Under the background of innovation 3.0 and open innovation, integrating and utilizing the resources of innovation ecosystem by embedding in it has become an important way for firms to enhance their competitive advantages. Hence, more and more firms conduct collaborative innovation to improve their performance and competitive advantages by embedding in innovation ecosystems. However, although it is widely believed that innovation ecosystem embeddedness is conducive to improving firms′ performance, previous studies on the relationship between them have not reached a consistent conclusion. In general, there are three different views on this relationship, including linear positive correlation, linear negative correlation and nonlinear relationship. In addition, previous studies have not yet explored the impact of innovation ecosystem embeddedness on firms′ performance from a multi-dimensional perspective, and hardly explore the boundary conditions of innovation ecosystem embeddedness to improve firms′ performance from the perspective of research sample characteristics at a macro level. Therefore, in order to achieve a more comprehensive and accurate research conclusion on the relationship between innovation ecosystem embeddedness and firms′ performance, this study performs a meta-analysis using the CMA 3.0 software package to integrate prior empirical studies on such relationship. The meta-analytic procedures are as follows. First, relevant studies are comprehensively searched by keywords, and a series of criteria are established to screen the literature that met the data requirements of this study. After screening, this study finally involves 53 independent empirical studies including 43114 samples covering years from 2010 to 2023. Second, to ensure the reliability of the results, funnel plot and fail-safe numbers are used to test the publication bias, and the significance of Q statistic and I2 value are used to test the heterogeneity among effect values. The results show that, there is no publication bias in this study; and that there exists heterogeneity among the effect values of the relationship between innovation ecosystem embeddedness and firms′ performance, revealing that this study is suitable for meta-analysis with random effects model, and it is necessary to conduct moderating effects analysis. Third, the overall effect meta-analysis and subgroup analysis are used to test the effect of innovation ecosystem embeddedness and its different dimensions on firms′ performance, as well as the moderating effects of industry type and national culture. Fourth, sensitivity analysis and meta-analysis without outliers are used to confirm the robustness of the results. Some important conclusions have been reached. On the one hand, the results of main effects show that innovation ecosystem embeddedness strongly and positively correlates with firms′ performance, and the three dimensions of innovation ecosystem embeddedness, namely strategic orientation, constituent elements, and characteristic attributes, all positively affect firms′ performance. On the other hand, the results of moderating effects reveal that firstly, the moderating effect of industry type on the relationship between innovation ecosystem embeddedness and firms′ performance is not significant; secondly, the positive relationship between innovation ecosystem embeddedness and firms′ performance is stronger for firms in countries with a high degree of power distance than for firms in countries with a low degree of power distance; thirdly, the positive relationship between innovation ecosystem embeddedness and firms′ performance is stronger for firms in countries with a collectivist culture than for firms in countries with a individualistic culture; fourthly, the positive relationship between innovation ecosystem embeddedness and firms′ performance is stronger for firms in countries with a long-term orientation than for firms in countries with a short-term orientation; fifthly, the positive relationship between innovation ecosystem embeddedness and firms′ performance is stronger for firms in countries with a tight culture than for firms in countries with a loose culture. To sum up, this study provides a more comprehensive and accurate research conclusion for in-depth exploration of the relationship between innovation ecosystem embeddedness and firms′ performance by using meta-analysis to comprehensively integrate previous studies on this relationship, which expands the theoretical research of firm-level innovation ecosystem.