar level of CO2 hydrogenation performance to Pd/In2O3. This work provides us with an opportunity to examine the nature of Pd/In species for optimal methanol yields since the In data in particular will be less dominated by In2O3. We propose to perform XAFS measurements during activation (in H2) and under operando conditions (during CO2 hydrogenation) to follow the structure of Pd and In speciation and to correlate this with performance (activation vs. deactivation).