Revenue dropped 14% yesterday. Do you know why — or are you guessing?
Niblin decomposes every revenue move into traffic, conversion, and AOV — automatically. When something shifts, you get the mathematical root cause, not a hunch. Built for operators who need to explain the number, not just read it.
Revenue Decomposition
See exactly which lever moved — traffic, conversion, or AOV
Every revenue change is broken into its component parts. When TSP drops $4K, Niblin shows you whether it was fewer sessions, lower conversion, or smaller carts. No spreadsheet gymnastics. No Slack threads asking "does anyone know what happened yesterday?"
- Automatic decomposition into traffic, conversion rate, and average order value
- Channel-level breakdown — Shopify vs Amazon, app vs web, organic vs paid
- Comparison against 4-week, 8-week, and 13-week baselines so you know what "normal" looks like
- Mix effect vs performance effect separation — was it the channel mix that shifted, or did the channel itself underperform?
Anomaly Detection
Get alerted the day revenue moves — not the day you notice
Statistical anomaly detection runs every morning across revenue, orders, AOV, and every sub-metric. When revenue is 2.3 standard deviations below your 30-day baseline, Niblin flags it with severity scoring so you know what actually needs attention vs what is just noise.
- Daily anomaly scans across all revenue metrics and sub-metrics
- Severity scoring from Low to Critical so you can triage your morning
- Expected vs actual values with deviation percentages — not just "revenue is down"
- Persistent anomaly tracking — see if yesterday's drop is a one-day blip or a three-day trend
Root Cause Analysis
Trace every spike and drop to its source — deterministically
This is not an LLM guessing why revenue dropped. Niblin runs mathematical causal decomposition through your metric tree. It traces a revenue drop to the exact product, geography, or customer segment that caused it. Same inputs, same answer, every time.
- Deterministic root cause analysis — not probabilistic guesses from a language model
- Dimensional breakdown across products, geography, and customer segments
- Narrative roles for every node: hero, culprit, defender, drag — so you can tell the story in seconds
- Drill from top-line revenue down to the SKU or state that drove the change
Dimensional Breakdown
Which products, which states, which customers — in one click
When revenue spikes $5K, the first question is "where did it come from?" Niblin answers with a multi-dimensional view: top products by contribution, geographic hotspots, and new vs returning customer splits. No pivot tables. No analyst requests.
- Top 20 products ranked by contribution to the revenue change
- Geographic breakdown by state with contribution percentages
- New vs returning customer revenue split with period-over-period comparison
- Every dimension loads in parallel — full picture in under 5 seconds
Stop explaining revenue with anecdotes
Connect Shopify in 15 minutes. Tomorrow morning, you'll know exactly what happened and why — before anyone asks.