A process to compare results of building daylighting computer simulations to the results obtained using scale-model photometry under real skies is outlined. The process uses the illuminance values in a subject space under study and normalizes the values with those obtained in a space considered to be the base case. The normalized factor is termed Normalized Daylight Performance Index (NDI). An example on how this can be done is applied to an open plan office space with two different furniture layouts. Sources of experimental errors are outlined. NDI was found to be a viable metric for comparing daylighting simulations with scale-model photometry. © 2014 Elsevier B.V. All rights reserved.