This project is a full-stack analytics dashboard built to track and explore business performance using real-world data. It provides key performance indicators (KPIs), interactive filtering, and drill-down capabilities to support data-driven decision-making.
The dashboard integrates embedded analytics components using Embeddable, while maintaining custom frontend and backend logic.
The dashboard was developed as part of my work experience and is presented here as a portfolio project to demonstrate full-stack development and data integration skills.
Frontend
- React (Vite)
- JavaScript (ES6)
- HTML, CSS
Backend
- Node.js
- Express
Data & Analytics
- BigQuery
- SQL
- Cube.js (semantic layer / analytics API)
- Embeddable (embedded analytics framework)
Tools
- Git, GitHub
- VS Code
- dotenv
- KPI tracking (Revenue, Deals, Conversion Rate, Efficiency)
- Interactive filters and drill-down analysis
- Time-based performance analysis (date range & granularity)
- Secure API-based authentication for data access
- Integration with BigQuery for real-time querying
- React frontend for UI and interactivity
- Node.js backend for API and authentication
- BigQuery as the data warehouse
- Cube.js as the semantic layer for data modelling and querying
- Embeddable for embedded analytics components and dashboard rendering
- Designed and implemented the dashboard structure and UI components
- Built frontend functionality and integrated backend APIs
- Connected and queried data from BigQuery
- Implemented authentication and handled API communication
- Debugged SQL, API, and data consistency issues
- Resolved BigQuery permission and query errors
- Handled inconsistencies between different date fields (e.g. created vs closed dates)
- Implemented secure token-based authentication
- Improved usability through better layout and interaction design
This repository contains a cleaned and portfolio-ready version of the project. Sensitive data, credentials, and company-specific details have been removed.
This project is complete as an MVP and serves as a demonstration of my ability to build full-stack, data-driven applications.