Automated KPI Report Pipeline – Simulated Sales Dashboard Refresh

Overview

This project simulates an automated KPI reporting pipeline using Python, demonstrating key skills in data automation, data cleaning, KPI calculation, and reporting workflows. It reflects the kind of real-world automation I developed at Smarsh, GoRefime, and during freelance projects.

Tooling Summary (Project-to-Role Alignment)

This project simulates a backend analytics pipeline aligned with common requirements for Analytics Engineer and Data Platform Engineer roles. It includes:

• ETL modularity: Python pipeline simulating generate → clean → calculate → notify structure
• Schema enforcement: Null filtering, data coercion, and validation logic matching real-world CRM sync constraints
• Power BI simulation: Pre-aggregated CSV output for use in dashboard/reporting environments
• Async logic simulation: Time-delayed refresh and error alerting replicates retry-safe orchestration (as used in ADF + Power Automate)
• Job-ready architecture: Matches HSM-eligible data roles in validation-first ETL, reporting infrastructure, and dashboard reliability

All code is fully modular and executable via main.py, with output stored in the output/ directory.

Use Case

The project showcases how to build a lightweight automated reporting solution without needing heavy backend infrastructure. It combines data cleaning, KPI calculation, and a simulated refresh process, similar to how BI and RevOps teams streamline reporting workflows.

Key Features

• Data Generation: Simulates messy, real-world CRM data (incomplete, duplicated, and inconsistent).
• Data Cleaning: Standardizes data, handles nulls, and removes duplicates.
• KPI Calculation: Computes conversion rates, average deal size, and average deal velocity.
• Automated Refresh Simulation: Uses a simple schedule to mimic a weekly update.
• Failure Alert Simulation: Provides basic error handling through simulated alerts.

Technologies Used

• Python: Core logic and automation simulation.
• Pandas: Data manipulation and cleaning.
• NumPy: Data generation and randomization.
• Time: Simulated scheduling.

Project Structure

automated-kpi-dashboard/
├── data_generator.py       # Simulates messy CRM data
├── data_cleaning.py        # Cleans and standardizes data
├── kpi_calculator.py       # Calculates key sales metrics
├── report_refresher.py     # Legacy runner (replaced by main.py)
├── notification_sim.py     # Simulates failure notifications
├── output/                 # Folder for saved .csv report
├── main.py                 # Runs the full pipeline
└── README.md               # Project documentation

Running the Project

1. Clone the repository:

   git clone https://github.com/regankight/automated-kpi-dashboard.git

2. Run the main script:

   python main.py

3. View the KPI report in your terminal and as a CSV in the output/ folder.

Why This Matters

This project demonstrates my ability to combine BI logic and process automation to produce repeatable, reliable insights. These are skills I developed at Smarsh, GoRefime, and through freelance work. By simulating real-world data challenges, it highlights how to maintain data quality while automating reporting.

Connect with Me

Check out more of my work on GitHub or LinkedIn.