The present article aims to explore the influence of quality management (QM), including quality commitment and practices, on product and service innovation. Design/methodology/approach – The authors examine the relationship of quality as a determinant of innovation using empirical evidence collected through the European Manufacturing Survey, 2021 edition, to generate nine binary logistic models. Findings – The results reveal differences among the considered innovation outputs. Soft quality practices (employee training) are significant and positive in relation to new-to-firm and new-to-market product innovation and service innovation. In contrast, hard quality aspects (working methods) only affect new-to-firm product innovation and service innovation. The findings also show that quality commitment does not drive innovation. Research limitations/implications – The data used in this paper corresponds to four countries, with Spain and Lithuania being moderate innovators and Slovakia and Latvia being emerging innovators, but all countries are performing below the EU average. Further research expanding the analysis in countries considered strong innovators would bring additional evidence complementing the present findings. Practical implications – This paper has a series of practical implications targeting general managers/CEOs, quality, production, human resource, and innovation managers, with findings showing the innovation impact of practices associated with their function, managed and implemented at the manufacturing site, but with impact and implications that often go beyond their area of responsibility. Originality/value – The main value of the present research consists in fine-graining the relationship of a variety of quality management practices, going from strategy-level aspects (quality commitment as strategic priority) to operational level (hard QM practices) and human facets (soft QM practices), with different innovation outputs, including both service and product innovations. The recent, multi-country, industry-wide dataset is another strength of the analysis.