> ## 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.

# Data diagnostics

> Diagnose data quality issues and identify root causes.

**Data diagnostics** helps you understand shape, distribution, and freshness problems inside datasets without exporting everything to a notebook first. Use it when dashboards look wrong but pipelines show green checks.

## Diagnostic tools

<Steps>
  <Step title="Select a dataset or run artifact">
    Pick a table, stage output, or sample file from a recent run.
  </Step>

  <Step title="Launch profiling">
    Compute column statistics: null rates, distinct counts, min/max for numerics, top values for categoricals.
  </Step>

  <Step title="Compare baselines">
    Contrast the current profile to a saved baseline or prior run to spot drift.
  </Step>

  <Step title="Trace upstream">
    Jump to the pipeline nodes that produced the artifact and open recent logs.
  </Step>
</Steps>

## Data profiling

Profiling is read-only against connections your role can access. Large tables use sampling strategies documented in the UI—note sample percentages when interpreting rare-value counts.

<Tabs>
  <Tab title="Key integrity">
    Check primary key uniqueness and foreign key presence when diagnostics expose join cardinality hints.
  </Tab>

  <Tab title="PII awareness">
    Mask or hash columns flagged as sensitive before sharing diagnostic exports externally.
  </Tab>
</Tabs>

## Anomaly detection

**Anomaly detection** models seasonality on volume, null rate, or numeric aggregates. When today’s batch diverges from the forecast band, diagnostics opens an incident card with likely upstream nodes.

<Info>
  Seasonal businesses (retail holidays, fiscal close) need custom calendars; configure blackouts or adjusted baselines before major events.
</Info>

<AccordionGroup>
  <Accordion title="False positives">
    If legitimate marketing spikes trigger alerts, widen bands temporarily and document the business event in the alert note.
  </Accordion>

  <Accordion title="True negatives">
    Silent failures can occur when bad data still matches schema—pair profiling with business reconciliation queries monthly.
  </Accordion>
</AccordionGroup>

## Related topics

<CardGroup cols={2}>
  <Card title="Dead letter queue" icon="triangle-exclamation" href="/observability/dead-letter-queue">
    Inspect rows that failed validation or loads.
  </Card>

  <Card title="Data contracts" icon="file-contract" href="/governance/data-contracts">
    Encode rules diagnostics surfaces informally today.
  </Card>
</CardGroup>