The oocyte grows within a follicle composed of layers of somatic cells. It undergoes with the cumulus cells that form the innermost layer a dialogue that is critical for its maturation. Based on the assumption that the transcriptome of the cumulus cells reflects the physiology of the oocyte, it may prove a useful non-invasive tool in embryo selection to improve assisted reproduction outcomes. During the past decade, various studies have been conducted with the objective of identifying cumulus biomarker genes as prognosis tools for oocyte quality and competence. Remarkably no common biomarkers stand out among all these studies. In this review we perform a critical analysis of the literature in order to reveal some of the parameters that may account for these discrepancies, such as patients' inclusion criteria (maternal age, stimulation protocols), day of embryo transfer (day 3 or 5), outcome criteria (oocyte potential, embryo competence, pregnancy). Moreover there is a lack of standardization in the experimental designs used for RNA extraction and gene expression assessment (microarrays, RT-qPCR) and for the statistical analyses. In conclusion, critical analyses such as the present one are indispensable to pave the way for future searches of predictive biomarkers of pregnancy.