This project merges business strategy with analytics to identify underutilized software licenses and proactively flag at-risk clients in the mechanical software industry. Using Python and Power BI, it simulates a real-world scenario account managers face in enterprise software sales.
- Optimize license usage across clients by identifying idle inventory
- Flag churn-prone accounts based on usage, support volume, and engagement gaps
- Provide actionable insights for QBRs and retention efforts
- Python (Pandas, Seaborn, Matplotlib)
- Power BI for dashboards
- Microsoft Excel & mock CSV data modeling
| Dataset | Description |
|---|---|
| CustomerProfile | Client metadata, industries, license counts |
| LicenseUsage | Product logins per user |
| SupportTickets | Support requests and resolution metrics |
| EngagementData | Meetings, follow-ups, QBRs |
- Usage-based tiering (top 20%, bottom 20%)
- Rule-based churn prediction logic
- Engagement score modeling
- Power BI dashboard with custom KPIs, filters, and visual storytelling
- 30% of software licenses remain unused across top accounts
- Risk-prone clients show low touchpoints and high support volume
- Rule-based flags identified 25% of accounts needing retention action
/notebooks/ClientLicenseAnalysis.ipynb/dashboard/License_Risk_Dashboard.pbix/data/*.csvmock input files
Developed by Parth Jani, Account Manager | Data & Digital Transformation Learner
This is a self-initiated portfolio project using fictional data for demonstration purposes.