How we work.

Where our cases come from, and what we ask of the people in them

We write about things that did not go as planned, often involving people who are still working. That puts an obligation on us to be clear about our method and about what we ask of anyone we approach. This page is that.

How a case is built

Finding candidates

We search public databases of governance and participation projects, alongside the academic literature on policy transfer. Software helps us sort thousands of entries down to a few dozen worth looking at. It suggests. It does not decide, and it does not write.

Research

A person reads the evaluations, the coverage and the primary documents, and works out what the case actually rests on. Evaluation reports matter more to us than press releases, because evaluators name problems and press releases do not. If we cannot find at least two independent sources, we drop the case.

Checking with the people who were there

Where we can identify someone involved before publishing, we send them the draft first. Often we cannot. In that case we publish from public sources, say plainly on the page that nobody involved has reviewed it yet, and then go looking for the people who were there. Anything they tell us gets corrected on the page, and the correction is noted.

Publication

Every case carries its sources. Every case names what went wrong, including cases we consider successes.

If we contact you

We may write to you because you were involved in something we are documenting. Here is exactly what that means.

You will receive either a draft or a link to something already published, not a request for an interview. Read it, correct it, or ignore it.
You choose how you appear: by name, by role only, by institution only, or not at all.
You can decline to be named after seeing the draft, and after publication.
We are not asking you to admit anything. We are asking what you would do differently, which is a question people are usually glad to answer.
If you agree to be contactable, we introduce you to individual people who are attempting something comparable. You approve each introduction. There is no ongoing commitment, and you can withdraw at any time.
Where a case is already online, we correct it on the page and note the correction. We do not quietly edit.

What we do not do

We personally edit and check every case text before it is published. We do use an AI-assisted database internally to help manage and search candidates.
We do not run a reporting system. Nobody submits their own failures to us, and we could not offer legal protection if they did.
We do not rank countries, cities or governments. A case that did not transfer is not evidence that anyone was incompetent.
We do not publish anonymous accusations.

When we get it wrong

We will. If you find an error, write to us and we will correct it and note the correction on the page. If you were involved in a case and disagree with how we have described it, tell us and we will either revise it or publish your objection alongside it.

Working with people under eighteen

Direct introductions to practitioners begin at eighteen. Below that age we work only through schools, youth organisations and group settings, with an adult from that organisation present. We do not put a minor in one to one contact with an adult we have introduced.

Who pays for this

Nobody, at the moment. There is no funder, no advertising, and nothing is behind a paywall. If that changes we will say so here, and we will name who is funding what.

Questions about any of this: iwantto@learnfromtheworld.org