Standard sound power test methods have existed for numerous years to allow for appropriate noise labeling of products for validation or for monitoring of changes. More recently, advanced methods such as acoustic holography and beamforming have also been successfully used for measurement of sound power and noise source identification. Sound power is a standard requirement for off-highway and agricultural vehicles, construction and power generation equipment, refrigeration and cooling devices, and many other consumer products. In the automotive industry, the engine and a few accessories (AC compressor, power steering pump) are tested for sound power. While sound power testing methods are well known and tests are conducted in most labs by efficient and often automated test procedures, the root-causing strategy in the case of lack of compliance to a specification is still mostly based on trial-and-error. This is likely due to the fact that engineers testing for sound power typically look at global, spatially and time averaged 1/3rd octave spectra of sound pressure and/or sound power, which are too coarse to provide meaningful diagnostic information. This paper describes a simple strategy that has been successfully used by many engineers to focus the troubleshooting and identification of countermeasures on the dominant contributions to the total sound power. Examples of this approach for a few different cases will be reviewed and its application to a large piece of construction equipment will be described with some details. The overall objective here is to contribute to the product development process in terms of noise reduction.