In its broadest sense, water productivity (WP) is the net return for a unit of water used. Improvement of water productivity aims at producing more food, income, better livelihoods and ecosystem services with less water. There is considerable scope for improving water productivity of crop, livestock and fisheries at field through to basin scale. Practices used to achieve this include water harvesting, supplemental irrigation, deficit irrigation, precision irrigation techniques and soil–water conservation practices. Practices not directly related to water management impact water productivity because of interactive effects such as those derived from improvements in soil fertility, pest and disease control, crop selection or access to better markets. However, there are several reasons to be cautious about the scope and ease of achieving water productivity gains. Crop water productivity is already quite high in highly productive regions, and gains in yield (per unit of land area) do not necessarily translate into gains in water productivity. Reuse of water that takes place within an irrigated area or a basin can compensate for the perceived losses at the field-scale in terms of water quantity, though the water quality is likely to be affected. While crop breeding has played an important role in increasing water productivity in the past, especially by improving the harvest index, such large gains are not easily foreseen in the future. More importantly, enabling conditions for farmers and water managers are not in place to enhance water productivity. Improving water productivity will thus require an understanding of the biophysical as well as the socioeconomic environments crossing scales between field, farm and basin. Priority areas where substantive increases in water productivity are possible include: (i) areas where poverty is high and water productivity is low, (ii) areas of physical water scarcity where competition for water is high, (iii) areas with little water resources development where high returns from a little extra water use can make a big difference, and (iv) areas of water-driven ecosystem degradation, such as falling groundwater tables, and river desiccation. However, achieving these gains will be challenging at least, and will require strategies that consider complex biophysical and socioeconomic factors.