To reduce dimensional and mechanical property variations in a component manufactured from a phosphorus alloyed P/M steel, a design of experiments approach was used to identify the most important green compact variations. Responses evaluated were unnotched Charpy impact energy total dimensional change, and sintered density. The experimental array was defined using a fractional factorial design, and the variables investigated were: phosphorus content; median particle size of Fe[3]P-powder; iron powder fines content; added graphite content; green density; and sintering temperature. The design of experiments proved powerful in detecting the most important variables; it pinpointed graphite and phosphorus content, and green density variations as being detrimental.