Among green technologies addressed to metal pollution, phytoextraction has received increasing attention in recent years as an alternative to physical and chemical methods of decontamination. Since 1998, as part of an Italian multidisciplinary research team on phytoremediation, we have been carrying out several agronomic investigations with field crops in agricultural soil and pyrite waste, both markedly contaminated by heavy metals. Phytoextraction was rarely an efficient process, requiring a long time even to remove merely the bioavailable metal fraction, but the great metal stock in roots suggests exploring the effectiveness of long-term in planta stabilisation. Poor above-ground productivity was the main factor constraining metal removals, especially in wastes. Without assisting the process, only zinc (Zn), manganese (Mn) and copper (Cu) were harvested by the canopy in substantial amounts, with an estimated maximum of ~8 kg of metals from an hectare base with rapeseed in the agricultural soil and only 0.33 kg with fodder radish in pyrite waste. Root growth was a key trait in species and genotype selection, in view of the close relationship between root length and metal uptake. The auxins, humic acids and chelators tested on the model plant fodder radish generally increased metal concentrations in plant tissues, but reduced growth and removals. It is currently concluded that phytoremediation efficiency with crop species may be improved through increased productivity by suitable soil management, involving mineral and organic fertilisation, contaminant dilution, soil capping, and metal immobilisation with inorganics and biochar.