KPI Monitor

A simple Python script that tracks business metrics and flags potential issues.

What This Does

Monitors key performance indicators (KPIs) from a CSV file and alerts you when:

• Revenue drops significantly
• Profit margin falls too low
• Active users decline

No dashboards, no web apps. Just clear terminal output showing your metrics and any problems.

How It Works

1. Load data - Reads CSV with date, revenue, costs, profit, and active users
2. Calculate KPIs - Computes profit margin, revenue growth, user growth
3. Check alerts - Compares against configurable thresholds
4. Print report - Shows period-by-period metrics and flags issues

Setup

Install dependencies:

pip install -r requirements.txt

Usage

Run with a file path:

python kpi_monitor.py sample_kpi_data.csv

Or run and enter the path when prompted:

python kpi_monitor.py

Dataset Requirements

Your CSV should have these columns:

• date - Date in YYYY-MM-DD format
• revenue - Total revenue for the period
• costs - Total costs for the period
• profit - Net profit (revenue - costs)
• active_users - Number of active users

Example:

date,revenue,costs,profit,active_users
2024-01-01,100000,70000,30000,5000
2024-02-01,105000,72000,33000,5100
2024-03-01,98000,71000,27000,4950

Generate Sample Data

Test the script with sample data:

python sample_data_generator.py

This creates sample_kpi_data.csv with 12 months of data including some intentional issues to trigger alerts.

Configure Thresholds

Edit the top of kpi_monitor.py to adjust alert thresholds:

REVENUE_DROP_THRESHOLD = 10  # Alert if revenue drops more than 10%
MIN_PROFIT_MARGIN = 15       # Alert if profit margin falls below 15%
USER_DECLINE_THRESHOLD = 5   # Alert if active users drop more than 5%

Example Output

📅 Period: 2024-06-01
--------------------------------------------------------------------------------
Revenue:       $  85,234.50  (-15.2%)
Costs:         $  59,664.15
Profit:        $  25,570.35
Profit Margin:        30.0%
Active Users:       5,234  (+2.1%)

🚨 ALERTS:
   ⚠️  REVENUE DROP: Revenue decreased by 15.2% ($100,276 → $85,235)

Limitations

This is a simple monitoring tool with real constraints:

• No forecasting - Only looks at historical data, doesn't predict future
• No anomaly detection - Uses fixed thresholds, not statistical methods
• No data validation - Assumes clean, complete data
• No visualization - Text output only, no charts or graphs
• No database support - CSV files only
• No notifications - Prints to terminal, doesn't send emails/Slack messages
• Simple period comparison - Only compares consecutive periods
• No seasonality handling - Doesn't account for seasonal patterns

Future Improvements

If I continue this project, I might add:

• Export alerts to a log file
• Email/Slack notifications when alerts trigger
• Support for multiple KPI files
• Visualization with matplotlib (simple charts)
• Moving averages for trend analysis
• Year-over-year comparisons
• Configurable alert rules via config file
• Database support (SQLite, PostgreSQL)
• More KPIs (customer acquisition cost, churn rate, etc.)

Why This Exists

I wanted to:

1. Learn basic business analytics with Python
2. Practice working with time-series data
3. Build something practical for monitoring
4. Keep it simple enough to understand and modify

License

MIT - Use this however you want.