Understanding tensions in the Planck data

In 2015, the Planck collaboration published results on cosmological parameters based on the Planck second data release. The Planck preferred best fit model is somewhat different from the one obtained using the WMAP satellite data or the SPT telescope data, and presents interesting mild tensions with other astrophysical datasets (e.g. direct measurements of H0), while being in very good agreement with others (e.g. the latest BAO measurements from BOSS). It has been advocated that some of these differences might be driven by inconsistent results of the Planck measurements at small scales.
In this talk, I will review some of the aforementioned differences and tensions and present our recent work to assess the consistency of the Planck data. I will then discuss the features in the Planck power spectra that lead to shifts in cosmological parameters, arguing that these are consistent with statistical fluctuations.