Jason Summers

Context

End-to-end analytics project simulating my real stretch-role work at CarMax. Uses mock data to mirror how labor overcharges, overlapping labor, and operational inefficiencies show up in automotive reconditioning workflows.

Problem

Leaders needed a clearer picture of where labor overcharges and misapplied operations were occurring, which technicians and categories were driving cost risk, and whether issues were isolated mistakes or systemic patterns.

Approach

• Cleaned and modeled repair order issue data in Excel, adding calculated fields for billed vs. correct labor, cost overcharge, and severity scoring.
• Loaded data into PostgreSQL and built a reusable view with derived fields for action quality, severity score, and operational risk buckets.
• Used SQL (including window functions) to analyze weekly trends, category rollups, technician-level variance, and high-risk operation groups.
• Performed exploratory data analysis in Python (Pandas) to quantify distributions, outliers, and technician performance indices.
• Built a Power BI dashboard with KPIs, technician and category breakdowns, trend lines, and slicers for severity, week, and operation group.

Outcomes & Insights

• Identified that incorrect labor time and overlapping labor were the primary drivers of total cost overcharge in the dataset.
• Surfaced a small subset of technicians and operation groups that contributed a disproportionate share of high-severity overcharges.
• Demonstrated how a severity-weighted risk score can prioritize coaching and process review instead of treating all issues equally.
• Produced an interactive dashboard that supports “where should we look first?” conversations for leaders and production managers.
• Showcased my full-stack analytics workflow: Excel → SQL → Python → Power BI, tied to a real operational domain.

Context

End-to-end business analytics project simulating the real-world performance of a single-technician mobile mechanic operating in Maricopa County, AZ. Built using mock data to reflect pricing strategy, service mix, customer retention, and the operational impact of mobile travel.

Problem

The business needed clear visibility into which services were most profitable, how repeat customers impacted revenue, whether mobile travel distance reduced profit margins, and how monthly demand patterns affected business stability.

Approach

• Modeled a full 12-month dataset (320 repair orders) including labor hours, labor revenue, parts cost, parts markup, travel distance, travel cost, and total profit.
• Loaded and analyzed the data in PostgreSQL, writing KPI-driven SQL queries to evaluate service profitability, customer lifetime value, monthly trends, and mobile efficiency.
• Used Python (Pandas) for exploratory data analysis to validate distributions, profit margin variability, travel vs profit relationships, and revenue consistency.
• Designed a Tableau executive dashboard featuring KPI cards, monthly revenue & profit trends, top services by profit, customer retention breakdowns, and travel efficiency visualizations.
• Structured the project using a professional multi-tool analytics workflow: CSV → SQL → Python → Tableau.

Outcomes & Insights

• Identified diagnostics and A/C compressor replacement as the highest-margin services in the business.
• Confirmed that repeat customers generated the majority of total revenue and profit.
• Found that long-distance jobs (11–15 miles) produced higher average profit due to larger repair scope, while mid-range travel (6–10 miles) showed weaker margin efficiency.
• Determined that premium vehicle brands did not consistently outperform standard brands in average profitability.
• Produced an executive-ready interactive Tableau dashboard that supports pricing, service focus, and retention strategy decisions.