Subhankar Das
Subhankar Das Data Analyst · Siliguri, IN
Open to Data Analyst roles
📍 Siliguri, India · IST

Subhankar Das

Data Analyst | SQL, Power BI, PostgreSQL, Python

I'm a Data Analyst based in Siliguri, India, with a BCA (Honours) in Computer Applications. I build end-to-end analytics solutions — from raw, messy data to dashboards that answer real business questions — across SaaS, e-commerce, and finance domains. My work spans SQL, PostgreSQL, Power BI, DAX, and Python, and every project on this site is built on real or real-style datasets, not toy exercises. After graduating in 2024, I spent 2025–26 deep in hands-on analytics work — the four projects below are the result. Open to full-time Data Analyst roles.

SQL · PostgreSQL · Window functions Excel · Pivot tables · Power BI dashboards

Featured Projects

End-to-end analytics work from real business-style datasets.

Independent Projects
💳 SaaS Analytics
SaaS Revenue & Churn Analysis
SQL · Power BI · Excel · Early 2026

Analyzed subscription billing data for 5,000+ SaaS customers to track MRR, ARR, churn rate, and retention by plan and industry using Power BI and DAX. Identified 23% revenue leakage in at-risk customer segments and uncovered that Basic plan customers churn 3x faster than Pro/Enterprise. Dashboard built with time intelligence measures for month-over-month churn trend tracking

🛒 E-commerce
E‑commerce Profitability Analysis
SQL · Power BI · Excel · Mar-Apr 2026

Built an end-to-end E-commerce analytics solution using PostgreSQL and Power BI. Analyzed 10,000+ transactions across product categories, regions, and time periods to identify revenue and profit margin patterns. Discovered a 15% margin improvement opportunity hidden in high-return product categories. Dashboard includes KPI cards, drill-down slicers, and category-level profitability breakdown.

🏦 Credit Risk
Loan Default Risk Analysis
SQL · Power BI · Excel · Apr-May 2026

Analyzed 8,000+ loan applications using PostgreSQL and Power BI to identify high-risk borrower segments by credit score band, DTI ratio, employment type, and income level. Built a 3-dashboard risk suite covering default risk overview, borrower profiles, and loan characteristics. Key finding: borrowers with DTI above 43% default at 4.5x the rate of low-DTI applicants.

🔁 Cohort Analysis
E-commerce Sales & Customer Cohort Retention Analysis
PostgreSQL · Python · Aug-Sep 2026

Analyzed 99,000+ orders from a real Brazilian marketplace (Olist) using PostgreSQL — cohort retention, rolling revenue trends, and delivery SLA analysis via CTEs and window functions across a 6-table schema. Key finding: only 3% of customers repeat-purchased, yet they carried 88% higher lifetime value — growth was driven by acquisition, not retention.