Benchmarks
Every number below is computed live from the catalogs published to LLMRank, using the same scoring engine that answers agent queries. No merchant is named and no individual listing is identifiable — this is the yardstick for judging your own scores.
Median score
56.3
out of 100
90th percentile
60.2
top 10% of listings
Listings scored
109
9 merchants
Agent click-through
5.8%
0 queries in 30 days
Score distribution
Median by category
| Category | Listings | Median | 90th pct |
|---|---|---|---|
| Apparel > Tops > Sweaters | 29 | 56.3 | 59 |
| Electronics > Audio > Headphones | 29 | 53.8 | 57.4 |
| Food & Beverage > Coffee > Whole Bean | 25 | 57 | 60.5 |
| Food, Beverages & Tobacco > Beverages > Coffee | 5 | 59.6 | 59.7 |
| Services > Home | 4 | 58.9 | 62.9 |
| Services > Tailoring | 4 | 49.8 | 49.9 |
Field coverage
The share of published listings that actually carry the fields an AI needs. Gaps here are the cheapest points available in the Completeness component.
Most common missing fields and what they cost
Point cost is measured from the data itself — the average score gap between listings that carry the field and those that don't.
| Field | Missing | Avg point cost |
|---|---|---|
| Brand | 15 (13.8%) | -0 |
| In stock | 14 (12.8%) | -1.5 |
| Image | 12 (11%) | -10.5 |
| Description | 10 (9.2%) | -1.6 |
53.2% of published listings are ready for agent checkout — image, description, brand, price and in stock.
Method: each listing is scored with its baseline query through the published six-component model. Percentiles use nearest-rank. Categories with fewer than three listings are omitted. Generated Thu, 29 Oct 2026 01:28:43 GMT.