A full Business Intelligence case study combining Excel preprocessing, data modeling, KPIs, and interactive dashboards.
This project presents an analysis of global car sales data using a complete analytics workflow:
Excel โ Data Cleaning & Preparation Power BI โ Data Modeling, DAX, KPIs, Dashboards & Storytelling
The goal is to understand:
Sales distribution
Engine specifications & performance
Pricing behavior
Resale value retention
Regional market preferences
Manufacturer competitiveness
The project includes multi-page interactive dashboards for Asia, Europe, and the USA with a professional Dark Mode design.
Car dealerships, manufacturers, and distributors often face challenges such as:
Identifying top-performing models
Understanding price vs performance relationships
Discovering regional sales patterns
Evaluating resale value retention
Comparing engine efficiency vs demand
This analysis transforms raw car sales data into valuable insights that support pricing decisions, inventory planning, and marketing focus.
Car-Sales-Analysis/
โ
โโโ data/
โ โโโ Car Sales Analysis.xlsx
โ
โโโ dashboards/
โ โโโ Car Sales Dashboard.pbix
โ
โโโ images/
โ โโโ Dashboard Main.png
โ โโโ Dashboard EU.png
โ โโโ Dashboard Asia.png
โ โโโ Dashboard USA.png
โ โโโ line-chart.png
โ โโโ bar-chart.png
โ โโโ scatter-plot.png
โ
โโโ README.md
The dataset includes key automotive metrics:
Manufacturer โ Car brand
Model โ Model name
Unit Sales โ Number of units sold
Price โ Selling price
Resale Value โ Post-launch retained value
Retention % โ Resale Value รท Price
Engine Metrics โ Engine Size, Horsepower, Fuel Efficiency
Vehicle Type โ Passenger, Car, Truck, etc.
These fields help analyze performance, pricing, demand, and long-term value.
All preprocessing was performed inside Excel using:
โ Removal of duplicates
โ Fixing inconsistent manufacturer/model names
โ Standardizing numeric fields (Price, Engine Size, HP)
โ Converting date formats
โ Creating calculated fields:
-Retention % Category (High / Medium / Low)
-Total Units Sold
-Price Bands
โ Pivot tables for initial summaries
โ Filtering invalid or missing entries
This step simulates real-world workflows where analysts clean data before BI modeling.
The cleaned Excel file was imported into Power BI and the following steps were performed:
Creating relationships (if multiple tables)
Writing DAX measures:
-Total Revenue
-Average Price
-Retention %
-Units Sold
-Engine-Performance Score
-Creating summary tables
-Characterizing models by price tiers
-Adding region-based slicers
-Building multi-page narrative dashboards
๐น 1. Main Dashboard (Global Overview)
KPIs: Unit Sales, Revenue ($ Billions), Avg Price
Correlation Heatmap: Price โ Horsepower โ Engine Size
Retention Donut Chart: Distribution of resale value strength
Top Manufacturers: Global sales performance ranking
Model-Level Insights: Price bands, average HP
๐น 2. Regional Dashboards (Asia โ EU โ USA)
Each region includes:
Treemaps: Top 10 models by sales
Scatter Plots: Price vs Resale Value (bubble size = units sold)
Performance Trends: Area/ribbon charts
Engine Preference Analysis: HP & Engine Size distribution
๐น 3. Excel Exploratory Charts (Initial Analysis)
These helped validate patterns before developing the BI model.
๐ 1. Strong correlation between car performance and pricing
Price increases consistently with engine size & horsepower (correlation = 0.85).
๐ 2. Asian manufacturers lead in value retention
Brands such as Toyota & Honda show high resale percentages compared to competitors.
๐ 3. Regional preferences vary significantly
-USA: High demand for high-horsepower trucks & SUVs
-Europe: Preference for compact luxury & fuel-efficient cars
-Asia: Balanced market with strong mid-range pricing
๐ 4. Some models generate high revenue despite lower unit sales
Due to premium pricing โ valuable insight for pricing strategy.
โ Increase marketing and inventory for top-retention Asian models
โ Adjust pricing strategy for power-heavy vehicles in USA market
โ Expand hybrid/efficient models for EU market
โ Promote high-HP models in regions showing preference
โ Use retention analysis to optimize long-term pricing strategy
These insights help manufacturers refine strategy and dealerships optimize stocking.
Microsoft Power BI
DAX
Data Modeling
Interactive Visualizations
Multi-page Dashboards
Microsoft Excel
Cleaning
Pivot Tables
Initial EDA
Add predictive modeling (Machine Learning) to forecast prices
Use Python (pandas, seaborn) to perform deeper statistical analysis
Automate report refresh via Power BI Servic
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
ู ุดุฑูุน ุชุญูููู ุดุงู ู ูุจูุงูุงุช ู ุจูุนุงุช ุงูุณูุงุฑุงุชุ ููุฏู ุฅูู ููู ุณููู ุงูุฃุณุนุงุฑุ ุฃุฏุงุก ุงูู ูุฏููุงุชุ ูู ูุงุตูุงุช ุงูู ุญุฑูุงุชุ ู ุน ุงูุชุฑููุฒ ุงูุขู ุนูู ููุญุงุช ุงูู ุนููู ุงุช ุงูุชูุงุนููุฉ (Dashboards) ูุชูุฏูู ุฑุคู ุฃุนู ู.
