Despite the importance of customer loyalty, no comprehensive, empirical work has attempted to assess the general findings across academic studies. The study intends to fill that void by conducting a meta-analysis of empirical findings on the predictors of customer loyalty. Although findings of this study support all the hypothesized main effects, they indicate stronger effect size for trust than for other determinants of loyalty. The study also tests the robustness of previous findings across various research and measurement contexts. The analysis of moderating effects reveals several interesting findings. For instance, attitudinal loyalty measures seem to be a plausible surrogate for behavioral loyalty measures. The effects of customer satisfaction and trust on loyalty are less prominent when products are purchased on a regular and relatively short (as opposed to an irregular and relatively long) purchase cycle. Factors that largely relate to product performance (e.g., satisfaction, quality) have a weaker impact on loyalty in B2B than in B2C settings. Some relationships (e.g., the effect of quality on loyalty) become stronger over time. Furthermore, our results detect consistently weaker effects from studies using single-item (relative to multi-item) loyalty measures. © 2011 Elsevier Ltd.