Energy‐dispersive X‐ray fluorescence (XRF) is an attractive analytical method to determine the level of air pollution by heavy metals. The concentration of the filter in ng/cm 2 is obtained by direct comparison of the net characteristic line intensity of an element with that of a thin film standard. As the sampled area on the filter and the area of the standard are larger than the area analysed by the instrument, the distribution of the elements on the surface of both samples and standards have to be sufficiently uniform. If this is not the case, biased concentration estimates are obtained. Two scanning macro‐XRF setups with a beam diameter of ~0.5 mm were used to investigate the distribution of elements in (1) commercially available (Micromatter) standards, (2) in‐house quartz filter standards obtained with an aerosol generator and (3) particulate matter (PM 10 ) collected on quartz filters by a Leckel SEQ 47/50 sampler. The uniformity of the Micromatter standards was better than 2%. At least some in‐house standards showed a concave distribution with less material at the edges. The maximum bias introduced by this is less than 5%. Because of the limited sensitivity of scanning XRF compared with conventional XRF, the distribution of only a few common elements like Ca and Fe could be determined reliably in aerosol filters. The distribution of some heavy elements could only be measured in filters sampled in polluted regions. In general, the loading of particulate matter over the filters was uniform. Copyright © 2017 John Wiley & Sons, Ltd.