Much research has been devoted to improving the process of identifying active factors from designed experiments. Generally, the proposed methods rely on an estimate of the experimental error. Here we present a method based on the TTT (total time on test) plot, where the scaled TTT transform enables an evaluation of the contrasts independently of the experimental error. The method can be separated into two parts. The first part consists of a transformed TTT plot for a visual evaluation of data. The second part is more formal and utilizes the cumulative TTT statistic for testing the significance of contrasts. A simulation study shows the power of the method compared with competing methods. Five data sets are used to show that the conclusions drawn are consistent with those obtained using other suggested methods.