John Geer

• Helping companies understand what resonates most with their audience by automating ad A/B tests and combining results into clear feedback about ad features
• Using causal inference to explain which ad features drive results and shape model structure

Techniques used: Design of Experiments, A/B Testing, Generative AI, Automated Variable Selection, Explainable Boosting Machine

• Analyzed revenue, LTV, and marketing spend patterns; frequently wrote internal articles and gave presentations to stakeholders
• Built data pipelines combining revenue, marketing, and supply chain data across 10+ platforms and 15 countries
• Forecasted customer lifetime value daily for each customer, enabling marketing improvement

Techniques used: BTYD / RFM Models, BigQuery, Airflow, ETL Pipelines

• Lifted revenue by over $5 million by optimizing regional marketing using experiments, causal inference, and media mix modeling
• Wrote 100+ articles and presented to the CEO and Management Team weekly, clearly explaining insights about products, promotions, and marketing
• Sped up financial reporting from monthly to daily by building a program to combine several cost and revenue datasets
• Made company data accessible to the 100+-person organization by building data pipelines and self-serve reporting systems
• Prioritized projects for the 4-person Data Science team and helped team members grow their skills

Techniques used: Causal Inference, Media Mix Modeling, Design of Experiments, BI Dashboards

• Improved the conversion rate, leading to $20M+ in additional annual revenue, by writing web experiment analysis software using Stan, R, and Bayesian bandit algorithms
• Produced daily sales forecasts and explained revenue changes with Bayesian time series analysis

Techniques used: Bayesian A/B Testing, Time Series Analysis, Hierarchical Models, Survival Analysis, Probabilistic Programming

• Co-wrote a program producing 6,000+ natural language medical clinic reports; 80%+ of clinic managers responded that reports made understanding data easier
• Built systems to optimize written content for conversion rates using contextual Bayesian bandit algorithms

Techniques used: NLP, Time Series Analysis, Contextual Bandit Algorithms, Random Forests

• Focused on Predictive Analytics and Data Mining
• Thesis: Built a predictive model of the number of views a TED talk will receive

• Thesis: "Skepticism Regarding the External World"
• Meaning: "Are we sure we know what's going on?"