Unused index review
Index audit
PostgreSQL index audit for small teams
Database Optimization Tool separates evidence from advice: index recommendations must be tied to scan counts, query behavior, and human-review risk notes.
Composite index candidates
EXPLAIN required before production changes
Missing index candidates
The audit looks for scan-heavy tables and slow query patterns, then asks for EXPLAIN evidence before treating an index as production-ready.
Unused index candidates
Large non-primary, non-unique indexes with low idx_scan can become review candidates, but Database Optimization Tool does not recommend blind drops.
Why Evidence Gate matters
AI can suggest plausible indexes. Evidence Gate keeps unsupported claims out of the official findings list unless metrics support the recommendation.
Related topics
RelatedLinks for neighboring audit problems
Use these focused guides to compare query pressure, index decisions, and maintenance signals before you change production.
FAQ
Frequently asked questions
These answers stay inside the current Database Optimization Tool product boundary: read-only collection, evidence-gated findings, and human-reviewed next steps.
Does Database Optimization Tool create indexes for me?
No. It keeps index work in the review lane and asks for EXPLAIN evidence before a team changes production.
Can it warn about missing and unused indexes in the same report?
Yes. Missing index candidates and low-usage index candidates are treated as separate review problems with different risks.
Why is statistics window coverage mentioned for unused indexes?
An index can look unused during a short or incomplete observation window, so the report warns against blind drops.
What should I read next if my problem starts with slow queries?
Start with the slow query and sequential scan guides, then come back to index-specific recommendations after you inspect the workload.