Upside is a digital marketplace that connects millions of consumers with brick and mortar retailers nationwide. Since 2016, Upside's personalized offers have empowered individuals to gain greater purchasing power for their essential needs and helped businesses achieve measurable profits. Billions of dollars in commerce run through the Upside platform every year, and that value goes directly back to our retailers, the consumers they serve, and towards important sustainability initiatives.

• Championed experimentation of the Dealer.com brand by introducing A/B testing for all major product rollouts. This resulted in over 50 tests launched per year.• Collaborated with Product, UX, and Engineering leads in a dual-track agile environment to ensure products are designed with qualitative and quantitative research represented throughout the customer journey.• Introduced new statistical methodologies to the Dealer.com platform, like Causal Impact time series modeling and Entropy Balancing as alternatives to pre/post measurement.• Led the migration of Google Universal Analytics event tracking to GA4 so product analytics measurement could continue uninterrupted.• Designed and launched surveys and screen recording studies to complement the quantitative "what" with the qualitative "why" to ensure products meet consumers' needs.

• Increased bookings and Marriott Rewards enrollments by leading experiment design and test analysis for over 200 A/B and multivariate tests on Marriott.com.• Supervised a team of analysts focused on A/B testing, dashboard creation, and other analysis methods focused on measuring released changed to Marriott.com.• Researched and introduced a statistical method, the Alpha Spending Approach, to understand when A/B tests should be ended early due to strong performance.

• Designed multivariate regression models to audit the sampling work of United States Postal Service statisticians.• Created sampling and estimation methodologies so auditors could validate Department of Transportation grant money was being used as intended.

• Redesigned cut-off based samples for the Annual Survey of Public Employment and Payroll and the Annual Survey of State and Local Government Finance in order to reduce costs associated with data collection.• Implemented decision-based estimation for cut-off sampling that improved the precision of survey statistics.