> For the complete documentation index, see [llms.txt](https://docs.hivel.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hivel.ai/ai-adoption/claude-enterprise-analytics.md).
# Claude Enterprise Analytics
Claude Code offers insights into user activity through IDE telemetry and Anthropic’s Usage API. Hivel captures and visualises these metrics across four areas: adoption, engagement, effectiveness, and cost. This guide documents every metric, its exact definition as shown in the dashboard, and the logic behind it.
#### Metric Categories
These categories break down how Claude metrics are measured:
| Area
| Metrics Covered
|
| ------------- | -------------------------------------------------------------------------------------------------------------------- |
| Adoption | Total Active Users, Total Inactive Users, Total Cost, User Segments (Power / Steady / Opportunistic / Inactive) |
| Engagement | Suggestions Acceptance rate by mode (Edit / Write / Multi-Edit), Sessions per user |
| Effectiveness | Edit Acceptance Rate, Write Acceptance Rate, Multi-Edit Acceptance Rate, Commits, PRs Opened, LOC Added, LOC Removed |
#### Key Metric Definitions
**Summary Metrics**
1. Total Active Users
* Lines of code written using AI for the selected duration.
* Displayed: Headline KPI card on the Overview tab.
2. Total Inactive Users
* Users with no lines of code generated using AI the selected period.
* Displayed: Headline KPI card on the Overview tab.
#### Suggestions Acceptance
Tracks the total number of times users have accepted AI-generated suggestions from Claude. Measured separately for three interaction modes.
1. Edit Acceptance Rate
* Percentage of suggestions accepted while editing an existing code.
* Formula: Accepted Edit Suggestions ÷ Total Edit Suggestions Shown × 100
* Tracked Per: User, per day
2. Write Acceptance Rate
* Percentage of suggestions accepted while writing a new code.
* Formula: Accepted Write Suggestions ÷ Total Write Suggestions Shown × 100
* Tracked Per: User, per day
3. Multi-Edit Acceptance Rate
* Percentage of suggestions accepted while making multiple code changes in parallel.
* Formula: Accepted Multi-Edit Suggestions ÷ Total Multi-Edit Suggestions Shown × 100
* Tracked Per: User, per day
* Notes: A rate of 0% across the period indicates the Multi-Edit mode is not yet in use by the team.
#### Developer Productivity Metrics
These metrics correlate Claude usage with developer throughput. The key question they answer: in periods of high Claude usage, is developer output - measured by commits, PRs, and lines of code - also higher?
1. Commits
* Number of git commits created through Claude Code's commit functionality.
* Purpose: Correlates Claude usage with commit frequency. High Claude activity alongside high commit counts indicates Claude is contributing to faster code delivery.
* Tracked Per: User, per week (or selected granularity)
* Attribution: Via git email matched to Hivel user account.
2. PRs Opened
* Number of pull requests created through Claude Code's PR functionality.
* Purpose: Correlates Claude usage with PR throughput. Use alongside Commits to understand whether Claude is accelerating the full delivery cycle, not just code writing.
* Tracked Per: User, per week (or selected granularity)
* Notes: Counts the PR creation event. Does not require the PR to be merged.
3. LOC Added
* Total number of lines of code added across all files by Claude Code.
* Purpose: Shows the volume of code written so you can correlate it with Claude usage in the same period. High LOC Added alongside high token usage suggests Claude is directly contributing to code volume.
* Tracked Per: User, per week (or selected granularity)
4. LOC Removed
* Total number of lines of code removed across all files by Claude Code.
* Purpose: Shows how many lines were deleted or refactored. Correlate with Claude usage to see whether Claude is also contributing to code cleanup and technical debt reduction.
* Tracked Per: User, per week (or selected granularity)
👥 User Classification
Every billed user is automatically classified into one of four segments based on their Claude token usage during the selected period. Segments are calculated dynamically using a rank-based percentile formula applied to token consumption across all users in scope.
#### Segment Definitions
| Segment
| Percentile Rule
| Tooltip Definition
|
| --------------------------------- | -------------------------------------------------- | ------------------------------------------------------------------------ |
| **Power Users** | Percentile ≥75% (top 25 percentile by token usage) | Top 25% of users by Claude token usage. |
| **Steady Users** | 50th–75th percentile token usage | Regular Claude usage, within the 50–75th percentile. |
| **Opportunistic Users** | <50th percentile, >0 tokens | Occasional Claude usage, below the 50th percentile. |
| **Inactive Users** | 0 tokens in selected period | User who has licence but has no Claude usage during the selected period. |
#### Usage Tab: User-Level Table
The Usage tab provides a sortable, per-user breakdown of every Claude metric. Switch to the Usage tab at the top of the Claude Adoption dashboard to access this view.
#### User Tags
Each user row can carry one or more tags based on their activity profile. Tags appear on individual rows and are filterable at the top-right of the Usage table (Usage > Output · High Delivery · New User).
| Tag
| Definition
|
| ----------------------------- | ---------------------------------------------------------------------------- |
| **New User** | First Claude session occurred within the selected time range. |
| **High Delivery** | Users who are part of the top 25% by commits and pull requests created. |
| **Usage > Output** | High Claude activity compared to peers, with lower relative delivery output. |
Clicking a tag button at the top-right of the table filters the entire list to show only users matching that cohort. Multiple filters can be combined.
🧱 Technical Foundations
* Freshness: Metrics reflect activity up to the previous calendar day (T-3 lag). Data is finalised at the end of each UTC day.
* Retention: Historical data is available for the period configured in your Hivel workspace (default: 30 days).
This guide provides a complete reference for all metrics in the Claude Adoption dashboard. Use it alongside the dashboard to diagnose adoption patterns and build data-driven enablement strategies.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://docs.hivel.ai/ai-adoption/claude-enterprise-analytics.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.