The Esther Duflo
Framework
Randomised Trials & Poverty Economics
Esther Duflo represents one of the most distinctive thinking patterns in their field. Esther Duflo's approach, captured in this framework as mental models, decision heuristics, and the signature questions they would ask.
The randomised trials & poverty economics framework distils Esther Duflo's documented approach into a .md skill file you can load into Claude, ChatGPT, or any LLM in seconds.
Where most AI personas generate generic output, this framework is grounded in real documented work: interviews, writing, published frameworks, and the specific decisions Esther Duflo made when it mattered. Use it for decision-making, writing, strategy, or any situation where you want your AI to think with Esther Duflo's distinctive patterns rather than a default model voice.
A glimpse of Esther Duflo's thinking.
The poor are not a monolith defined by their poverty; they are rational agents navigating constrained choices, and understanding those constraints precisely—not romantically—is the precondition for helping them. Intuition about what should work for the poor is almost always wrong, and the cost of that wrongness is paid by the people you intended to help. Evidence generated through randomised controlled trials is not a bureaucratic hurdle; it is the ethical obligation of anyone wielding resources and power over vulnerable lives.
- 01
- 02
- 03
- 04
- 05
Five specific models, each named and explained with concrete examples, unlocked with purchase.
"What is the counterfactual, and is it credible given how this program was assigned?"
4 more questions in Esther Duflo's voice are unlocked with purchase.
Full list of anti-patterns unlocked with purchase.
The full framework adds: detailed explanations of all five mental models with real-world examples, the decision heuristics Esther Duflo actually used, remaining signature questions, the complete list of anti-patterns, a copy-paste activation prompt, and a worked example.
Unlocked the moment you purchase. ~2,000 words total. Delivered as a Claude Skill and plain .md within 60 seconds of checkout.
- ✦Core philosophy: 2–3 sentences grounded in Esther Duflo's documented worldview
- ✦Five signature mental models: each named, with concrete examples from their work
- ✦Decision heuristics: the concrete rules they actually used
- ✦Signature questions: five questions that sound like Esther Duflo, not a generic MBA
- ✦What they would NOT do: the anti-patterns this framework rejects
- ✦Copy-paste activation prompt: drop straight into Claude or ChatGPT
- ✦Worked example: the framework applied to a realistic scenario
- →esther-duflo.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →esther-duflo-framework.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
What it does with a real question.
A founder is deciding whether to take a large but difficult client.
Don't decide on the story, decide on the counterfactual. The founder is comparing "take the client" against a vague picture of "don't," and that comparison will mislead them. What actually matters is what happens to the company in each world. Take the client: revenue jumps, but so do concentration risk, support load, and the odds that your best people spend a year on one account. Pass: you keep the roadmap, but you also keep whatever growth problem made this client tempting. Name that problem first, because it determines whether the client is a cure or a distraction.
Then notice the selection effect. Large, difficult clients are usually difficult for a reason: they have unusual requirements, slow procurement, or a history of churning vendors. Treat that history as data about how this relationship will go, not as a challenge you will be the exception to. Ask for it directly: how many vendors did they use before, and why did each leave?
My position: take the client only if you can cap the downside cheaply. The best-case reading is real, but the founder's confidence is not evidence.
Next move: propose a paid, scoped pilot of 60 to 90 days with written success criteria and an exit clause, and tell them the terms are non-negotiable. How they respond to a small, structured test will tell you more about the full engagement than any amount of deliberation. If they refuse any pilot, you have your answer.
The full framework carries five models, decision heuristics, signature questions and a worked example.
Written by claude-sonnet-5-5 with this framework loaded, 2026-10-11. Checked automatically for length and for copied framework text.
Save up to 40% with a bundle
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Inspired by the documented thinking of Esther Duflo. A mental-model toolkit, not a literal representation. Delivered as a .md file compatible with Claude, ChatGPT, Gemini, and any LLM.