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

# Dashboards & Metrics

> Build custom dashboards over your journeys, users, and signals

The [Journeys](/docs/journeys/overview) page is for looking at one journey at a time. **Executive Dashboards** is the aggregate view: counts, trends, funnels, and timings across all of them.

## Widgets

A dashboard is a company-level board of widgets. Each widget is one chart or tile, and comes in one of four kinds:

| Kind           | What it shows                                                                                          |
| -------------- | ------------------------------------------------------------------------------------------------------ |
| **Number**     | A single number for the window                                                                         |
| **Line chart** | The same number, as a trend over time                                                                  |
| **Funnel**     | An ordered sequence of steps, checked per journey: did this journey hit step 1, then step 2, in order? |
| **Cycle time** | How long it typically takes to get from one event to another (the median)                              |

Number and Line chart widgets also take a **measure**: what you are counting. Journeys and users are the journey-side measures; traces, total trace cost, average trace latency, and trace error rate are the trace-side ones, so project metrics like LLM spend or error rate are ordinary widgets you can filter like any other.

## Building a filter with Selectors

Number, Line chart, and Funnel widgets are driven by a **Selector**: a query builder over your signal events and properties, no code required. Point it at an event name, add property filters, and the widget updates.

## Drilling into a widget

Widgets are clickable. Select a point, bar, funnel step, or tile and a panel opens listing the exact journeys or traces behind that number, so a spike or a drop-off leads straight to the records that produced it. On the traces view you can select traces and route them onward with the same actions as the [Findings](/docs/findings/overview) trace list — assign them for review, add them to an [evaluation dataset](/docs/judges/datasets-and-evaluations), or add them to a [generation seed dataset](/docs/concepts#where-seed-data-comes-from) — so a drop-off you spot in a funnel can become a review batch or a generation seed without leaving the dashboard. **Open full page** carries the same filter into [Journeys](/docs/journeys/overview) or Traces.

## Funnels and value

Beyond step counts, each funnel step can carry what one journey reaching it is worth: an amount, plus an optional label for it. That lets you layer a simple value or cost model on top of the funnel — for example, attaching a dollar value or a cost to a step — so the funnel shows not just how many journeys got through, but what that flow was worth.

## Next steps

<Card title="Journeys" icon="route" href="/docs/journeys/overview">
  Browse the individual journeys behind these numbers
</Card>