This paper describes a personal journey in ESR dating and gives some insights of the progress made over 35 years in ESR dating of tooth enamel. When I started, samples were irradiated with four additive dose steps of 5, 10, 15 and 20 krad and the linear extrapolation yielded the dose value. An assumed dose rate of 100 mrad/a yielded the age of the sample. Since then I learned that one actually has to measure the dose rate. I also learned that the dose response can be described by a single saturating function, or perhaps two, and that errors can be appropriately calculated. Rather than having a single anisotropic CO2− radical, tooth enamel consists of two anisotropic CO2− radicals and two isotropic CO2− radicals, one stable, one unstable. Also, the dose response shows spatial differences in the production of these different radicals in response to beta and gamma rays. Then there is the problem of uranium uptake over time. We solved this by combining ESR and U-series analyses. We can even measure appropriate beta attenuation factors including sequential laser ablation U-series analyses. Even better, all analyses can be carried out on small enamel fragments, which can be glued back into their original position with little to no visible damage. Still, when using the most sophisticated dose rate calculations, they're often close to 1000 μGy a−1.