In order to be competitive, suppliers of wheat must ensure that their products have the required quality profile demanded by customers, and consistently deliver that quality. In Australia, an integral component of being competitive is the grade classification system. A two-tiered approach, this system involves detailed analysis to determine a ‘genotypic' quality classification, and harvest testing procedures to measure ‘environmental' quality traits. Wheat in Australia is grown across diverse environments therefore different ‘genotypic' classification decisions are made on a regional basis. Most regions though are political divided along state boundaries and are not fully reflective of agro-ecological influences. Despite this the classification regions have worked well in the past, but the effectiveness ‘genotypic' classification decisions is now being placed under pressure with changes in the Australian wheat-breeding sector. Companies now have national targets, instead of the traditional regional or state focus. Furthermore, companies are introducing a greater diversity of germplasm, and at the same time more economical composite sample quality testing regimes are replacing site-by-site assessment. Consequently a national database of quality results has been assembled to determine whether more appropriate boundaries can be identified for the ‘genotypic' classification process. The database contains the core measurements of test weight, hardness, kernel size, protein content, flour yield, farinograph water absorption and dough development time, and extensograph extensibility and strength, originating from testing of advanced stage wheat trials and specialised experiments conducted between 1983 to 2003. The data were initially used to confirm the level of genotype x environment interactions, using balanced data sets for each of the separate, regional components. However, due to the unbalanced nature of the national data, different approaches have been used to identify common environments. These have included the use of glutenin and puroindoline gene family groupings, and newer statistical methods such as residual maximum likelihood (REML).