ูุนุชู ุฏ ุงูู ุดุฑูุน ุนูู ุชุญููู ุจูุงูุงุช ุงูุณูุงุฑุงุช ูุงุณุชุฎุฑุงุฌ ุงูุฃูู ุงุท ูุงูุงุชุฌุงูุงุช. ุชู ุชุทููุฑ ุงูุนู ู ูููุชูู ู ู ุงูุชุญููู ุงูุชูููุฏู ุจุงุณุชุฎุฏุงู Excel ุฅูู ุจูุงุก ูุตุต ุจูุงููุฉ ุชูุงุนููุฉ ุจุงุณุชุฎุฏุงู Power BI.
ุชู ุฅุถุงูุฉ ู ุฌูุฏ ุฎุงุต ุจู ููุงุช Power BI ูุตูุฑ ุงูููุญุงุช ุงูุฌุฏูุฏุฉ ูู ุง ูู ู ูุถุญ ูู ุงููุณู ุงูุฅูุฌููุฒู ุฃุนูุงู.
ุชุดู ู ุงูุจูุงูุงุช: ุงูุดุฑูุฉ ุงูู ุตูุนุฉุ ุงูู ูุฏููุ ุงูู ุจูุนุงุชุ ุงูุณุนุฑุ ููู ุฉ ุฅุนุงุฏุฉ ุงูุจูุนุ ูุณุจุฉ ุงูุงุญุชูุงุธ ุจุงูููู ุฉุ ูู ูุงุตูุงุช ุงูู ุญุฑู (ุงูุญุฌู ุ ุงูููุฉุ ุงูููุงุกุฉ).
ุชู ุชุตู ูู ุชูุฑูุฑ ู ุชุนุฏุฏ ุงูุตูุญุงุช ุจูุธุงู ุงููุถุน ุงูุฏุงูู (Dark Mode):
- ู ุคุดุฑุงุช ุงูุฃุฏุงุก (KPIs): ุฅุฌู ุงูู ุงูู ุจูุนุงุช (ุจุงูู ููุงุฑ ุฏููุงุฑ) ูุนุฏุฏ ุงููุญุฏุงุช.
- ุฎุฑูุทุฉ ุงูุงุฑุชุจุงุท (Correlation Heatmap): ุชูุถุญ ุงูุนูุงูุฉ ุงููููุฉ ุจูู ุงูุณุนุฑุ ุงูููุฉ ุงูุญุตุงููุฉุ ูุณุนุฉ ุงูู ุญุฑู.
- ุชุญููู ุงูููู ุฉ: ุชูุฒูุน ูุณุจุฉ ุงูุงุญุชูุงุธ ุจุงูููู ุฉ (Retention Value).
- ุงูุฎุฑุงุฆุท ุงูุดุฌุฑูุฉ (Treemaps): ุชุนุฑุถ ุฃูุถู 10 ู ูุฏููุงุช ู ุจูุนุงู ูู ูู ู ูุทูุฉ.
- ู ุฎุทุทุงุช ุงูุชุดุชุช (Scatter Plots): ุนูุงูุฉ ุงูุณุนุฑ ุจููู ุฉ ุฅุนุงุฏุฉ ุงูุจูุน (ุญุฌู ุงูุฏุงุฆุฑุฉ ูู ุซู ุญุฌู ุงูู ุจูุนุงุช).
- ุชูุฒูุน ุงูู ุญุฑูุงุช: ุชุญููู ุงูู ุจูุนุงุช ุจูุงุกู ุนูู ุงูููุฉ ุงูุญุตุงููุฉ ูุญุฌู ุงูู ุญุฑู.
ุชู ุงุณุชุฎุฏุงู ุงูู ุฎุทุทุงุช ุงูุฎุทูุฉุ ุงูุนู ูุฏูุฉุ ูSunburst ูุงุณุชูุดุงู ุงูุจูุงูุงุช ูุจู ููููุง ุฅูู Power BI.
- ุงูุงุฑุชุจุงุท: ูุฌูุฏ ุนูุงูุฉ ูููุฉ ุฌุฏุงู (0.85) ุจูู ุณุนุฑ ุงูุณูุงุฑุฉ ูููุชูุง ุงูุญุตุงููุฉ.
- ุงูููู ุฉ: ุงูุณูุงุฑุงุช ุงูุขุณูููุฉ (ู ุซู ุชูููุชุง ููููุฏุง) ุชุญุงูุธ ุนูู ููู ุฉ ุฅุนุงุฏุฉ ุจูุน ุฃุนูู ู ูุงุฑูุฉ ุจุงูู ูุงูุณูู.
- ุงูุชูุถููุงุช ุงูุฅูููู ูุฉ: ุงูุณูู ุงูุฃู ุฑููู ูู ูู ููุณูุงุฑุงุช ุฐุงุช ุงูููุฉ ุงูุญุตุงููุฉ ุงูุนุงููุฉุ ุจููู ุง ุงูุณูู ุงูุฃูุฑูุจู ูู ูู ููููุงุกุฉ ูุงููุฎุงู ุฉ ุงูู ุฏู ุฌุฉ.
- Microsoft Power BI (ุชุตู ูู ุงูููุญุงุชุ ู ุนุงุฏูุงุช DAX).
- Microsoft Excel (ุงูุฌุฏุงูู ุงูู ุญูุฑูุฉ Pivot Tables).
- ุจูุงุก ูู ูุฐุฌ ุชุนูู ุขูุฉ (Machine Learning) ูุชููุน ุฃุณุนุงุฑ ุงูุณูุงุฑุงุช.
- ุงุณุชุฎุฏุงู Python ูุฅุฌุฑุงุก ุชุญูููุงุช ุฅุญุตุงุฆูุฉ ู ุชูุฏู ุฉ.







