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

# Monitor AI Usage and Spend Across Your Organisation

> Monitor AI spend, requests, tokens, run hours, storage, and active agents across your org with filters, time ranges, and group-by breakdowns.

The Usage dashboard gives you a complete picture of where your organisation's AI resources are going. From a single screen you can see total spend, the number of AI requests made, tokens consumed, agent run hours accumulated, file storage used, and which agents were actually active — all broken down by model, agent, department, or individual user. Use the rich set of filters and time-range controls to answer specific questions: which team drove the most spend last month, which agent consumed the most tokens, or how this week's usage compares to the week before.

<Note>
  The Usage dashboard is accessible to **Org Owners and Admins only**. Non-admin users see an **Admin access required** message when they navigate to this page.
</Note>

***

## Metric tiles

The top of the Usage dashboard shows six metric tiles. Each tile displays:

* A **headline total** for the selected time range
* A **change indicator** comparing the current period to the equivalent previous period
* A **small stacked bar** showing the relative contribution of top items
* A **top contributors legend** identifying those items by name

| Tile              | What it measures                                     |
| ----------------- | ---------------------------------------------------- |
| **Spend**         | Total AI cost in your billing currency               |
| **Requests**      | Number of AI requests made                           |
| **Tokens**        | Combined input and output tokens across all requests |
| **Run Hours**     | Cumulative time agents spent running                 |
| **File Storage**  | Storage consumed, always broken down by department   |
| **Active Agents** | Agents that were actually invoked during the period  |

### Expanding a tile

Click the **expand icon** (maximise) on any tile to open a larger detailed view of that metric's trend over time. This full-width view makes it easier to spot patterns across longer date ranges or across many contributors.

### Showing and hiding tiles

Click the **gear icon** in the top-right of the dashboard to open a panel of on/off switches for each tile. Toggle any tile off to declutter the view for your current investigation. This preference is saved in your browser.

***

## Breakdown section

Below the metric tiles, the Breakdown table lists the heaviest contributors for the selected grouping axis. Each row contains:

| Column          | Description                                                  |
| --------------- | ------------------------------------------------------------ |
| Grouping axis   | The Model, Agent, Department, or User name                   |
| Requests        | Number of AI requests attributed to this row                 |
| Tokens          | Tokens consumed (input + output)                             |
| Spend           | AI cost attributed to this row                               |
| Run Hours       | Agent run time attributed to this row                        |
| Storage         | File storage attributed to this row                          |
| Trend sparkline | A small line chart showing activity over the selected period |

Click **Load more** at the bottom of the table to fetch the next page of results.

***

## Controls

### Time range

<Tabs>
  <Tab title="Presets">
    Click any preset button to instantly change the reporting window:

    | Preset | Window                 |
    | ------ | ---------------------- |
    | 1d     | Last 24 hours          |
    | 7d     | Last 7 days            |
    | 30d    | Last 30 days (default) |
    | 90d    | Last 90 days           |
    | 1y     | Last 12 months         |
  </Tab>

  <Tab title="Custom range">
    Click **Custom…** to open the date picker. Enter a **From** date and a **To** date, then click **Apply range** to update the dashboard.
  </Tab>
</Tabs>

### Group by

Use the **Group by** dropdown to change the grouping axis in the Breakdown table and the stacked bar legends on the metric tiles. Available options:

* **Model** — group by the AI model used
* **Agent** — group by individual agent
* **Department** — group by organisational department
* **User** — group by individual user

### Filters

Apply one or more filters to narrow the data shown across all tiles and the Breakdown table. Active filters appear as **removable badges** below the filter bar. Each badge has an **×** to remove that individual filter. Click **Clear all (N)** to remove every active filter at once, where N is the number currently applied.

| Filter      | What it narrows                                                          |
| ----------- | ------------------------------------------------------------------------ |
| Search box  | Filter by user, agent, or model name (free text)                         |
| Models      | Limit to selected AI models                                              |
| Agents      | Limit to selected agents                                                 |
| Departments | Limit to selected departments                                            |
| Users       | Limit to selected users                                                  |
| Mode        | Limit to agent execution mode: Quick / Deep / ReAct / Graph / Persistent |
| Status      | Limit to request outcome: Success / Error / Cancelled                    |

<Tip>
  Combine **Group by: Department** with a **Mode** filter to see which departments are making the heaviest use of long-running agent modes like Deep or Persistent.
</Tip>

***

## Common admin questions

<Accordion title="Which department spent the most on AI last month?">
  Set the time range to **30d**, set **Group by** to **Department**, and look at the **Spend** column in the Breakdown table. The first row is the highest spender.
</Accordion>

<Accordion title="Which model is consuming the most tokens?">
  Set **Group by** to **Model** and sort or scan the **Tokens** column. The Tokens tile legend also shows the top contributing models at a glance.
</Accordion>

<Accordion title="Is a specific agent running too long?">
  Set **Group by** to **Agent**, apply an **Agents** filter to select the agent in question, and review the **Run Hours** column and its sparkline trend.
</Accordion>

<Accordion title="How many requests failed this week?">
  Set the time range to **7d**, apply a **Status: Error** filter, and read the **Requests** tile headline total.
</Accordion>

<Accordion title="How does this month compare to last month?">
  Set the time range to **30d**. Each metric tile shows a change indicator comparing the current 30-day window to the previous 30-day window. Expand any tile for a side-by-side trend view.
</Accordion>

***

## URL state

Your current time range, filters, and grouping selection are automatically saved to the URL. You can bookmark or share a specific Usage view — anyone with admin access who opens the link will see the same filtered dashboard.