This is not your Metabase
Acme Analytics is a fictional company. The 540 questions, 18 duplicate groups, and 73 stale items are illustrative of what a typical mid-stage SaaS Metabase looks like after 2 years of self-serve growth. Your real X-Ray is generated from your live instance in under 90 seconds — no editing, no curation.
Run X-Ray on your Metabase →Acme Analytics (sample) — X-Ray Report
Scanned on April 29, 2026
Fair — significant issues to address
Your Metabase has accumulated 540 questions across 47 core tables, but 14% are stale or broken and 18 duplicate groups are eroding trust. The fastest path to an A grade is archiving the 73 unviewed-90-day questions, then merging the 4 largest duplicate groups (18% of total questions). That alone moves the score from C+ to B+ and removes most of the chaos analysts complain about.
Score Breakdown
47 core tables identified — 22 have no description. Auto-generate baseline docs from your dbt manifest or schema comments, then ask owners to review.
73 stale questions detected. Bulk-archive in Settings → Auto-Archive, or run our quarterly cleanup workflow to surface them in your inbox.
18 duplicate groups found. The top 4 groups account for 96 questions — merging them would remove 18% of total clutter.
73 questions failed last execution. Most reference dropped columns from the user_events_v1 → v2 migration. Bulk-fix or archive.
Strong overall — only 24 questions have draft/test/copy in their name.
Top Priority Actions
Archive 73 stale questions
Unviewed in 90+ days, not embedded in any dashboard, not used as a base question for SQL snippets.
+8 health score, removes 14% of clutter
Merge top 4 duplicate groups
Acme Revenue MoM (12 copies), New User Activation (24), Churn by Plan (31), Top Customers This Quarter (29). Each group has one canonical version with the highest view count.
+6 health score, 96 fewer questions to maintain
Document 22 undocumented core tables
Tables include orders, subscriptions, accounts, events, customers — the highest-traffic tables in your instance. Use the AI doc-gen or pull from dbt.
+9 health score, dramatic drop in 'is this column right?' questions
Fix or archive 47 broken queries from the user_events migration
Reference dropped columns: event_payload, event_user_uuid, raw_event. Most can be auto-rewritten against user_events_v2 — the rest should be archived.
+4 health score, eliminates a long tail of red-banner errors in your dashboards
Set up the quarterly cleanup workflow
MetaLens auto-surfaces new stale and duplicate questions every 90 days into your inbox, so the instance never re-accumulates.
Prevents future regression — keeps the score steady at B+ or higher
This is what your X-Ray will look like
Connect your Metabase, get a real audit in 90 seconds. Free X-Ray scan, no card needed. 14-day money-back if you upgrade.
Run X-Ray on my Metabase