SellerScore explained
The SellerScore is the headline of every result row. One number. 0–100. Higher is better.
This page explains how it’s computed, what each sub-score contributes, and how to read it.
In this guide:
- The formula
- The six sub-scores
- The default weights
- Interpreting the number
- When to trust it less
The formula
SellerScore = weighted average of the six sub-scores, each on a 0–100 scale.
SellerScore =
weight_demand × demand
+ weight_margin × margin
+ weight_competition × competition
+ weight_trend × trend
+ weight_seasonality × seasonality
+ weight_cross_platform × cross_platformThe weights normalize automatically — they don’t have to sum to 100 (we just present them as percentages because that’s how people think).
The six sub-scores
| Sub-score | Measures | Source |
|---|---|---|
| Demand | How many people want this product | Amazon SFR (Brand Analytics) where available; BSR + Google Trends fallback. → Demand |
| Margin | Profitability per unit | Sell price − COGS estimate − FBA fee − referral fee. → Margin |
| Competition | How crowded the niche is | Review counts, ratings, listing density. → Competition |
| Trend | 12-month direction | Google Trends 12-month delta. → Trend & seasonality |
| Seasonality | Spiky vs. smooth | Google Trends monthly variance. → Trend & seasonality |
| Cross-platform | Pricing leverage from non-Amazon platforms | Crawlee (AliExpress, TikTok, Google Shopping). Coming soon |
Each sub-score is itself a 0–100 number. You can see all six on the result row as a chip strip — useful for quickly understanding which sub-score is dragging the composite up or down.
The default weights
| Sub-score | Default weight |
|---|---|
| Demand | 25% |
| Margin | 20% |
| Competition | 20% |
| Trend | 15% |
| Seasonality | 10% |
| Cross-platform | 10% |
These reflect a balanced “general high-conviction product” prior. Override per-search if you have a different strategy. → Filters & weights
Interpreting the number
| Range | Read it as |
|---|---|
| 80–100 | Excellent. High-conviction; worth deep evaluation. |
| 60–79 | Good. Worth a second look; may need one weak sub-score addressed. |
| 40–59 | Marginal. Probably not worth pursuing without a clear edge. |
| 20–39 | Weak. Multiple sub-scores are negative. |
| 0–19 | Avoid. The composite is dragged by structural issues. |
A 75 is the “I should look at this carefully” threshold for most sellers. An 85+ is rare; when it shows up, expect to find data-quality caveats or a niche that’s about to blow up (so other sellers will catch on too).
When to trust the SellerScore less
The composite hides which sub-scores were on real data vs. fallback. Always glance at:
- Data-quality badge on the row. → Data quality
- The chip strip for the six sub-scores. If three or more are missing or red, the composite isn’t well-grounded.
- The score-reasoning explanation on the product detail page — this is where the AI tells you what’s solid and what’s a guess.
A high SellerScore + low data quality = an informational signal, not a decision signal. Verify before acting.
How weights affect the composite
If you bump a sub-score’s weight, you’re saying “this dimension matters more than usual.” A product with that sub-score very high will rise; a product with that sub-score very low will sink.
Examples:
- Bump margin to 35% → high-margin products climb the rankings; low-margin products fall.
- Bump trend to 30% → climbers climb; flat / declining products fall.
- Bump competition to 35% → uncrowded niches climb; saturated niches fall.
The composite is always a weighted average; what changes is which products end up at the top.