Sandisiwe M.
Data Analyst
Notable Highlight
Developed Python and R-based models to predict account payment probability based on historical traits, enabling legal and collections teams to escalate non-paying accounts and reduce unnecessary operational outreach.
Experience Summary
Sandisiwe is a highly analytical Data Analyst and aspiring Data Scientist with 5+ years of experience working across collections, revenue enhancement, CRM analytics, and financial performance optimization within consulting environments servicing municipal clients. She holds a Master’s degree in Applied Statistics, advanced certifications in Data Science, and is currently pursuing a PhD in Data Science. Her expertise lies in transforming large-scale operational and collections data into actionable business insights that improve revenue recovery, reduce operational costs, and enhance decision-making. She has hands-on experience with Power BI, Excel, Python, R, SQL, and CRM systems, using predictive modeling, segmentation, and automation techniques to optimize collections strategies and customer engagement workflows. Sandisiwe has worked extensively with collections and revenue teams, analyzing payment behaviors, identifying high-probability recovery segments, and building predictive models to determine which accounts are most likely to pay versus escalate to legal action. She also partnered closely with stakeholders across operations and leadership teams to simplify technical insights into commercially actionable recommendations. In addition to her analytics background, she has experience presenting executive-level insights, conducting handover and portfolio analyses for new municipal books, and driving process improvements during major CRM transitions.
Key Achievements
- During a major CRM migration project, Sandisiwe identified a critical flaw in the new system’s “Promise to Pay” (PTP) logic, where customer grace periods were not being correctly applied. She independently investigated the issue, presented findings to stakeholders and external CRM vendors, and successfully demonstrated that the system configuration was inaccurately classifying valid payments as broken promises, helping protect collections reporting accuracy and operational decision-making. - Conducted a detailed operational cost analysis that uncovered excessive SMS communication spend caused by duplicate messaging across multiple customer contact numbers. Her findings led to the implementation of an automated verification process that reduced unnecessary outbound messaging, improved customer communication efficiency, and generated significant cost savings for the business. - Built predictive collection models using Python and R to help segment municipal debt portfolios based on payment likelihood, enabling the business to prioritize high-recovery accounts while reducing operational effort and unnecessary engagement costs on low-probability accounts. - Played a key role in transitioning reporting workflows from legacy CRM systems into self-service SQL-based data extraction pipelines integrated with Power BI dashboards, improving reporting accessibility and insight generation for operational stakeholders.
Skills
Data Analysis & Business Intelligence, Power BI Dashboard Development, Advanced Excel & Reporting, Python & R for Data Science, SQL & Data Extraction, Predictive Modeling & Segmentation, Revenue Enhancement Analytics, Collections & CRM Analytics, Stakeholder Reporting & Insights, Financial & Operational Analysis, Customer Behavior Analysis, Data Visualization & Storytelling, Process Optimization & Cost Reduction, Statistical Analysis & Forecasting, Cross-functional Stakeholder Collaboration