Variations in many aspects of physical and chemical grade are often of concern to mining operations and mineral processing plants. In most circumstances these amounts of variation are predictable. At the time of planning new operations, we should be able to predict the amount of sampling and testing variation which our sampling devices, laboratory procedures and on-stream analysis devices will achieve. We should know how well our stockpiles, reclaimers and bins will homogenise material (or whether segregation will actually increase the amount of variation). We should be able to predict the amounts of variation that will occur on the important time or tonnage scales at the crucial p oints in our proposed operation. And once these amounts of variation are reliably known, mine planning will seem a little bit more like dealing with known knowns rather than like dealing with unknown unknowns.The first step towards quantifying amounts of variation for planned operations is to assess amounts of variation for existing operations. How many operations have estimates of amounts of variation in their databases? How often are people able to partition amounts of variation into amounts due to sampling error, amounts due to sample preparation, amounts due to assaying, and the amount of real variation in the stream of material?CITATION:Robinson, G K, 2012. Making sure that the amount of grade variation is a known known, in Proceedings Project Evaluation 2012 , pp 159-162 (The Australasian Institute of Mining and Metallurgy: Melbourne).