I need a data-savvy partner who can look at our archive of customer support interactions, uncover why customers leave, and translate those findings into concrete retention tactics that directly lower our churn rate. You will handle everything from cleaning the text logs to extracting sentiment and issue themes, then quantifying how each theme correlates with cancellations. Once the insights are clear, I expect a concise playbook that ranks the most impactful interventions—think proactive outreach scripts, self-service improvements, or escalation triggers—along with simple KPIs so my team can track results. Python, SQL, and a lightweight dashboarding tool such as Tableau or Power BI are all fine; use whatever you are fastest with as long as the workflow can be reproduced later in a Jupyter notebook or similar. Deliverables: • A well-commented notebook (or script) that ingests the support data and outputs the analytical results. • A brief slide or PDF report summarising key churn drivers, their quantified impact, and the recommended actions. • A live dashboard (share-link or packaged file) that highlights churn-risk signals and can be refreshed with new data. Acceptance criteria: findings must cover at least 90 % of our historic cancellations, recommended actions are tied to measurable metrics, and all code runs plug-and-play on my side. If this first phase goes smoothly we can expand the scope later to include purchase or website usage data, but the immediate focus is the support interactions and a clear plan to keep more customers from walking away.