Use Case
See what your team knows — and what they don't
Track search patterns, identify documentation gaps, and measure knowledge health across your entire organization.
Search analytics
See the top 100 queries, which ones return results, and which ones hit dead ends. Fix gaps before they cost you.
Usage patterns
Which teams search most? Which docs are most referenced? Which sources are never touched? Data-driven knowledge management.
ROI measurement
Calculate time saved, support tickets deflected, and onboarding days reduced — real metrics for leadership.
“You can't improve what you can't measure. Knowledge analytics turns tribal knowledge into organizational intelligence.”
Capabilities
How it works for knowledge analytics
Gap analysis dashboard
Queries with no results or low-confidence answers reveal exactly where your documentation falls short.
Team-level insights
Compare search behavior across engineering, product, support, and sales. Each team has different knowledge needs.
Trending topics
Spot emerging questions before they become bottlenecks. A spike in "deployment process" queries means your docs need updating.
Source health scores
Each connected tool gets a freshness score. Stale Notion pages drag down your overall knowledge health.
Weekly reports
Automated email to knowledge managers with key metrics: searches, results quality, gaps identified, and docs updated.
Export & API
Pull analytics data into your existing BI tools. CSV export and REST API for custom dashboards.
Under the hood
The shape of a safe query.
Three guarantees, made visible. Reads run on the role you already trust. Writes pause for a person. The query is the receipt.
The write-gate, demonstrated
Two paths for two intentions — reads ship, writes wait.
A SELECT executes the moment it parses. Anything that mutates state pauses for a person on the other side of the gate. Same query surface, two doors.
Result rows return inline. Audit-logged, replayable, but the agent does not stop to ask.
INSERT, UPDATE, DELETE — all stop at the gate. A named human reviews the exact statement, then approves or denies.
The query is the citation
Question in, annotated SQL out, rows attached, rerunnable.
A natural-language question becomes SQL the analyst can read, with the schema-aware decisions called out. Result rows return with the exact query attached — replay it and you get the same answer.
Where the 1.4 seconds goes
Schema, plan, execute, cite — accounted for, segment by segment.
Most BI questions sit in queues for days. Here, the whole round-trip is under a second and a half — and every millisecond is labelled.
Inspect schema
~13% of the round-trip
Plan the SQL
~50% of the round-trip
Run on the warehouse
~25% of the round-trip
Cite + render
~13% of the round-trip
Median over a six-month rolling window. Warehouse exec varies with the query — schema and citation costs do not.
Ready to unify your knowledge?
Join the waitlist to be first in line when we launch.
No credit card requiredSetup in 5 minutes