The Nana positioning protocol is widely used to position participants to minimize technical error when undertaking body composition scanning and analysis with a Dual energy X-Ray absorptiometry (DXA) machine. Once biological and technical errors are accounted for, the only variation in test–retest results is from statistical fluctuation or machine error. Therefore, the aim of this study is to assess the test–retest reliability of the Nana positioning protocol and establish the smallest real difference percentage (SRD%). A gender-balanced group of 30 participants (15 males, 15 females) underwent two scans in succession using the Nana positioning protocol, with repositioning between scans. Percentage change in mean with typical error, Intraclass Correlation Coefficients (ICC), and standard error measurement percentage (SEM%) were used to identify the test–retest reliability and error rate of these protocols. Additionally, SRD% was calculated to assess the point at which clinically important changes occurred in a participant. The reliabilities of the whole body and regional scans were excellent. Percentage change in mean ranged between 0.00% and 0.23%. High reproducibility of the Nana positioning protocol was evident through an ICC ranging between 0.966–1.000. Additionally, typical error, SEM%, and SRD% were all low. Interestingly, fat mass was associated with the largest fluctuations observed to be associated with any of the parameters assessed. When all sources of biological and technical errors have been accounted for, the Nana positioning protocol has excellent test–retest reliability and produces low SEM% and SRD%.