This presentation examines the quantitative effects of automated writing evaluation (AWE) on EFL learners' writing development using the complexity, accuracy, and fluency (CAF) framework. Our study compared two university-level EFL writing classes: one receiving combined AWE and teacher feedback, and another receiving only teacher feedback. Analysis focused on specific CAF metrics including syntactic complexity (mean length of T-units), lexical diversity (type-token ratio), grammatical accuracy (errors per 100 words), and fluency (total word count). Results revealed that the AWE-supported group demonstrated significantly greater improvements in grammatical accuracy (23% reduction in errors) and lexical diversity (18% increase), with more modest gains in syntactic complexity. These findings align with recent research by Barrot (2021) and Guo et al. (2022) on AWE's differential impact across writing dimensions. This presentation provides practitioners with a practical CAF-based benchmark for evaluating AWE's effectiveness in their own classrooms, including specific metrics and measurement techniques. Attendees will gain insights into which aspects of student writing most benefit from AWE integration and how to implement similar assessment approaches in their teaching contexts.