# Signals frequency

Share of conversations carrying each signal tag — intent signals and agent gaps — as a ranked bar list.

**Path:** Analytics → Dashboard → Conversations

Signals frequency shows the share of conversations carrying each **signal tag**, as horizontal bars sorted by frequency. It combines two kinds of tag in one view: **intent signals** and **agent gaps**.

## Signal tags

Every tag is a boolean flag on a conversation, and tags **can overlap** — one conversation can carry several, so the percentages do not sum to 100%. The footer reads "Based on N conversations · signals can overlap."

| Tag | Kind | What it means |
| :--- | :--- | :--- |
| **Information Request** | Intent | The user is asking for information. |
| **Action Request** | Intent | The user wants the agent to do something. |
| **Reported Issue** | Intent | The user is reporting a problem. |
| **Agent Knowledge Gap** | Gap | The agent was missing information it needed. |
| **Agent Action Gap** | Gap | The agent was missing an action it needed. |

Each bar shows the tag's percentage; hovering gives the raw count ("X of N conversations (Y%)").

## Act on the gaps

The two gap tags are the actionable ones:

- **Agent Knowledge Gap** is growing → add to the [Knowledge Base](/user-guides/knowledge-base/).
- **Agent Action Gap** is growing → add or fix an [action](/user-guides/authoring-actions/).

Track these monthly. In [Reports](/user-guides/analytics/reports/), the Agent Gap metric breaks down by gap (Action vs Knowledge) over time.

## Related

- [User intent](/user-guides/analytics/user-intent/): the intent signals on their own, over time.
- [Conversation metrics](/user-guides/analytics/conversation-metrics/): resolution and handoff rates.