Ultrasonic inspection of wall thickness in pipes is a standard technique applied widely in the petrochemical industry. The potential precision of repeat measurements with permanently installed ultrasonic sensors however significantly surpasses that of handheld sensors as uncertainties associated with coupling fluids and positional offsets are eliminated. With permanently installed sensors the precise evaluation of very small wall loss rates becomes feasible in a matter of hours. The improved accuracy and speed of wall loss rate measurements can be used to evaluate and develop more effective mitigation strategies. This paper presents an overview of factors causing variability in the ultrasonic measurements which are then systematically addressed and an experimental setup with the best achievable stability based on these considerations is presented. In the experimental setup galvanic corrosion is used to induce predictable and very small wall thickness loss. Furthermore, it is shown that the experimental measurements can be used to assess the effect of reduced wall loss that is produced by the injection of corrosion inhibitor. The measurements show an estimated standard deviation of about 20nm, which in turn allows us to evaluate the effect and behaviour of corrosion inhibitors within less than an hour.