What needed to be understood.
Customer information, journey activity, and regional mappings were stored separately. The analysis needed a connected model that could explain both overall performance and the sequence of customer interactions.
How the analysis was built.
- Prepared and related the customer, journey, and regional datasets.
- Built customer and marketing KPIs in Power BI.
- Analyzed funnel stages, touchpoint volume, and customer movement.
- Created a Sankey-style view to visualize common paths.
- Compared paid, earned, and owned media behavior and considered the correct time unit for customer lifetime value.
What the evidence showed.
- The project revealed which touchpoints acted as common entry points, connectors, and exit points.
- Channel performance could not be judged only by top-of-funnel volume; downstream movement and conversion context mattered.
- Journey analysis provided a more complete view than isolated campaign metrics.
What should happen next.
- Shift budget toward channels that contribute to qualified downstream movement, not merely traffic.
- Create recovery messaging for high-volume abandonment points.
- Track cohort-level retention and value over a consistent customer-lifetime time horizon.
What this project does—and does not—prove.
The case uses an academic online-eyewear dataset. Recommendations are framed around observed journey structure rather than claimed real-world revenue lift.