Features
Listing Quality Score
How the 0 to 100 Listing Quality Score is calculated, what each bucket checks, and how to read the breakdown.
Listing Quality Score
The Listing Quality Score (LQS) is Rufusly's own 0 to 100 rating of how complete, compliant, and AI-ready a listing is. It is calculated independently of the AI model that wrote the listing, using a fixed set of rules, so the score cannot be gamed by the AI marking its own work.
The five buckets
| Bucket | Max score | What it checks |
|---|---|---|
| Title | 20 | Length within the 75-character limit, no banned characters, key attributes present |
| Bullets | 30 | Five bullets present, each within the character cap, no keyword stuffing, no banned claims |
| Description | 20 | Length, readability, no banned characters or promotional claims |
| Backend search terms | 15 | Byte limit respected, no duplicate words already used in the title or bullets |
| AI readiness | 15 | How well the listing answers the questions Amazon's Rufus assistant is likely to ask a shopper |
Scores below the max in any bucket come with specific checks that failed, shown in the breakdown, and suggestions for fixing them.
Reading your score
| Score | Meaning |
|---|---|
| 80 to 100 | Good. Listing meets Amazon compliance and is well set up for AI discovery |
| 65 to 79 | Workable, but flagged for improvement |
| Below 65 | Needs attention before publishing |
AI readiness bucket
This bucket used to be labelled "Rufus" in some places and "Compliance" in others. Both now show as AI readiness everywhere in the app. It checks the same six criteria as the Rufus Optimisation Audit:
- Full ingredient names, not abbreviated
- Product format stated explicitly
- Serving size and frequency included
- Dietary suitability addressed (vegan, gluten-free, etc.)
- NRV percentages included, supplement category only
- Target shopper questions answered in the bullets or description
Where you see it
- Listing Studio, Step 3, as a ring and bucket breakdown
- Unified Product Editor, as a compact badge in the product header and a full breakdown in the Content tab
- MCP tool
get_lqs_score, for pulling the score from an AI client - Approvals, alongside the fields sent for review, so a reviewer can see the quality impact of a proposed change