As TiO2 photocatalysis has been widely investigated for its environmental application, many researchers aim at development of new and/or modified photocatalysts showing higher activities. The assessment of the activity of photocatalysts is an integral part in the development and characterization of various photocatalytic materials. However, how the photocatalytic activities can be represented and quantified and how the activities of different photocatalysts should be compared are not straightforward. In this study, we tried to reveal the multi-dimensional nature of the photocatalytic activity in a systematic way. The photocatalytic reactivities of 8 commercial TiO2 samples (both anatase and rutile) were quantified by employing 19 test substrates (phenols, organic acids, amines, chlorohydrocarbons, dyes, inorganic ions, etc.) in terms of their degradation rates. The measured activities of photocatalysts exhibited complex behaviors that depend on what compound was used as a test substrate. Even the relative activity order among the tested photocatalysts was markedly influenced by the kind of test substrates. This highly substrate-specific activity of TiO2 photocatalysts hinders the straightforward comparison of which sample is better than others. No single substrate for the activity test can represent the activity of a specific photocatalyst in a comprehensive sense. Although there seems to be no simple correlation between the activity and the property of photocatalysts, the substrate-specific activity will be analyzed and discussed in terms of various parameters such as surface area, crystallinity, surface charge, and substrate adsorption. The reaction rates for DCA and iodide oxidation are proportional to the surface area of TiO2 while the rates for phenol, RhB, and MB degradation are rather inversely proportional to the surface area. There was also rough correlation in photocatalytic activities among test compounds which have similar adsorption behavior. Meanwhile, the effect of surface charge and the particle size of TiO2 turned out to be less significant at given conditions resulting in no correlations with photocatalytic activities. This study clearly indicates that the photocatalytic activity assessment and standardization should be cautiously adjusted to the specific purpose and the condition of photocatalyst applications.