About Policy Wonk
Policy Wonk is a free, searchable library of real public policy interventions from around the world: what worked, what didn't, and why. It is built for people who design, fund, study or discuss policy and want to learn from what has already been tried.
What's in it
190 case studies across 15 policy areas, from housing and health to misinformation and disaster preparedness. Each one follows the same short format: the challenge, one paragraph on what was done and what happened, a short analysis of why, a headline result and links to the original sources.
How cases are chosen and checked
- Real and sourced. Every case is a real intervention documented in published research, official evaluations or reputable reporting, with links to the original sources.
- The authors' numbers. Figures are reported as the study's authors or publishers state them. We don't do our own calculations or rounding.
- Failures included. Interventions that didn't work, or had mixed results, are as useful as successes, so they're included and labelled clearly.
- Evidence shown. Each case shows how strong its evidence is: randomized trials and meta-analyses (strong), quasi-experimental studies (moderate) or observational data (limited).
- Checked and corrected. Cases are fact-checked against their sources, and errors are corrected when found.
What the labels mean
- Worked: the intervention achieved its main goals according to the evidence.
- Mixed: some goals were met and others weren't, or results varied by place or group.
- Didn't work: the intervention failed to achieve its goals or caused harm.
Please check the sources
Summaries are a starting point and may contain errors. Always read the original sources before citing a case or adapting a policy.
Independent and lightweight
Policy Wonk has no ads, no tracking and no accounts. The site is plain, fast pages with no images or video, which keeps it quick to load and light on energy use.
Get in touch
Know a case we should add, or spotted an error? Send feedback.