Embedded learning analytics mostly uses data from online learning platforms to analyze learner's activities and preferences. It's like a digital coach that watches how students interact with their lessons, then provides insights to both the student and the teacher.

The potential is huge, as it can make learning personal, adapting to each student's pace and style. Teachers get a real-time view of who's struggling and can step in with help before it's too late. For students, seeing their own data can make them more aware of their study habits, helping them learn how to learn. A great example of that is the recent update of Microsoft Teams "Reading Progress assignment", as it provides the child with the vocabulary misread and it gives on spot practice for the assignment to be marked completed.

However, using this technology comes with big challenges. First, there are serious privacy concerns, about collecting so much student data. It's crucial to make sure this information is safe and used ethically. In addition to that, there's a risk that teachers might start "teaching to the algorithm," focusing on what the data measures instead of what truly matters for learning. Ultimately, while learning analytics offers powerful tools, it must be implemented carefully and thoughtfully to avoid these pitfalls.

https://elearningindustry.com/data-driven-learning-how-analytics-and-gamification-are-personalizing-education