Measure Killer Measure Killer

End-to-end data lineage in Power BI (source → report)

Trace every data source through dataflows and semantic models to all the reports, paginated reports, and Excel files that depend on it — across your entire Power BI tenant.

Last updated · June 12, 2026

What you get

The Lineage tab gives you a searchable, cross-workspace dependency graph that traces every item from its data source through dataflows, semantic models, and all the way to the reports, paginated reports, Excel files, and downstream semantic models that consume it.

For column-level detail — following a single column rather than whole items — see what’s available today at each granularity in data lineage in Power BI — from source to visual.

The tab has two views, selectable from the dropdown in the top-left corner:

Data sources view

Lineage tab — data sources view showing downstream models and reports per source

Starts from every data source in the tenant and shows what flows downstream from it. Each data source row lists the type (SQL, File, Web, etc.) and the number of downstream semantic models and reports. Expand any data source to trace the path: source → semantic models → reports, paginated reports, Excel consumers, and downstream semantic models.

Use this view when you want to answer “what depends on this data source?” — for impact analysis before a schema change, a data source migration, or a retirement.

Models and dataflows view

Lineage tab — models and dataflows view showing upstream sources and downstream consumers for every model and dataflow

Starts from every semantic model and dataflow in the tenant and shows both directions: the upstream sources that feed into each item (other dataflows, models, or data sources) and everything downstream that consumes it (reports, paginated reports, Excel files, downstream models).

Expand any model or dataflow to see its full dependency tree in both directions. Use this view when you want to answer “where does this model get its data, and what consumes it?” — for understanding cross-workspace dependencies, planning workspace reorganizations, or tracing a number from report back to source.

Item-level vs column-level

The Lineage tab is item-level: it connects whole items — a data source, a dataflow, a model, a report — not individual columns. That is what’s available today, populated by Phase 1 of a scan.

One level finer — which models or dataflows touch a specific table or view — is also available today: search the M Expressions tab for the table or view name, and you get every item whose Power Query code references it (after Phase 2).

Column-level tracing exists today from the semantic model down to the visual: after Phase 2, the Where-used detail view shows every visual, filter, measure, and downstream reference for each column. What you cannot do yet is follow a single column upstream — from the source database through dataflows into the model. That piece ships in September 2026 as source-to-visual column lineage.

For the full picture of which lineage granularity is available today and which questions each one answers, see data lineage in Power BI — from source to visual.

Run the analysis

  1. Run a tenant-wide scan — lineage is populated automatically as part of Phase 1.
  2. Switch to the Lineage tab.
  3. Use the dropdown in the top-left corner to switch between Data sources and Models and dataflows.
  4. Use the search bar to find a specific data source, semantic model, or report by name.
  5. Expand any row to see its upstream and downstream dependencies.

Common workflows

  • Impact analysis before a schema change. A DBA wants to rename or drop a source column. Open the Lineage tab, search for the data source, and trace downstream to see every semantic model, dataflow, and report that depends on it — before making the change.
  • Find every report affected by a deprecated data source. When a data source is being retired, use lineage to build the complete list of models and reports that need to migrate to the new source.
  • Trace a number back to its source. An executive asks “where does this number come from?” Start at the report, follow the lineage upstream through the semantic model to the data source — no more guesswork.
  • Discover cross-workspace dependencies. Models and reports often depend on items in other workspaces. Lineage makes these invisible dependencies visible — critical before workspace reorganizations or capacity migrations.

What to do with the findings

  • Export as JSON — paid editions can export the full lineage graph as JSON for integration with data catalogs, governance dashboards, or change-management workflows.
  • Share the .measurekiller file — hand it to a colleague who can open it and browse the same lineage without re-running the scan.
  • Plan migrations — use the dependency graph to sequence migrations correctly: move data sources first, then dataflows, then models, then reports.