Objective: Sudden gains are robust predictors of outcome in psychotherapy. However, previous attempts at predicting sudden gains have yielded inconclusive findings. The aim of the present study was to examine a novel, transdiagnostic, transtherapeutic predictor of sudden gains that would replicate in different settings and populations. Specifically, we examined intraindividual variability in symptoms. Method: We examined data from a randomized controlled trial (RCT) of prolonged exposure therapy for posttraumatic stress disorder (PTSD) in children and adolescents (n = 63), an RCT of cognitive and behavioral therapies for obsessive–compulsive disorder (OCD) in adults (n = 91), and psychodynamic therapy delivered under routine clinical conditions in a naturalistic setting for diverse disorders (n = 106). In all 3 data sets, we examined whether a measure of variability in symptoms occurring during the first sessions could predict sudden gains. Results: Variability in symptoms was found to be independent of total change during treatment. Variability in symptoms significantly predicted sudden gains in all 3 data sets and correctly classified 81.0%, 69.2%, and 76.9% of individuals to sudden gain or nonsudden gain status, respectively. Conclusions: The present study represents the first examination of variability in symptoms as a predictor of sudden gains. Findings indicated that sudden gains are significantly predicted by intraindividual variability in symptoms, in diverse settings, contexts, and populations. Advantages of this predictor, as well as clinical and research implications are discussed.