// DATABRICKS COST INTELLIGENCE

Know what every data product actually costs.

Analytexa attributes Databricks spend to tables, pipelines, and data products — derived from lineage, not tags.

It runs inside your workspace as a Databricks App. Your billing and lineage data never leave your environment.

Request access // tell us about your workspace — accounts are onboarded directly contact@analytexa.com
// GRANULARITY

Cost per table, pipeline, and data product

Not another cluster dashboard. Spend allocated to the assets your business actually recognizes.

// MECHANISM

Lineage-derived, tag-free

Attribution follows your system tables' lineage — it works even where tagging policies never stuck.

// DEPLOYMENT

In your workspace, zero egress

Ships as a Databricks App. Read-only over system tables, no data leaves your environment.

// HOW IT WORKS
01 · read

Read-only over system tables

A scheduled job reads your workspace's billing, lineage, job-run, query, and storage system tables. No agents, no exports, no write scopes.

02 · attribute

Lineage does the allocation

Every DBU and byte is walked through table lineage to the tables, pipelines, and data products it produced — shared clusters and warehouses split by actual runtime. No tagging policy required.

03 · reconcile

The remainder gets named

What isn't attributable is classified — warehouse idle, reads, failed runs — and the total always reconciles to your bill. Nothing is smeared into an "other" bucket.

// SECURITY
Read-only SELECT on system.* — no write scopes requested
Zero egress by design — no external domains declared, built for SEG-restricted workspaces
No telemetry — diagnostics leave your workspace only if you export them
Results land in your own catalog — you keep the data, we never see it
Open data contract — attribution, per-product P&L, recommendations, and a FOCUS-conformant export are plain Delta tables in your Unity Catalog: query them, join them, build dashboards and models on top. The app shows you how; the app is a lens, not a cage.
// FREE X-RAY

One free notebook, ten minutes: import it into any Unity Catalog workspace and get your 20 most expensive tables end-to-end, your warehouse idle burn, and the SELECT-only spend nobody is charged for — reconciled to your exact bill, or the run fails. Source published for audit; nothing leaves your environment.

Get the X-ray on GitHub // two-region validation: zero reconciliation drift · 45% attributed to tables · 93% accounted by name
// GETTING ACCESS

Analytexa ships as a Databricks App you install from source in your own workspace — no SaaS signup, nothing to host, nothing leaves your environment.

01

Tell us about your workspace

Email contact@analytexa.com with your cloud, regions, and rough estate size.

02

You get the release

A private repo invite to a tag-pinned, human-readable release — inspect every line before anything runs in your account.

03

Install in your workspace

A guided ~30-minute install with a preflight that turns any missing grant into a copy-paste to-do. Results land in your catalog on a daily schedule.

Requirements: Unity Catalog–enabled workspace · system schemas enabled · account admin for one-time grants. Licensing: annual, per Databricks account, single tier — ask for a quote.