We examined the use of soil quality (SQ) assessment to predict soil productivity and stability as a component of site potential for rangelands. Two minimum sets of data were compared for the SQ assessment within an area of relatively uniform climate. Data set 1 consisted of total soil N, topsoil depth, effective profile depth (EPD), and grade of structure, thus incorporating only soil chemical and physical properties. Data set 2 included exchangeable soil potassium, EPD, soil water retention capacity at wilting point, a soil slake test, and a nutrient cycling index. The interrelationships between soil properties and plant growth characteristics (i.e. total and herbage yield) were investigated and interpreted by statistical analysis and expert knowledge. By performing multiple regressions for each data set against the plant growth characteristics, we identified the contribution of each data set variable to the variability in plant characteristics and, thus, the predictive potential of each variable and data set. Within data set 1, EPD was important and in data set 2 the nutrient cycling index, which is a landscape function index derived from soil surface attributes, played the most important role in predicting potential. Principal component analysis was used to provide weighting factors for each indicator. We then transformed and combined observed indicator values for each data set using weighting factors and scoring functions into an additive soil quality index (SQI) varying in value from 0 to 1. The SQIs, with values greater than 0.8, provide optimum conditions for high yield.