Recent developments in international trade theory give growing emphasis to the quality of the exported products, showing that it affects both the direction of trade and the countries’ export performances. However, as quality is unobservable, a measurement problem clearly emerges. In this paper we measure product quality relying on a nested logit demand structure developed by Berry (1994) and then applied to trade data by Khandelwal (2010). Our main goal is to investigate the reliability and the properties of the estimated qualities, focusing on the EU food sector, where the growing attention on quality and safety issues is leading to an increase in the demand for high quality products. Main results give credence to the accuracy of the quality estimates, which display some interesting properties. Indeed, the quality rankings we draw are in line with the expectations, and quality growth proves to be strongly correlated with TFP growth. Moreover, results reveal that the competitive strategy of countries (high-quality vs. low-price) tends to change when moving from OECDs to non-OECDs. Finally, we provide evidence that the quality and price components of export unit values behave differently when testing their relationship with trade costs.