Results-driven Senior Data Analyst with 3+ years of experience in analytics, specializing in revenue forecasting, tender modeling, regulatory data automation, and competitive intelligence. Skilled in programming languages and advanced visualization tools, with a proven record of transforming complex datasets into insights that drive investment strategies, pricing models, and portfolio decisions across global markets.
Revenue Forecasting
· Built a Python-based forecasting model for 40 companies, integrating lifecycle dynamics, exclusivity timelines, pipeline launches, and residuals — achieving 94–95% forecast accuracy versus historical sales and outperforming financial analyst forecasts, strengthening portfolio valuation decisions.
Regulatory Database Automation & Quality Control
· Developed automated pipelines and QC frameworks that added new product entries, classified therapies via label-text algorithms, and validated data across regulatory sources - cutting processing time from 1 month to 1 week, reducing manual review effort by 80%, and improving accuracy and coverage by extracting attributes from 1M+ clinical trial records.
Market & Competitive Analysis
· Led portfolio and competitor analyses across US, EU, and Middle Eastern markets, evaluating 20 products to identify portfolio gaps, therapeutic overlaps, and licensing opportunities— insights that reduced manual research effort by ~40% and directly supported 60+ licensing and M&A evaluations.
Tender Modelling & Pricing Analytics
· Developed a Python-driven tender modelling dashboard with built-in integrity checks, ranking algorithms, and scenario-based revenue simulations, deployed via a multi-page Dash interface with interactive visualizations— transforming Excel modelsinto a scalable, automated decision-support system.
Graph Databases & Advanced Analytics
Developed Neo4j and TypeDB knowledge graphs to unify 40,000+ US & EU productswith complex attributes (strengths, formulations, multi-level indications), replacing flat Excel structures and reducing data redundancy by ~70%, enabling accurate representation of real-world product relationships.