Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.humanintelligence.com/llms.txt

Use this file to discover all available pages before exploring further.

The Methodology Copilot is an AI assistant built into Human Intelligence that helps you work with governed definitions — without bypassing the review controls that keep those definitions trustworthy.
Methodology Copilot

What It Helps With

Explain a formula

Ask the copilot to walk you through what a metric’s formula does in plain language. Helpful when you inherit definitions from another team.

Suggest refinements

Describe what you want to change and the copilot will suggest an updated formula. You review the suggestion before anything is saved.

Help create segments

Tell the copilot which scope you want to isolate and it will help you build the right segment definition.

Translate business questions

Describe a business need in everyday language and the copilot will propose the methodology changes needed to answer it.

Guardrails

The copilot can draft and explain, but it can’t publish changes on its own. Every suggestion still goes through the standard validate, preview, and save lifecycle — the same process as any manual edit.
This means your team gets the speed of AI assistance without giving up the governance controls that make your definitions trustworthy. Every change is validated, previewed for impact, and captured in the audit log.

Change Requests

Sometimes you’ll spot something that needs fixing but you don’t have permission to publish the change yourself. In that case, the copilot can help you create a change request. A change request captures what you want to change and why, then routes it to someone with the right permissions to review and publish. It’s a lightweight way to contribute to your organization’s methodology without needing full edit access.
Not sure if a metric is defined the way you expect? Ask the copilot to explain it first. You might find the current definition already covers your use case.