<div>Titanium dioxide is by far the most utilised semiconductor material for photocatalytic applications. Still, it is transparent to visible-light. Recently, it has been proved that a type-II band alignment for the rutile−anatase mixture would improve its visible-light absorption.</div><div>In this research paper we thoroughly characterised the real crystalline and amorphous phases of synthesised titanias – thermally treated at different temperatures to get distinct ratios of anatase-rutile-amorphous fraction – as well as that of three commercially available photocatalytic nano-TiO2. </div><div>The structural characterisation was done via advanced X-ray diffraction method, namely the Rietveld-RIR method, to attain a full quantitative phase analysis of the specimens. The microstructure was also investigated via an advanced X-ray method, the whole powder pattern modelling. These methods were validated combining advanced aberration-corrected scanning transmission microscopy and high-resolution electron energy-loss spectroscopy. The photocatalytic activity was assessed in the liquid- and gas-solid phase (employing rhodamine B and 4-chlorophenol, and isopropanol, respectively, as the organic substances to degrade) using a light source irradiating exclusively in the visible-range.</div><div>Optical spectroscopy showed that even a small fraction of rutile (2 wt%) is able to shift to lower energies the apparent optical band gap of an anatase-rutile mixed phase. But is this enough to attain a real photocatalytic activity promoted by merely visible-light?</div><div>We tried to give a reply to that question.</div><div>Photocatalytic activity results in the liquid-solid phase showed that a high surface hydroxylation led to specimen with superior visible light-induced catalytic activity (i.e. dye and ligand-to-metal charge transfer complexes sensitisation effects). That is: not photocatalysis <i>sensu-strictu</i>.</div><div>On the other hand, the gas-solid phase results showed that a higher amount of the rutile fraction (around 10 wt%), together with less recombination of the charge carriers, were more effective for an actual photocatalytic oxidation of isopropanol.</div>