Measure Killer Measure Killer

Fabric notebooks inventory

Every Fabric notebook in the tenant — workspace, run and session durations, last run, schedule, lines of code — with full-text search across all notebook content.

Last updated · July 7, 2026

What you get

Notebooks tab — Fabric notebooks with run and session durations, last run status, run schedule, lines of code, and searchable content

Fabric notebooks across the tenant — where they live, how often and how long they run, when they last ran, and how big they are. Searchable across all notebook content — useful for finding which notebooks reference specific packages, libraries, or keywords.

For each notebook you see:

  • Name
  • Workspace and Workspace type (e.g. Premium)
  • Avg. run duration — average run time (HH:MM:SS); N/A if it has never run
  • Avg. session duration — average session length (HH:MM:SS)
  • Last run (UTC) — timestamp of the last run with a success/failure status indicator (N/A if it has never run)
  • Lines of code — total lines across all cells
  • Run schedule — the configured run schedule (e.g. Weekly on Monday at 15:30), or No schedule
  • CU (s) (new in v2.9.5) — the notebook’s Capacity Unit consumption for the default 14-day window, populated after fetching Capacity metrics (which holds the longer history)

Right-click a notebook and choose View job instances to drill into its full run history.

Run the analysis

  1. Run a tenant-wide scan and complete Phase 2 by selecting items and clicking Analyze.
  2. Switch to the Notebooks tab.
  3. Use the search bar to find specific packages, keywords, or data references across all notebook content.

Common workflows

  • Find notebooks that reference a specific package. Search for a package name (e.g. pandas, pyspark, delta) to find every notebook that imports it. Useful before upgrading or deprecating a library.
  • Audit notebook sprawl. Sort by workspace to see how notebooks are distributed. A workspace with dozens of notebooks may need cleanup or reorganization.
  • Find data source references. Search for server names, database names, or connection strings to find every notebook that connects to a specific source — the same workflow as the M Expressions search but for notebook code instead of Power Query.
  • Estimate migration effort. Sort by lines of code to identify the largest notebooks. These will take the most effort to review, refactor, or migrate.

What to do with the findings

  • Export as JSON — paid editions can export the full notebook inventory, including content, for downstream analysis.
  • Share the .measurekiller file — hand the scan to a colleague for review.