Ecotoxicogenomics is developing into a key tool for the assessment of environmental impacts and environmental risk assessment for aquatic ecosystems. This review aims to report achievements and drawbacks of this technique and to explore potential conceptual and experimental procedures to improve future investigations. Ecotoxicogenomic literature evidences the ability of genomic technologies to characterize toxicant specific gene transcriptome patterns that can be used to identify major toxicants affecting aquatic species. They also contribute decisively to the development of new molecular biomarkers and, in many cases, to the determination of new possible gene targets. Primary transcriptomic responses obtained after short exposures provided more information of putative gene targets than secondary responses obtained after long, chronic exposures, which in turn are usually more accurate to describe actual environmental impacts in natural populations. Several problems need to be addressed in future investigations: the lack of studies (and genomic information) on key ecological species and taxa, the need to better understand the different transcriptomic responses to high and low doses and, especially, short and long exposures, and the need to improve experimental designs to minimize false transcriptome interpretations of target genes.