Government & Policy
Causalis — Impact Evaluation Platform
Design, run, and report impact evaluations without specialist expertise
Launch ApplicationAbout this project
Impact evaluation traditionally requires deep methodological expertise — randomised controlled trials, difference-in-differences, regression discontinuity. Most government policy teams don't have that bandwidth in-house, so evaluations either don't happen or get outsourced expensively.
Causalis was built in collaboration with LKY School of Public Policy to compress that workflow. AI guides the user through method selection, study design, data preparation, analysis, and reporting — surfacing the right approach for the policy question without requiring the user to know the methodology by name.
Key Features
- Four-stage workflow: agenda setting → design → analysis → reporting
- Two-knowledge-base RAG architecture (methodology + policy domain)
- AI guides method selection based on the policy question
- Built with LKY School of Public Policy
- Designed for non-specialist policy researchers
Technology
Claude with RAG, React, Node.js, Python/R, structured methodology knowledge base
Sectors
Want something like this built for your organisation?
Every app starts with a conversation about the problem, not the technology.