Setting practice goals, which helps students regulate their effort toward achieving engagement and mastery, can enhance the benefits of active learning. However, traditional goal-setting approaches, such as homework contingency contracts, often lack frequent feedback and require substantial human intervention. The presented research investigates integrating intelligent and scalable goal-setting support within active learning environments. In recent research, we have evaluated the impact of goal-setting contracts, performance feedback, and scalable goal support with rewards on student effort and learning. The proposed research aims to study student achievement trajectories related to achievement and the distinct impact of adaptive goal feedback. To that end, we propose developing an adaptive goal-setting dashboard that automates feedback and recommendations to support students in setting and refining their goals. Findings from a 12-week study with 110 middle school students in a hybrid tutoring program show that data-supported goal setting led to about a 25% increase in weekly practice time and about a 40% increase in skills mastered per week. These results indicate that intelligent goal setting can enhance engagement and learning while minimizing teacher workload. This research contributes to practical advancements in AI-assisted active learning and theoretical insights into self-regulatory processes related to student effort in active learning.