Zircon Raman dating is an emergent thermochronological method. It exploits the disruption of the zircon lattice due to α-disintegration of trace amounts of 238U, 235U, 232Th, and their daughters. The radiation damage broadens the Raman bands and shifts them to lower wavenumbers. The Raman bandwidths provide a sensitive measure for the accumulated lattice damage (Nasdala et al., 1995; Nasdala et al., 2001). The measured bandwidth and the effective uranium concentration define the Raman age (Härtel et al., 2021). Radiation damage anneals upon heating and the meaning of the Raman age depends on the zircon’s thermal history. The Raman age is a formation age if no annealing has taken place, or a reset age if all pre-existing damage has been annealed by a geological heating event. In the case of partial annealing, however, the Raman age is a mixed age with no obvious geological significance. Mixed ages are difficult to interpret and cannot be distinguished from reset ages using the standard procedure for annealing detection (Nasdala et al., 2001). On the other hand, inhomogeneous damage distributions due to actinide zoning within zircon grains present a problem for zircon Raman dating. Overlapping signals from more and less damaged zones lead to asymmetric Raman bands and overestimated bandwidths (Nasdala et al., 2005). We discuss Raman spectra of zircon from partially annealed samples and spectra with asymmetric bands. We introduce discrimination plots based on the 356, 439, and 1008 cm-1 bandwidths that provide a means for detecting and distinguishing asymmetry and partial annealing. We discuss examples of zircon Raman dating and present a measurement protocol.