Documentation Index
Fetch the complete documentation index at: https://docs.datagenie.ai/llms.txt
Use this file to discover all available pages before exploring further.
What Explorer is
Explorer is interactive, multi-dimensional exploration of any KPI across time, segments, and contributors. Where Top Stories surfaces what changed, Explorer is where you validate why — confirming a root cause, comparing segments, or testing a hypothesis before acting on an autonomous insight.When to use Explorer
- A Top Story surfaced a change and you want to pressure-test the root cause
- You want to compare two KPIs side-by-side without building a dashboard
- You want to rank segments by Top N or Bottom N contribution to a change
- You need a freeform view that doesn’t fit a standard story or breakdown
What you get
Correlation Matrix
See which KPIs move together — including metrics auto-connected across sources by Nirvana.
Dimensional Analysis
Rank dimension values by Top N or Bottom N contribution — understand which segments drive or resist a change.
Data Playground
Freeform KPI × dimension exploration. Compare timeseries, overlay filters, toggle chart/table views.
AD Model Group selector
Switch between anomaly detection baselines on the fly — useful when comparing default vs custom AD Groups with Multi Yhat.
How it works
| Situation | Start with |
|---|---|
| ”Which KPIs tend to move together?” | Correlation Matrix |
| ”Which regions / channels carry the most impact?” | Dimensional Analysis (Top N) |
| “Which segments are underperforming most?” | Dimensional Analysis (Bottom N) |
| “I want two KPIs on the same chart” | Data Playground |
| ”I need raw numbers for a specific period” | Data Playground → Table view |
| ”Compare this KPI for UK vs US” | Data Playground → Dimension filter |
What’s next
Correlation Matrix
Explore multi-source KPI relationships.
Dimensional Analysis
Rank contributors and opposers.
Data Playground
Freeform exploration.
Top Stories
Start here — stories give you the what, Explorer gives you the why.
Filter Presets
Once you’ve found the right segments, save them as a Filter Preset.
Wisdom
Ask a follow-up question in plain English.