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

# Dashboard

> View pipeline health, run history, and key metrics at a glance.

The **DataOps dashboard** is the control tower for pipeline operations. You see which jobs are green, which are trending slower, and where data quality checks failed—all without opening each pipeline individually.

## DataOps dashboard

Widgets typically include:

* **Run outcomes** – success, failure, skipped over the last 24 hours / 7 days
* **Latency** – schedule drift (did the 06:00 job start at 06:00?)
* **Throughput** – rows or bytes processed relative to trailing averages
* **Ownership** – teams responsible for the noisiest failures

Filter by environment (dev/stage/prod), domain tags, or connection to focus incident response.

<Tabs>
  <Tab title="Executive view">
    Collapse to SLA tiles: on-time loads, critical dataset freshness, open incidents.
  </Tab>

  <Tab title="Engineer view">
    Expand to per-pipeline sparklines, error fingerprints, and last deploy markers.
  </Tab>
</Tabs>

## Health metrics

**Health** combines recency of success, error budget consumption, and dependency availability (for example, warehouse incidents). A pipeline can be “degraded” if it completes but misses freshness targets.

<Tip>
  Pair health metrics with on-call rotations so degraded states page someone even when jobs technically “succeed” with partial data.
</Tip>

## Quality scores

When you attach data quality nodes or contracts, the dashboard surfaces **quality scores** or failure counts per dataset. Drops often precede BI incidents; investigate before executives notice.

## Schema monitoring

**Schema drift** indicators compare observed columns to expected schemas or contracts. New unexpected columns, missing required fields, or type widening/narrowing appear as events you can route to alerts.

<AccordionGroup>
  <Accordion title="Noise control">
    Tune sensitivity so expected vendor experiments do not page nightly; whitelist known sandbox sources separately.
  </Accordion>

  <Accordion title="Cross-pipeline correlation">
    When multiple pipelines fail together, check shared infrastructure (warehouse, identity provider, DNS) before deep-diving each job.
  </Accordion>
</AccordionGroup>

## Related topics

<CardGroup cols={2}>
  <Card title="Diagnostics" icon="stethoscope" href="/observability/diagnostics">
    Drill into profiling and anomalies.
  </Card>

  <Card title="Alerts" icon="bell" href="/observability/alerts">
    Turn dashboard signals into notifications.
  </Card>
</CardGroup>