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

# Analytics — Users

> Per-developer activity: tasks, cost, tokens, models and run quality

|              |                                                     |
| ------------ | --------------------------------------------------- |
| **URL**      | `/analytics/users`                                  |
| **Sidebar**  | Analytics → Users                                   |
| **Roles**    | Admin · Analytics Viewer                            |
| **Requires** | Enabling per organization · gated — contact support |

Users shows activity per developer — what they ran, what it cost, and how their agent runs went.

<Note>
  **Users** reads to **Admins and Analytics Viewers**.
</Note>

## What's on the page

### Headline figures

Four cards across the top, covering everyone the current filters match rather than the page on screen.

| Card                   | Shows                                                          | Underneath                                                                                                                     |
| ---------------------- | -------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| **Active Users**       | Developers with activity in the range                          | How many tasks they worked on                                                                                                  |
| **Tokens Used**        | Tokens across those developers                                 | How many developers that is                                                                                                    |
| **Agentic Users**      | Developers whose clean completion reaches **Steady** or better | The average clean-completion rate across developers with enough runs to score, as *N% avg strength* — or *Awaiting run signal* |
| **Peak User (Tokens)** | The largest single consumer's token count                      | Their email address. Selecting the card opens their page                                                                       |

### Table

Headed **Users** with the total count.

| Column                    | Contents                                                                                                                                                                                                            |
| ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Name**                  | The developer, with their email                                                                                                                                                                                     |
| **Longest hands-off run** | The longest single AI turn that ran without user input · sortable                                                                                                                                                   |
| **Median run**            | The typical AI-turn length · sortable                                                                                                                                                                               |
| **Clean completion rate** | The share of turns that ended in a clean completion rather than a request for clarification, an error, an interruption or an unrecorded outcome, badged **Power**, **Steady**, **Emerging** or **Light** · sortable |
| **Total Cost**            | Estimated spend · sortable                                                                                                                                                                                          |
| **Total Tokens**          | Tokens used · sortable                                                                                                                                                                                              |
| **Tasks**                 | Tasks worked on · sortable                                                                                                                                                                                          |
| **Models**                | How many distinct models were used · sortable                                                                                                                                                                       |

Twenty developers to a page. An info icon on each of the three run-quality headers explains what the number measures, and a column picker chooses which of Total Cost, Total Tokens, Tasks and Models are shown.

## Filters and controls

### Time range

The standard Analytics range selector applies.

### Filters

| Filter           | Values                              |
| ---------------- | ----------------------------------- |
| **User**         | The people you can read             |
| **Tool Type**    | The AI tools your organization uses |
| **Department**   | Departments in your organization    |
| **Team Manager** | Managers in your organization       |

## What you can do

| Action             | Result                                                                                                      |
| ------------------ | ----------------------------------------------------------------------------------------------------------- |
| Sort a column      | Orders every developer by that measure, not just the page on screen                                         |
| Select a developer | Opens their own page at `/analytics/users/[userId]`                                                         |
| **Export CSV**     | The developers matching your current filters, from the **Export Users** dialog · Admin and Analytics Viewer |

### A developer's page

**Total Cost**, **Total Tokens**, **Total Tasks** and **Avg Cost/Task** for the range, a **Model Usage by Cost** breakdown showing what share of their spend each model took, and their tasks — with the work type editable on your own tasks, and never for an Analytics Viewer.

Alongside those, the run-quality picture: weekly trends for clean-completion rate, median run and longest hands-off run, and an activity heatmap that always covers the last 12 months regardless of the range you picked.

<Tip>
  **Longest hands-off run** and **Clean completion rate** show how much your developers trust their agents. Both rising usually means the policies and setup are working.
</Tip>

<Note>
  Run-quality figures stay blank until a developer has enough agent runs in the window for the number to mean something: a developer with some runs but too few to score reads as a dash. A developer with no agent runs at all reads as zero, badged **Light**. To read either end of the distribution, sort **Total Tokens** — descending for your heaviest users, ascending for your lightest.
</Note>

## Messages you may see

| Message                      | What it means                                                 |
| ---------------------------- | ------------------------------------------------------------- |
| `No Users Found`             | Nothing matches the current range and filters. Widen either   |
| `Failed to update work type` | A work-type change made from a developer's page was not saved |

## Related

| Page                                          | For                                  |
| --------------------------------------------- | ------------------------------------ |
| [Users](/dashboard/users)                     | Roles, groups and invitations        |
| [Cost Management](/dashboard/cost-management) | Spend against limits, and exemptions |