The water-insoluble phosphorus (P) component (‘residue source’) was separated from four commercial triple superphosphate (TSP) products and its agronomic effectiveness evaluated. Two of the TSP products were sourced from the USA and two from the UK. Effectiveness was measured using a pot trial technique with ryegrass as the test crop. The residue sources were evaluated on 13 soils varying in key properties: readily plant-available P, texture, pH and organic matter content. Four of the soils were from Germany, one from Spain, five from the UK and three from France. Grass dry-matter yield and P offtake were measured at every cut on every soil. For both of these variables, significant responses to applied P were obtained on all soils. Mono-calcium phosphate (MCP) was used as the comparative P source and effectiveness of the residue sources was estimated relative to that of MCP. Using dry-matter data, the residue sources were 44–87% as effective as MCP; P offtake data provided corresponding estimates of 35–79%. Both methods of estimation indicated the same ranking of residue sources in terms of relative effectiveness, those derived from USA products being more effective than those derived from UK products. The ranking was the same for all soils. The effectiveness of the residue sources relative to that of MCP appeared little affected by measured soil properties: texture, pH, available P or organic carbon content. The results indicate there is little difference in agronomic effectiveness of TSP products with water solubility greater than 85%.