Lift Test Designer for ChatGPT: Define Treatment, Control, and Spillover Risk
Plan an incrementality test with a treatment, control, outcome, duration, sample needs, and analysis. Consider how exposure might leak between groups.
中文Install in ChatGPT
Use an eligible ChatGPT workspace with permission to create skills and, for uploads, upload permission. The installation path for this text-only adaptation was checked against OpenAI documentation. Workspace runtime verification is pending.
- Download docs/research/external/imports-20260422/autonomous_marketing_agent_math_ai_proof_pack/skills/lift-test-designer/SKILL.md from the pinned GitHub source. Keep the name, description, and instructions for lift-test-designer together as a standalone text skill.
- In a ChatGPT workspace with skills and uploads enabled, open Plugins, Skills, Create, and Upload from your computer. Choose a skill package format accepted by the file picker. You can also choose Create with editor and enter the original name, description, and instructions.
- After the scan completes, check that lift-test-designer appears in Installed. Type @ in a new chat, select the skill, and use the fictional example on this page to check the output.
Example task
A fictional library wants to know whether a reminder adds attendance. Registrants can be randomized and the event stays the same. Draft a one-week test and account for household members forwarding the reminder.What to check
Keep reminder assignment and the original process distinct. Define attendance among eligible registrants. Record forwarding as spillover and leave sample size to be estimated from baseline data.
Original SKILL.md on GitHub · Updated
Frequently asked question
Can groups be defined by who actually clicks?
Analyze original assignment first and record observed exposure separately. Clickers may already have higher intent, so their behavior is suitable for a descriptive breakdown.