The study utilises cross-sectional data from utilities of the National Water and Sewerage Corporation of Uganda to test application of extrinsic incentive theory on non-revenue water (NRW) reduction. Applying a multiple regression model, we find that the effect of level of incentive payments on revenue water (billing efficiency) depends on the amounts of staff inputs, service coverage and water supplied per connection. Accordingly, utilities with higher levels of promised incentive payments have a higher likelihood of improving their billing efficiencies, depending on the number of staff employed, level of service coverage and production per connection. The evidence also suggests that other extrinsic factors of incentive theory like coercion and threat of punishment may be useful modulators of incentive payment effects. The study increases understanding about the use of incentive theory in the design of NRW reduction plans and how corresponding incentive policies can be structured.