Several light pollution indicators are commonly used to monitor the effects of the transition from outdoor lighting systems based on traditional gas-discharge lamps to solid-state light sources. In this work we analyze a subset of these indicators, including the artificial zenithal night sky brightness in the visual photopic and scotopic bands, the brightness in the specific photometric band of the widely used Sky Quality Meter (SQM), and the top-of-atmosphere radiance detected by the VIIRS-DNB radiometer onboard the satellite Suomi-NPP. Using a single-scattering approximation in a layered atmosphere we quantitatively show that, depending on the transition scenarios, these indicators may show different, even opposite behaviors. This is mainly due to the combined effects of the changes in the sources spectra and angular radiation patterns, the wavelength-dependent atmospheric propagation processes and the differences in the detector spectral sensitivity bands. It is suggested that the possible presence of this differential behavior should be taken into account when evaluating light pollution indicator datasets for assessing the outcomes of public policy decisions regarding the upgrading of outdoor lighting systems.