Quantitative PCR (qPCR) offers potential to analyse the microbial composition (component ratios) of mixed-species and mixed-strain fermentations, for example in the preparation of mixed-strain starter cultures for fermented foods or during the maturation and shelf-life of those foods. However, achieving genuinely reliable results requires the identification of suitable DNA sequence targets and the minimization of sources of systematic error. These issues were examined using the Pfaffl approach to relative quantitative analysis and strain mixtures of Lactococcus lactis subsp. cremoris as a model system. Quantitatively distinguishing strains of the same species required (i) establishment of an efficient cell lysis method giving quantitatively equivalent extraction of high-quality DNA from all strains (ii) design of primers (conserved and different between the strains to be analysed) that give high amplification efficiencies, (iii) optimisation of reaction conditions, and (iv) determining DNA target sequences chosen were not prone to variable dosage due to DNA replication. Sequences for strain discrimination were identified via analysis of genomic sequences to identify unique priming sites (conventional species-specific PCR) or restriction enzyme cleavage sites (in a novel restriction-PCR assay). In laboratory trials with quantitatively defined two-strain mixtures, culture component ratios estimated in this way correlated well with cell counts (colony counts adjusted for coccal chain length). Practical application in the field may be limited by the complexities of customising the system for particular strains and species.