Root cause driver tree
When a metric drops, the AI walks a key-driver tree column by column, drawing its path to the segment that explains the change, then opens an evidence sheet with a waterfall, a timeline and a hypothesis it can test.
Data and insightNo. 2332 variationsFree with an account
Playing on its own
Variation
- The idea
- Root cause as a visible search: the AI's walk through metric, dimension, segment and detail, then a testable hypothesis
- Autoplay
- Auto explore pulses the metric, reveals Platform, Region and Payment method with their share of the drop, follows Payment method, then Wallet, then Android 14.2 with a drawn path; the evidence sheet slides in; Test hypothesis runs three checks and stamps Likely cause; Mark as cause tags the node; then the cursor drills into Platform and Android by hand.
- When you click
- Click any node to expand it or open its evidence; tree keys (Up, Down, Right to go in, Left to go back, Enter); Auto explore (A); Test hypothesis, Mark as cause, Rule out; Esc closes the sheet and returns focus. On phones the tree is a sliding drill list with breadcrumbs and the evidence is a bottom sheet.
- Look
- hairline nodes with thin bars, sliding sheet, path draw, staggered columns, sans with mono changes type
- Variations
- Ice, Night teal
- Files
- index.html, style.css, script.js and prompt.md. No framework, no CDN, works offline.
The prompt and the files
Paste the prompt into Claude, ChatGPT, Lovable or any coding agent to rebuild it in your own product, or take the three files as they are.
Loading the files