📊 Lead Quality Analysis

🧠 Project Overview

This project analyzes a dataset of approximately 8.8K leads from a single publisher with multiple traffic sources delivered to one advertiser.
Each row represents one lead and its final disposition.
The goal is to understand lead quality trends, key influencing factors, and cost–performance trade-offs to improve advertiser ROI.

---

🎯 Objectives

1. Trend Analysis: Determine if lead quality is improving, declining, or stable over time, and test for statistical significance.
2. Drivers of Lead Quality: Identify major factors affecting lead outcomes — including ad placement, data quality, geography, and creative format.
3. CPL Trade-off Analysis: Evaluate whether a +20% CPL increase (from $30 → $33) can be justified by a +20% uplift in lead quality (from 8.0% → 9.6%).
4. Optimization Goal: Recommend actionable strategies to maximize lead quality and minimize wasted spend.

---

🧮 Key Methods

• Data Cleaning & Preprocessing:

  • Handled missing values for CallStatus, AddressScore, and PhoneScore.
  • Removed non-essential columns like IP Address.

• Feature Categorization:
  Leads grouped into Closed, Good, Bad, and Unknown for consistent evaluation.

• Exploratory Data Analysis (EDA):

  • Trend analysis of lead quality over time.
  • Comparison of call disposition across campaigns and widgets.
  • Distribution of address and phone quality scores.

• Statistical Testing:

  • Conducted hypothesis tests to identify statistically significant differences in lead quality.

---

📊 Visualizations

Below are the main visual insights derived from the analysis:

---

1. Lead Outcome Distribution

Distribution of leads categorized as Good, Bad, or Closed.

---

2. Lead Quality Trends Over Time

Trend showing how lead quality evolved over time.

---

3. Top Widgets by Good Rate

Top-performing widgets ranked by percentage of Good leads.

---

4. CPL vs Quality Trade-Off (Simulation)

Simulation showing trade-off between CPL and overall lead quality improvement.

---

💡 Key Insights & Final Recommendations

Key Insights

1. Quality Trend: Overall lead quality has remained flat, with no statistically significant improvement or decline over time.
2. Top Drivers of Quality: Key influencing factors include widgets, publisher campaigns, data hygiene (address and phone scores), and geography.
3. Performance Segmentation: Certain widgets and campaigns consistently outperform others, delivering higher proportions of Good and Closed leads.
4. Data Quality Correlation: Leads with strong address and phone scores show a clear positive correlation with conversion rates.
5. CPL–Quality Trade-off: A +20% increase in CPL (from $30 → $33) is justifiable when targeting segments achieving ≥10% Good Rate or higher.

---

Actionable Recommendations

• 🎯 Attribution: Minimize “Unknown” lead sources by improving tracking and tightening attribution channels.
• 📈 Campaign Optimization: Pause or restructure underperforming campaigns and scale the top 5 high-quality widgets/publishers.
• ✅ Data Validation: Implement address and phone validation at lead capture to eliminate low-quality entries early.
• 🗺️ Geographic Targeting: Redirect budget toward top-performing regions/states and reduce spend in weak zones.
• 💰 Pricing Strategy: Negotiate higher CPL rates for pre-qualified, validated leads that historically convert better.
• 📊 Continuous Monitoring: Develop monthly dashboards to track Good/Closed rates and detect performance drifts in real time.

---

Overall Conclusion

The analysis indicates that lead quality optimization is achievable without major cost escalation, provided focus remains on
data hygiene, campaign refinement, and performance-based pricing.
Strategic realignment of budget and validation filters can yield a sustainable uplift in ROI and lead quality.

---

🛠️ Technologies Used

• Python
• Pandas, NumPy
• Matplotlib, Seaborn
• Jupyter Notebook

---

⚙️ Setup & Installation

1. Clone this repository:

git clone https://github.com/indu-explores-data/Lead-Quality-Analysis.git

2. Navigate to the project folder

cd Lead-Quality-Analysis

3. Install required dependencies:

pip install -r requirements.txt

4. Open the notebook:

jupyter notebook "Lead Quality Analysis.ipynb"

---

▶️ Usage

• Run the notebook cells sequentially to:
• Clean and preprocess data
• Perform exploratory analysis
• Visualize patterns and trends
• Interpret insights and business recommendations

You can also adapt the notebook for your own lead datasets by updating the input file path and rerunning the workflow.

---

🔗 Connect with Me

Let’s connect on LinkedIn for project discussions or data-driven collaborations:

---

🙌 Feedback & Support

If you found this project helpful, please ⭐ star the repository and share your thoughts. Suggestions and contributions are always welcome!