Recent well-publicised problems with the start-up of new mining projects have continued to feed the perception that the mining business in general and Ore Reserves in particular are a poor risk. New mines commonly experience difficulties when the true variability of ore characteristics becomes apparent during commissioning. At this critical stage, the impact of lower metal recovery, reduced throughput or increased consumption of reagents and energy can have a catastrophic effect on economic returns, as illustrated by the premature closure of some operations. There appear to be several common reasons contributing to the poor performance of mining projects, of which three are examined here: lack of early risk assessment before feasibility studies are started, inadequate sampling for metallurgical test work and variability of the run-of-mine (ROM) ore. An underlying feature of these issues is the lack of effective interaction between geologists, mining engineers and metallurgists/ process engineers. Geologists need to be more aware of ore processing issues and the range of characteristics that need to be considered as part of the resource estimate. Metallurgists need a greater understanding of the variability of the run of mine ore in terms of all the characteristics that influence the design and operation of the plant. Mining engineers need to determine the short-term variability of the ore in the mine schedule, so that there is confidence that the quality of the run-of-mine ore will be within the design specifications of the plant. A substantial overlap between these disciplines and sharing of areas of responsibility is necessary if these and other technical problems are to be better understood, production estimates are to be improved and commercial risks are to be reduced. Although the JORC Code (1999) provides guidelines for reporting all the issues that are material to an Ore Reserve estimate, some issues are rarely fully addressed in Ore Reserve reports. In particular there is no formal requirement for a metallurgist to assure his or herself of the appropriateness of the Ore Reserve classification. Consequently the level of uncertainty related to recovered grade and ore processing characteristics may be understated. It is suggested that to meet the concerns of bodies funding new mineral developments a stronger multi-disciplinary approach to the analysis of mineral deposits and reporting of Ore Reserves is required. The culmination of this approach would be the formal requirement for a geologist, mining engineer and a metallurgist to co-sign Ore Reserve reports.