Block Programming Environments (BPEs) are becoming popular tools for introducing novices to programming, due in part to their connection with students' interests in games, apps and stories. This has led to increasing use of BPEs outside of classroom settings, where knowledgeable instructors are not always available. Intelligent Tutoring Systems (ITSs) can keep students on track in the absence of instructors by providing hints and warnings to students in need of help. Further, data-driven techniques can generate this feedback automatically from previous students' attempts at a problem. This research focuses on the integration of this data-driven, ITS-style feedback into a modern BPE and the evaluation of its impact.