Using the Usage Tab
Once BI connectors or query log mining are configured, each product's detail page gains a Usage tab showing real-world consumption data.
Accessing the Usage Tab
- Open any data product from the catalog
- Click the Usage tab
Usage Metrics
Query Frequency
See how often your product is queried over time. The query count covers all detected queries — from BI dashboards, saved queries, ad-hoc analysis, and warehouse query logs.
Top Users
View the users who query this product most frequently. This helps identify key stakeholders and consumers who should be notified about schema changes or deprecation.
Popular Columns
See which columns are most referenced in WHERE, GROUP BY, and SELECT clauses. This reveals:
- Which columns are critical for filtering and aggregation
- Which columns may need documentation or quality checks
- Which columns are candidates for indexing or optimization
Common Query Patterns
Anonymized examples of frequent query patterns against this product. Literals are replaced with ? for privacy. These patterns help new analysts learn how others use the data.
Popularity Score
Every product receives a popularity score from 0 to 100 based on:
| Factor | Weight | Description |
|---|---|---|
| Query volume | 50% | Number of queries in the last 30 days (log-scaled) |
| Unique users | 30% | Number of distinct users querying the product |
| Recency | 20% | How recently the product was last queried |
Higher popularity scores boost the product in search results, ensuring analysts find the most-used datasets first.
Unused Products
Products with zero queries over a configurable lookback window are surfaced as unused. You can view unused products from the catalog filters to identify candidates for:
- Deprecation — Products that nobody uses may be safe to archive
- Documentation — Undocumented products may be unused simply because nobody knows about them
- Review — Check with the owner to confirm whether the product is still needed
Dashboard Lineage
When a product is consumed by BI dashboards (detected via BI scraping), dashboard nodes appear in the product's lineage graph:
- Dashboard nodes show the BI tool icon (Superset, Metabase, etc.)
- Clicking a dashboard node opens the original dashboard via its link-back URL
- Edges from tables to dashboards complete the end-to-end lineage from warehouse to report