> ## Documentation Index
> Fetch the complete documentation index at: https://closedloop.sh/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Glossary

> Definitions for every key term in ClosedLoop AI: insights, themes, features, opportunities, context, facets, severity, RIC scoring, and more.

### Behavior

Product usage data from analytics platforms (Amplitude, Mixpanel). Shows what customers actually do, complementing what they say in conversations.

### Churn Risk

An insight flagged as threatening an existing customer relationship. Identified from expressed dissatisfaction, competitor comparison, or explicit churn language.

### Context (Strategic Intelligence)

Business intelligence extracted alongside product feedback: satisfaction levels, churn reasons, competitor mentions, decision criteria, win/loss reasons. Answers "why" behind what customers say. In the REST API this is the `/context` resource; each record resolves to a `customer_id`, so it lines up with insights from the same customer.

### Deal Blocker

An insight flagged by the AI as preventing or delaying a purchase. Identified from explicit statements ("this is a deal breaker") or escalation language in conversations.

### Facet

A filter dimension and its value counts over the full dataset: categories, severities, sources. The API's `/facets` endpoint returns every available value with its count, useful for building filter menus that only offer values that actually exist in your data.

### Feature

A buildable, shippable child of a theme. A broad theme like "Payment issues" may contain features like "Double charges", "Payment timeouts", and "Refund failures". Features are the roadmap-ready layer of the hierarchy; in the REST API, `/features` returns them (filter by `theme_id`).

### Feature Area

The part of your product an insight relates to. Assigned by the AI, not from a predefined list but discovered from the language customers use.

### Insight

A single piece of product feedback extracted from a customer conversation. Contains the verbatim quote, who said it, how urgent it is, and what product area it relates to. One conversation may produce multiple insights.

### MCP (Model Context Protocol)

The protocol that connects ClosedLoop AI to AI development tools like Claude Code and Cursor. Lets AI agents query customer evidence while building features.

### Opportunity

A group of related insights automatically clustered by ClosedLoop AI. Represents a product problem or theme, scored by business impact. Updated daily as new evidence arrives. At the top level of the hierarchy, an opportunity is what the REST API and the roadmap call a theme.

### RIC Score

**R**each x **I**mpact x **C**onfidence: the first three dimensions of RICE prioritization, computed automatically from customer evidence. Effort is excluded because only your team knows what things cost to build.

### Severity

Business urgency of an insight: **Critical** (immediate revenue impact), **High** (significant pain), **Medium** (notable friction), **Low** (minor inconvenience).

### Theme

A top-level problem cluster of related insights, RIC-scored, rolling up one or more features. "Theme" is the name the REST API and the roadmap use for what the concept docs call an opportunity at the top level of the hierarchy.

### Velocity

Growth trend of an opportunity: **Accelerating** (getting worse), **Active** (steady), **Cooling** (declining), **Dormant** (no recent evidence).

### Won Deal Value

Total value of closed-won deals for a customer, in their CRM's home currency. Used to weight opportunities by revenue impact.