# Field-level Lineage

Source: https://userguide.qualytics.io/lineage/deep-dive/field-level-lineage/

Field-level lineage refines the container view by connecting individual fields between containers, giving you column-level visibility into how data flows and transforms across your pipeline.

## Expanding a Node's Fields

Each node shows a field count and how many of those fields have lineage connections, so you can tell at a glance whether there is field-level detail to explore. Click the expand control on a node to load its fields. Scroll down to load more, or use the search box to filter by name.

Fields with connections come first. When the list also holds fields without any, those follow under a divider such as **4 fields (no lineage)**.

A connection stays on the row of its field as the list scrolls. When a linked field scrolls out of the list, its connection moves to the edge it left through, and a pill such as **2 more linked above** or **1 more linked below** counts the fields hidden on that side. Click the pill to scroll to them.

In the lineage between two datastores, the count names the other end, as in **7 of 9 fields linked to Sales Warehouse**, counting only connections to that end. The divider there reads **2 fields (not linked)**.

You can also toggle all nodes at once from the toolbar's **Expand fields** / **Collapse fields** action to see all field-level connections across the graph in one view.

## Field Details

Each expanded field shows:

- **Type**: the column type, when available.
- **Active anomaly count**: how many active anomalies are open against that field.
- **Computed field marker**: the type icon carries it when the field is a computed field rather than one read from the source.
- **Lineage indicator**: a marker on fields that have at least one connection.

## Focal Field

The **focal field** is a single field the graph is centered on. While active, it highlights only the connections that touch that field, including connections in other containers. This gives you a focused view of a single column's flow without the noise of every other column in the same containers.

You set the focal field by clicking a field row in an expanded node. Clicking that same row again turns it off, and so does clicking another field, clicking a node, clicking the canvas background, or pressing **Esc**.

If the focal field is also connected to containers that are not drawn, a **+N** chip appears on its row. Its tooltip says how many containers, and clicking it adds them to the graph.

## Creating Field-level Connections

Field-level connections are created from the **Add** menu on a node, which needs **Editor** permission on the current datastore or the **Admin** role. Saving the connection requires Editor on both datastores for Members and Managers. The panel has a **Field** tab where you pick the field on each side. The menu is offered on a container's own lineage, not on a datastore's read-only views.

The transformation note that a connection can carry is not part of that form. It is set and read through the API, as the connection's `description`, and the graph does not display it.

For the full walkthrough, see [Add an Upstream Connection](https://userguide.qualytics.io/lineage/how-tos/connections/add-upstream-connection/) or [Add a Downstream Connection](https://userguide.qualytics.io/lineage/how-tos/connections/add-downstream-connection/) and switch to the **Field** tab.

## See Also

- **Impact Analysis**

  What a container feeds and what feeds it, ranked across a datastore and exportable as CSV.

  [Impact Analysis](https://userguide.qualytics.io/lineage/deep-dive/impact-analysis/)
- **Examples**

  Enrolling a warehouse, a medallion pipeline across datastores, a remediation output, and a hop only a person can record.

  [Examples](https://userguide.qualytics.io/lineage/deep-dive/examples/)
- **Best Practices**

  Which source to lean on, how to run collection, and how to keep a graph you can trust.

  [Best Practices](https://userguide.qualytics.io/lineage/deep-dive/best-practices/)
- **Permissions**

  What each role can do, which team permissions apply, and why a deleted connection can come back.

  [Permissions](https://userguide.qualytics.io/lineage/deep-dive/permissions/)
- **How It Works**

  The connection model, the two levels of granularity, the rules a connection has to obey, and how to find the containers that have one.

  [How It Works](https://userguide.qualytics.io/lineage/deep-dive/how-lineage-works/)
- **Lineage Sources**

  The five categories a connection can come from, and how each is created, updated, and removed.

  [Lineage Sources](https://userguide.qualytics.io/lineage/deep-dive/lineage-sources/)
- **Reading the Graph**

  Direction, nodes, assets outside Qualytics, where a connection came from, and expanding the graph.

  [Reading the Graph](https://userguide.qualytics.io/lineage/deep-dive/reading-the-graph/)
- **Datastore-level Lineage**

  The datastore map, the lineage between two datastores, and every container of a datastore that has lineage.

  [Datastore-level Lineage](https://userguide.qualytics.io/lineage/deep-dive/datastore-level-lineage/)

*Last updated: October 27, 2026.*
