⚫Built and maintained real-time dashboards using Excel and SQL, analyzing over 2500 donations and 5000+ project interactions to drive insights on retention trends.⚫Extracted and processed live blockchain data from PotLock smart contracts, delivering weekly data reports that helped boost donor retention by an estimated 20%.⚫Collaborated with the data team in biweekly stand-ups and onboarded 2 new analysts, enhancing team productivity and knowledge-sharing.

⚫Utilized ETL processes to explore 25 million records of user activities on Procore Application platform using Snowflake’s SQL worksheet.⚫Introduced and measured "Active Rate" through A/B testing, evaluating user interactions with Procore marketplace applications and custom applications.⚫Identified 18 out of a total of 66 categories where the “Active Rates” of custom applications were over 30% higher than marketplace apps, driving competitor analysis with Google Analytics.⚫Developed 6 Tableau dashboards to visualize the differences between 500 marketplace and custom applications within the 18 identified categories. Highlighted key differences through HTTP method analysis.⚫Presented insights of competitor analysis and delivered actionable recommendations to optimize customer retention in the biweekly meetings, enabling product managers and engineers to prioritize high-impact feature designs on applications.

⚫Conducted ETL operations to extract 8 GB of unique search engine results page (SERP) data from 3 online marketplaces and applied A/B Testing and SQL to estimate the impact of advertising on product sales over 3 years.⚫Utilized difference in difference (DID) methodology in Python to estimate the causal impact of 4 advertising methods, including pay-per-click, revealing a 32% boost in total purchases and a 15% cannibalization of organic sales.⚫Created data visualization dashboards in Tableau and Domo to highlight advertising campaign insights, driving strategic ecommerce decisions.

⚫Applied Linear Regression and SARIMA models in Python to identify trends and patterns of company revenue using 12 million records of historical financial data, achieving a forecast accuracy of 84% for future revenue performances.⚫Utilized Monte Carlo simulations to improve revenue prediction accuracy and evaluated model performance by back-testing, reducing forecasting volatility by 31% with the enhanced model.⚫Developed 4 interactive dashboards in Power BI to present insights from financial data analysis, equipping stakeholders with intuitive tools for strategic decision-making and performance tracking.⚫Analyzed over 35 million daily stock price records, ensuring data integrity through SQL queries, utilizing ETL processes in Google Analytics to extract, transform, and load the data efficiently.⚫Designed and optimized a volatility measurement model using low-frequency data and rolling windows, improving VaR (Value-at-Risk) accuracy by 27% compared to baseline models.⚫Collaborated with cross-functional teams including finance and business development to build 4 visualization dashboards of key model metrics by Excel and Power BI, presenting analytical insights to stakeholders during weekly meetings.