Condition-Based Maintenance (CBM) is widely used to manage the condition of rotating machinery. They require a number of CBM tools to detect acous-tic and vibration signals. One such method is the Fast Fourier Transform (FFT). FFT converts a vibration signal from time domain to its equivalent fre-quency domain representation. Unfortunately, there are dramatic assumptions made related to the proper use of FFT. The paper will provide evidence that this approach may not be the perfect tool for fault detection and diagnosis. The paper celebrates with the limitations of FFT and does not muffle the culpa-bilities of our developed diagnosis culture. The aim is to challenge researchers to come up with something more developed to eventually take use of the processing power we have today.