Learntwin is a web-based learning analytics companion that creates a “digital twin” profile for each learner by aggregating data such as course activity, quiz scores, deadlines, attendance or self-reported feelings. The prototype consists of a backend API that ingests structured data (e.g. CSV exports or simple connectors to LMS tools), a database where learner profiles and indicators are stored, and a frontend interface that shows teachers and students a compact overview of progress, risk level and suggested actions. Short-term milestones include: (1) stable data ingestion (CSV + at least one LMS connector), (2) robust learner profiling with a small set of validated indicators, (3) a classroom-ready teacher view and student view, and (4) pilot tests with a small course or school. Mid-term, the plan is to add more data sources, refine indicators with real usage feedback, and provide exportable reports for schools. Main resources are development time, access to anonymised data samples and close collaboration with teachers. Barriers include data privacy constraints, heterogeneous school systems and limited teacher time. Learntwin enhances digital education by turning scattered data into timely, actionable insights that support feedback, differentiation and student self-regulation. Success can be measured through engagement with the tool (e.g. how often teachers and students consult their “twin”), improvement in course completion or assignment submission rates, and qualitative feedback from educators and learners on usefulness and clarity.