During the last two decades, Kalman filter-based process monitoring techniques have been rediscovered and widely applied in different areas of control and signal validation. When the physical model of the underlying system is known, the Kalman filter is sensitive enough to indicate small, unexpected changes either in the plant or in the measurement models. Although the innovation process that is generated by Kalman filters contains all the necessary statistical information for detecting certain malfunctions, performance degradation, or off-normal operation conditions, skillful hypothesis testing methods are needed for proper interpretation of the innovation’s behavior. The classical binary sequential probability ratio test (SPRT’), developed by Wald, is an optimal tool to judge between two concurring hypotheses. For more than two alternatives, the multiple-hypothesis testing method, the so-called M-ary SPRT, is recommended. In many cases, the situation cannot be represented as simply as a binary problem, however, and the M-ary scheme would be an overcomplication. For an illustration, consider leakage detection when the exact amount of the loss is not of interest. In such a case, the problem can be treated by a properly chosen binary test, and Wald’s classical SPRT framework can be applied. Thus, any binary SPRT and computer code can be used without any modification.