Evidence

Evidence is an open-source web app that enables the creation of data-driven websites and reports. It replaces traditional drag-and-drop interfaces with a code-based approach, allowing users to build business intelligence products using SQL and Markdown. This method treats reporting as a software development process, facilitating version control and reproducible data storytelling.

The software functions by generating a static website from markdown files. It can be deployed via a VSCode extension, CLI, or Docker, and is compatible with static site hosting platforms such as Netlify, Vercel, or private infrastructure. Users write queries directly in their documents, and the system transforms those results into interactive visual components that are served as a web application.

Key features

• Generation of reports and websites from markdown files
• Execution of SQL statements within markdown to query data sources
• Rendering of charts and components based on query results
• Use of templated pages to generate multiple pages from a single template
• Implementation of loops and conditional If/Else statements for content control
• Integration with VSCode via a dedicated extension for project initialization
• Support for multiple deployment options including Docker and static hosting
• Use of a code-based workflow to replace manual dashboard configuration

Evidence is designed for data engineers, data scientists, and analysts who prefer a "BI as Code" workflow over traditional visual editors. By utilizing SQL for data retrieval and Markdown for layout, it allows for the creation of complex reports that remain maintainable as the underlying data grows. The architecture focuses on transforming raw data queries into a published web format, making it suitable for exploratory data analysis and corporate reporting. It integrates with the developer's existing toolchain, allowing for the same rigor in data reporting that is typically applied to application code.

It serves as a tool for those seeking a developer-centric approach to data visualization and business intelligence, prioritizing transparency and reproducibility over visual drag-and-drop editors.