Back to Projects

Customer Retention Analysis

Predicting Churn and Personalizing Engagement to Strengthen Loyalty

Tools Used: Excel · Power BI · Jira

Project Overview

  • Customer retention is more than just keeping users — it’s about understanding their journey, anticipating their needs, and building lasting relationships. This project focused on analyzing behavioral data and purchase history to uncover what drives loyalty and what signals potential churn. The goal was to build a predictive framework that could identify at-risk customers and enable personalized interventions.
  • I worked with a mid-sized e-commerce platform to analyze several months of customer activity, including transaction logs, support interactions, and feedback ratings. The objective was to segment users by engagement level, model churn risk, and empower marketing and support teams with actionable insights.

    • Data Preparation & Exploration

    Using Excel, I cleaned and structured the dataset, which included:

  • Purchase history and frequency
  • Customer demographics and location
  • Support interaction logs from Jira
  • Feedback ratings and review sentiment

    •  

    I performed exploratory analysis to uncover:

  • Drop-off points in the customer journey
  • Correlations between support activity and retention
  • Patterns in repeat purchases and dormant accounts
  • Sentiment trends across product categories

    • This work helped define the KPIs we would track: retention rate, churn probability, repeat purchase frequency, and support impact.

    Dashboard Development & Visualization

    With Power BI, I built dashboards tailored to different teams:

  • Retention Heatmap: Highlighting churn risk across customer segments
  • Engagement Funnel: Visualizing drop-off stages and reactivation potential
  • Support Impact Tracker: Measuring how support interactions influenced retention

    • Jira integration allowed us to tag and quantify support tickets, revealing patterns in user frustration and satisfaction. Dashboards included dynamic filters, cohort comparisons, and trend lines to support weekly decision-making.

    Key Insights & Strategic Recommendations

    The analysis revealed several high-impact findings:

  • Customers with frequent support interactions had a 40 percent higher retention rate
  • Dormant users responded best to milestone-based rewards and reactivation emails
  • Negative reviews clustered around delivery delays and unclear return policies
  • High-value customers preferred early access to new products over discounts

    • These insights led to a segmented retention strategy based on user behavior and preferences.

    Operational Impact

    The project delivered measurable improvements:

  • Reduced churn by 22 percent within three months
  • Increased repeat purchase rate by 30 percent
  • Improved customer satisfaction scores by 18 percent
  • Enabled proactive outreach through automated retention workflows

    • Collaboration & Workflow

    I worked closely with cross-functional teams:

  • Marketing: To design personalized re-engagement campaigns
  • Support: To prioritize high-risk users and improve ticket resolution
  • Product: To address feedback-driven pain points and improve UX

    • Feedback loops were embedded in the dashboards, allowing teams to monitor the impact of retention strategies and iterate quickly.

    Lessons Learned

  • Support is a loyalty driver: Proactive service builds trust and retention
  • Segmentation unlocks personalization: Not all customers churn for the same reasons
  • Data storytelling matters: Visualizing churn risk helped teams act faster and smarter