Business Problem: This project explores the factors influencing customer churn and identifies opportunities to improve customer retention.
Dataset & Approach: Using IBM Cognos Analytics, I analyzed customer demographic, service, and account data to identify churn drivers and trends.
Key Findings: This section presents the most important insights and visualizations generated during the analysis.
Recommendations: Based on the findings, I developed recommendations aimed at reducing churn and improving customer retention.
The Finding: The analysis found that customers with month-to-month contracts and shorter tenure were more likely to churn than other customer groups.
Why Data Alone Wasn’t Enough: While IBM Cognos identified these patterns, the data could not determine the best business response. Detecting a trend is different from deciding how an organization should act on it.
The Human Judgement: To address this, I evaluated the business context and considered multiple retention strategies before recommending targeted retention initiatives for high-risk customer segments, including contract conversion incentives and proactive outreach. This experience reinforced the importance of human-in-the-loop decision-making, where analytics tools support—but do not replace—professional judgment.
Skills Demonstrated:
- IBM Cognos Analytics
- Data Visualization
- Marketing Analytics
- Business Analysis
- Data Storytelling
- Strategic Reccomendations
- Human-in-the-Loop Decision Making
William Dvorak | Marketing Analytics & AI Portfolio 