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

# Use Cases

> A hierarchical taxonomy of what users ask your agent, kept current automatically and browsable down to individual conversations.

The Use Cases page shows what your users actually ask the agent to do, organized as a hierarchy: broad **user areas** at the top, specific **user goals** as leaves. Open **Understand → Use Cases** to see where conversation volume concentrates, how each use case is trending, and which conversations belong to each node.

## How the taxonomy is built

Moda splits each conversation into segments — contiguous stretches of one user goal — and clusters segments into the hierarchy. Each node carries a label, a description, keywords, and conversation counts.

* The first taxonomy for a workspace is built once at least **50 analyzed segments** exist. Before that, the page shows the empty state "No task clusters yet".
* New segments are assigned to the existing taxonomy as they are analyzed, so counts stay current.
* The hierarchy itself is rebuilt automatically as your traffic shifts — you never configure or trigger it. Segments that match no existing use case are held aside and grouped into new nodes at the next rebuild, so a genuinely new use case becomes visible once that rebuild has run.

## What the page shows

Two view modes over the same hierarchy:

* **Overview** — a ranked list of use cases with label, description, child-goal count (or "Leaf use case"), conversation count, share of total, and trend.
* **Browse** — the task tree side by side with an inspector pane for the selected node.

A search box matches across the whole hierarchy, and a time range control scopes counts to All time, Last 3 days, Last 7 days, or Last 30 days.

## Drill into a use case

Click a node to open its detail page. The header shows whether it is a **User area** (branch) or a **User goal** (leaf), plus a **Conversations** metric and — when enough data exists — **Recent trend** and **Signals to review**.

* **View conversations** jumps to the **Representative conversations** section: sampled real conversations for this node, each opening in a new tab, with a load-more control.
* **Goals within \<area>** breaks a branch node down into its child goals.
* **What to investigate next** lists health signals detected inside this use case, each with a link to the evidence.
* **Related goals** and **Related terms** in the sidebar show nearby nodes and the vocabulary users employ for this use case.
* **Create eval set** generates a replay set from this node's conversations and opens it in the [Playground](/dashboard/evaluate), so you can test changes against exactly this kind of work.

You can also filter the [Conversations list](/dashboard/conversations) by use case cluster, and the [Home page](/dashboard/overview) shows the top use cases in its "What the agent handles" section.

## Use cases in the API

The same hierarchy is available through the Data API: [`GET /clusters`](/data-api/analytics) returns the taxonomy and `GET /clusters/:nodeId/conversations` returns a node's conversations.

## Next steps

* [Conversations](/dashboard/conversations) — filter the conversation list by use case cluster.
* [Replays and Playground](/dashboard/evaluate) — turn a use case into an eval set and compare changes against it.
* [Analysis](/concepts/analysis) — how segmentation and clustering fit into the rest of the pipeline.