SQL Financial Operations Analysis

A SQLite portfolio project analysing 12,000 synthetic financial transactions across products, markets, acquisition channels, and customer segments.

Business questions

• How are fee revenue and contribution changing over time?
• Which markets contribute the most?
• Where are failed transactions concentrated?
• Which customer groups generate the most contribution?
• Which acquisition channels produce the highest contribution per customer?

Skills demonstrated

• Relational schema design
• SQL joins-ready structure
• Common table expressions
• Subqueries
• Aggregations
• CASE WHEN
• LAG
• NTILE
• Window functions
• Customer segmentation
• Operational and financial KPI analysis
• Python and SQLite integration

Dataset

• 12,000 synthetic transactions
• 2,750 possible customers
• 5 markets
• 5 products
• 5 acquisition channels
• Completed and failed transaction outcomes

Selected results

• Total fee revenue: GBP 13,192
• Total contribution: GBP 8,096
• Overall failed-transaction rate: 3.5%
• Highest-contribution market: UK
• Highest observed market-product failure rate: Spain / Transfers (6.2%)

Visual preview

Repository structure

data/       CSV dataset and ready-to-query SQLite database
sql/        Schema and documented analysis queries
src/        Python runner
outputs/    Query results, charts, and executive PDF

Run the project

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

You can also open data/financial_operations.db with DB Browser for SQLite and run the queries in sql/analysis.sql.

Data protection

All data is synthetic. No real customer, transaction, employer, or financial information is included.

Author

Vansh Agrawal
MSc FinTech and Policy, University of Surrey
LinkedIn: https://www.linkedin.com/in/vanshagrawall