Marketing Funnel & Conversion Performance Analysis

Project Overview

This project was completed for Future Interns Data Science & Analytics – Task 3: Marketing Funnel & Conversion Performance Analysis.

The objective was to analyze campaign and customer response data, identify conversion drop-offs, compare channel performance, and recommend practical ways to improve lead-to-customer conversion.

This submission uses the Bank Marketing Campaign Dataset from the UCI Machine Learning Repository, a real-world dataset based on direct marketing campaigns conducted by a Portuguese banking institution.

Dataset Source

• Dataset: Bank Marketing Campaign Dataset
• Source: UCI Machine Learning Repository
• Citation: Moro, S., Rita, P., & Cortez, P. (2014). Bank Marketing [Dataset]. UCI Machine Learning Repository. https://doi.org/10.24432/C5K306
• File used in this analysis: bank-full.csv from bank+marketing.zip
• Records analyzed: 45,211

Business Problem

The business wants to understand:

• Where prospects drop off during the campaign funnel
• Which contact channels bring the best conversion efficiency
• Which customer segments convert best
• Which stages need optimization to improve conversion performance

Because the source dataset starts at the outbound contact stage rather than website traffic stage, the funnel in this project is defined as:

Campaign contacts → Engaged calls (≥120 sec) → High-intent calls (≥300 sec) → Conversions

Tools Used

• Python
• Pandas
• NumPy
• Matplotlib
• ReportLab
• Microsoft Word-compatible report generation
• Business KPI and conversion analysis

Key KPIs

KPI	Result
Total contacts	45,211
Conversions	5,289
Overall conversion rate	11.7%
Engaged leads (≥120 sec)	31,303
Engagement rate	69.2%
High-intent leads (≥300 sec)	12,329
High-intent share	27.3%
Engaged → High-intent conversion	39.4%
High-intent → Conversion	42.9%
Best scaled channel	Cellular
Best scaled channel conversion rate	14.9%

Dashboard

• Dashboard PNG
• Dashboard PDF

Main Insights

1. The biggest opportunity is improving conversion efficiency

Only 11.7% of campaign contacts converted. This indicates that growth will come more from better funnel design and targeting than from simply increasing campaign volume.

2. Cellular is the strongest scalable channel

The Cellular channel delivered 29,285 contacts, 4,369 conversions, and a 14.9% conversion rate, making it the best channel at scale.

3. Previous campaign success is the clearest quality signal

Prospects with a previous campaign outcome of success converted at 64.7%, far above the base rate. This segment should be prioritized.

4. Repeated contact attempts reduce efficiency

Prospects reached only once converted at 14.6%, while leads contacted 6+ times converted at only 5.8%.

5. Volume and conversion are not always aligned

May produced the highest contact volume (13,766) but only 6.7% conversion. By contrast, Mar delivered the best conversion rate at 52.0%, although at lower volume.

6. Some segments convert much more efficiently

Students converted at 28.7%, retirees at 22.8%, and customers with previous campaign success or stronger balance bands also showed better efficiency.

Recommendations

1. Shift more effort toward cellular outreach. It delivers the strongest mix of scale and conversion quality.
2. Create dedicated follow-up journeys for previously successful customers. This is the highest-quality segment in the data.
3. Reduce over-contacting. If a lead crosses 4 contacts without engagement, change the script, offer, or cadence instead of repeating the same approach.
4. Fix high-volume low-efficiency months. Review targeting, messaging, and campaign timing in months such as May and July.
5. Prioritize high-converting segments. Tailor campaigns for students, retirees, and higher-balance customers.
6. Use engagement as a live signal. Calls that pass 300 seconds show much stronger purchase intent and should move into priority follow-up workflows.

Repository Structure

FUTURE_DS_03_marketing_funnel_conversion_analysis/
├── analysis/
│   ├── age_band_performance.csv
│   ├── analysis_summary.json
│   ├── balance_band_performance.csv
│   ├── campaign_performance.csv
│   ├── channel_performance.csv
│   ├── funnel_summary.csv
│   ├── housing_loan_performance.csv
│   ├── job_performance.csv
│   ├── month_performance.csv
│   ├── previous_outcome_performance.csv
│   └── segment_conversion_drivers.csv
├── charts/
│   ├── channel_performance.png
│   ├── funnel_dropoff.png
│   ├── key_drivers.png
│   └── monthly_performance.png
├── dashboard/
│   ├── Marketing_Funnel_Conversion_Dashboard.pdf
│   └── Marketing_Funnel_Conversion_Dashboard.png
├── data/
│   ├── raw/
│   │   ├── bank+marketing.zip
│   │   └── bank-full.csv
│   └── processed/
│       └── bank_marketing_funnel_ready.csv
├── docs/
│   ├── analysis_report.md
│   └── submission_summary.md
├── reports/
│   ├── Marketing_Funnel_Conversion_Analysis_Report.docx
│   └── Marketing_Funnel_Conversion_Analysis_Report.pdf
├── src/
│   └── marketing_funnel_analysis.py
├── .gitignore
├── README.md
├── README_Marketing_Funnel_Conversion_Analysis.md
└── requirements.txt

How to Run

pip install -r requirements.txt
python src/marketing_funnel_analysis.py

Author

Umuhire Gatesi Lyse

GitHub: Lyse777