The quantification of relative and absolute taxa-specific abundances in complex microbial communities is crucial for understanding and modeling natural and engineered ecosystems. Many errors inherent to this quantification are, though well-known, still insufficiently addressed and can potentially lead to a completely different interpretation of experimental results. This review provides a critical assessment of next generation sequencing (NGS) of amplicons and quantitative real-time PCR for the quantification of relative and absolute taxa-specific genome abundances. Starting from DNA extraction, the following error sources were considered: DNA extraction efficiency, PCR-associated bias, variance of strain-specific 16S rRNA operon copy number per genome, and analysis of quantitative real-time PCR and NGS data. Tools and methods for estimating and minimizing these errors are presented and demonstrated on published data. In conclusion, amplicon sequencing and qPCR of 16S rRNA genes are valuable tools to determine relative and absolute taxa-specific genome abundances, but results can deviate by several orders of magnitudes from the true values if the reviewed error sources are ignored. Many of these errors can be minimized in a cost-efficient manner and large errors can be easily identified by plausibility checks as shown in this review. Finally, the accurate conversion of genome abundances to cell numbers and microbial biomasses was pointed out as an important future research topic for the integration of PCR-based abundances into mathematical models.