Current data are needed to manage our soil resources but there are few recent purpose-built surveys of soil fertility, a critical asset for natural resources management. Instead, data collected for other purposes are re-used as has been the case in our area of interest here: managing the threat of soil erosion on farm land in East Gippsland. The dynamic nature of soil fertility challenges this approach, so we conducted a new survey using 234 paddocks selected at random. These new data were compared to re-purposed legacy data. Plant tissue samples from a selected subset were used to assess micronutrients. Separately, 27 paddocks that had been sampled during 1975–89 were re-tested to examine temporal changes. We concluded that many paddocks are now strongly or very strongly acidic and deficient in molybdenum (Mo) and boron (B). More specifically, those under dairying, cropping and horticulture are likely to have adequate phosphorus (P) while those grazed for meat and wool are likely to be deficient in P. We concluded that grazing enterprises need to either re-adopt soil testing and fertiliser treatment or adopt management practices suitable for acid soils deficient in P, Mo and B, if their paddocks are to support sustainable productive agriculture. Data from the re-tested paddocks suggest that, in general, many were once less acidic, highlighting the difficulties in using legacy data. Also, contrasts in soil pH and available P between paddocks under different management regimes, highlighted the difficulties in using re-purposed data. We concluded that a statistically designed survey is a preferable basis for natural resource management.