Introduction
Ever waste twenty minutes asking Slack where customer_ltv lives? You're not alone. Most teams rely on tribal knowledge instead of a proper data dictionary. Bloom fixes this with an auto-generated catalog that stays up to date and searchable—in plain English.
Why a Living Catalog Beats Static Docs
- Self-Service: Analysts answer their own "where is X?" questions.
- Faster On-Boarding: New hires explore data like browsing an app store.
- AI Context: Bloom's agent references column meanings to write smarter SQL.
How Bloom Builds the Catalog
- Schema Crawl: On first connect Bloom scans Snowflake, BigQuery, or Postgres.
- Natural-Language Descriptions: AI summarises table purpose and key joins.
- Lineage Mapping: Upstream and downstream links are auto-drawn.
- Search & Chat: Ask "fields for churn model" and jump straight to columns.
Quick Start
Point Bloom at your warehouse, hit "Generate Catalog." In about the time it takes to fetch coffee, your team gains a shareable URL with full schema, docs, and lineage graphs.
Real-World Impact
A fintech startup reduced onboarding time by 50 %. New analysts now explore data assets on day one instead of week two.
FAQs
Q: How often does Bloom update the catalog?
A: Nightly by default or instantly on schema change.
Q: Can I export docs?
A: Yes—Markdown, HTML, or push to Confluence.
Conclusion
Stop digging for column names and start analysing. Bloom turns data sprawl into a living, AI-ready catalog your whole org can rely on.