📊 Property Tax & Assessment Data Quality Dashboard

Identifying Data Quality Risks to Improve Reporting Accuracy and Decision-Making

---

📊 Dashboard Preview

---

🚀 Overview

This project analyzes property tax and assessment data to identify data quality issues that can impact reporting accuracy, financial calculations, and decision-making.

Using a structured validation framework, I identified that 22.2% of records contained data quality issues, including tax mismatches, duplicate entries, and missing values.

💡 Key Insight: Data quality is not just a technical issue — it is a business risk that directly affects decision reliability.

---

🎯 Business Problem

Organizations rely on property tax data for:

• Financial reporting
• Revenue planning
• Policy and operational decisions

However, poor data quality can lead to:

• Incorrect tax calculations
• Duplicate or inconsistent records
• Misinformed business decisions

---

🔍 Objective

To design a data quality validation framework that:

• Detects inconsistencies and errors
• Quantifies data quality issues
• Improves overall data reliability

---

🧱 Dataset

The dataset contains structured property-level data, including:

• Roll Number (Unique Identifier)
• Municipality
• Recorded Tax Amount
• Calculated Tax Value
• Property Classification

---

🧰 Analytics Stack

• Excel (Core Analysis Environment)
• Power Query (ETL / Data Transformation Layer)
• Pivot Tables (Metric Aggregation Engine)
• Excel Dashboards (Visualization & Reporting Layer)
• GitHub (Documentation & Portfolio Hosting)

---

🔍 Data Quality Framework

The following validation checks were implemented:

1. Duplicate Detection

• Identified duplicate roll numbers
• Prevents overcounting and data redundancy

2. Tax Validation

• Compared recorded vs calculated tax values
• Detected inconsistencies in tax calculations

3. Missing Value Checks

• Identified incomplete records
• Ensured dataset completeness

4. Municipality Validation

• Checked for invalid or inconsistent location entries

5. Data Standardization

• Ensured consistent formats across fields

---

🧪 Validation Logic (Python)

import pandas as pd

df = pd.read_csv("data/processed/cleaned_data.csv")

# Duplicate detection
duplicates = df[df.duplicated(subset=["roll_number"])]

# Missing values
missing = df.isnull().sum()

# Tax mismatch
mismatch = df[df["calculated_tax"] != df["recorded_tax"]]

print("Duplicates:", len(duplicates))
print("Missing Values:\n", missing)
print("Tax Mismatches:", len(mismatch))

---

📈 Key Findings

• 22.2% of records contained data quality issues
• 9.4% tax mismatches identified
• Duplicate entries detected across records
• Missing values present in key fields

---

💼 Business Value

This system demonstrates how raw administrative data can be transformed into a governed analytical asset that supports:

• ✅ Improved revenue forecasting accuracy
• ✅ Better policy decision-making
• ✅ Early detection of data degradation
• ✅ Increased trust in reporting systems
• ✅ Reduced manual data validation effort

📌 Key Insight: Data quality issues are not reporting errors — they are decision risks.

---

📊 Dashboard Highlights

The dashboard provides:

• Data quality score overview
• Breakdown of issue types
• Tax mismatch analysis
• Duplicate record detection
• Data completeness tracking

---

⚠️ Limitations

• Dataset size may limit generalization
• Validation assumes correct reference logic
• No external benchmark dataset used

---

🔮 Future Enhancements

• 🔄 Automate ETL pipeline (scheduled ingestion)
• 🤖 Introduce anomaly detection using ML models
• 📡 Enable near real-time data quality monitoring
• 🧾 Expand to multi-region property tax benchmarking
• 📊 Add SLA-based data quality thresholds & alerts

---

📚 Key Learnings

• Designing data quality frameworks from raw administrative data
• Translating technical data issues into business KPIs
• Building monitoring systems instead of static dashboards
• Structuring analytics projects like real data products
• Communicating insights for non-technical stakeholders

---

📂 Repository Structure

data/ scripts/ dashboards/ outputs/ docs/ assets/

---

📄 Data Dictionary

Column	Description
roll_number	Unique property identifier
municipality	Property location
recorded_tax	Reported tax value
calculated_tax	Expected tax value
property_type	Classification of property

---

👤 Author

Abodunrin (Richard) Oketade

Data Analyst | Business Intelligence | Data Quality & Analytics

“Turning data into business decisions.”

🔗 GitHub: https://github.com/Richie-Rokka
🔗 LinkedIn: www.linkedin.com/in/abodunrin-oketade

---

⭐ If you found this project insightful, feel free to star the repository!