Gunal Krish

I’m an analytical data professional focused on commercial research and decision-support work—turning large transaction and customer-behaviour datasets into clear, actionable insights.

In payments and revenue leakage analysis, I analysed 3.5M+ transaction records to identify ₹4.84M in potentially recoverable revenue through systematic failure analysis and operational cross-validation. I also pinpointed payment failure spikes (15.59% vs 6% off-peak) and compared failure rates across UPI and card payments to surface infrastructure bottlenecks.

Across predictive analytics and growth use cases, I built an early-warning student churn prediction workflow from 1,701 activity records, engineering 15+ behavioural features and using structured cross-validation. I achieved 90.3% accuracy and AUC 0.978 with XGBoost, then translated technical findings into retention recommendations for non-technical stakeholders.

I’m equally comfortable with commercial analytics and experimentation-style thinking—like evaluating attribution distortion across channels and recommending reallocation strategies to improve ROAS. I work with SQL, Python, and BI tooling to maintain data verification, documentation discipline, and concise stakeholder communication, especially in remote environments.