BRIDGING THE LEARNING CURVE IN BDD: AN ADAPTIVE TRAINING MODEL FOR CROSS-FUNCTIONAL AGILE TEAMS

Abstract

Behavior-Driven Development (BDD) is widely recognized for improving communication between technical and non-technical stakeholders and enhancing requirement clarity in agile software development. However, its adoption faces significant challenges including steep learning curves, lack of Gherkin knowledge, confusion between BDD and Test-Driven Development (TDD), and poorly structured scenarios. Despite extensive research on BDD’s benefits, no existing study proposes a structured, role-based, adaptive training model to address these adoption barriers. This paper presents the Adaptive BDD Training Model (ABTM), a novel framework that personalizes learning paths based on learner role, performance, and identified weaknesses. We conducted an empirical evaluation with 48 participants (32 experimental, 16 control) across three roles: developers, testers, and product owners. Results demonstrate that ABTM significantly improves scenario quality (Scenario Quality Score improvement: +1.55 vs. +0.47, Cohen’s d = 2.38 vs. 0.71), BDD knowledge (+31.25% vs. +17.50%), and team collaboration (Cohen’s d = 2.13 vs. 0.79) compared to traditional training. The model reduced time to competency by 24% while achieving 96.9% completion rate and 89.6% automation readiness. These findings provide strong empirical evidence for ABTM’s effectiveness in reducing BDD learning barriers, contributing both a validated training framework and practical guidelines for BDD adoption in software organizations.