Data-Driven Insights and Cash Forecasting for Automotive Repair Operations

This project partnered with a small auto-repair shop to help reduce uncertainty in cash flow and customer behavior—two challenges that make budgeting and operational planning difficult for many small businesses. The primary objective was to build practical, data-driven tools that support more accurate cash forecasting and clearer insight into customer patterns.

Using two and a half years of operational data, five years of financial history, and economic indicators such as GDP and inflation, the project evaluated several forecasting approaches, including SARIMA, XGBoost, and LSTM models. In parallel, the project applied customer segmentation techniques to better understand how different groups contribute to revenue and shop activity.

The analysis uncovered predictable seasonal patterns in both income and expenses and revealed actionable customer segments, such as high-frequency low-value visitors and customers who request quotes but rarely schedule work. These insights provide the shop with clearer expectations around cash flow, improved resource planning, and the ability to target marketing efforts more effectively, demonstrating how applied analytics can directly support small business decision-making.