The present sub-permil precision of single zircon chemical abrasion, isotope-dilution, thermal ionisation mass spectrometry (CA-ID-TIMS) UePb dates often reveals age dispersions that are outside of analytical uncertainty. Interpreting these complex age distributions requires the ability to distinguish between protracted crystallization of zircon over a few 100 kyr, age bias due to radiation damage induced Pb-loss, and analytical artefacts. This is a particularly critical issue when a number of these factors occur together. To ensure geologically meaningful results, the complete eradication of Pb-loss is of paramount importance. The impact of Pb-loss can be removed by chemical abrasion (CA) applied prior to the dissolution of zircon. However, CA is an empirical approach that is used without a detailed understanding of how the temperature applied during the annealing step, or the temperature and duration of the partial dissolution step affect the radiation-damaged zones. In addition, the conditions of the CA procedures differ between laboratories making comparisons of age data problematic. This study presents an experimental approach to quantify how chemical abrasion affects the crystal structure and the chemical composition of zircon as well as its UePb age. For this experiment, we have chosen the Plešovice reference zircon, because of its known variation in trace element concentrations and especially the presence of domains rich in actinides. We performed CA experiments under different temperature-time conditions on fragmented Plešovice crystals. These were compared in respect to the changes in trace element concentration, lattice order and UePb date. The most reliable UePb results are obtained by chemically abrading Plešovice fragments at 210 °C for 12 h. Additionally, we demonstrate that the Plešovice zircon cannot be considered homogenous at the current level of precision achieved by CA-ID-TIMS dating due to a natural age variation at the ~900 kyr scale.