Scoring transparency
Scoring weights and source rules
Last methodology review: June 9, 2026
Mulebuy comparisons are built for spreadsheet buyers, so the ranking starts with workflow fit and then checks QC evidence, shipping flexibility, storage, support, fee visibility, and account-history tradeoffs. The goal is to explain a shortlist, not to claim an official score from any agent.
Scoring dimensions and weights
| Scoring dimension | Weight | What we check |
|---|---|---|
| Spreadsheet workflow fit | 30% | How easily a buyer can move from spreadsheet row to seller link, order submission, warehouse intake, QC review, and parcel submission. |
| QC evidence depth | 20% | Included photo count, QC usefulness score, extra-photo workflow, and whether a buyer can catch obvious option, size, color, or condition issues. |
| Shipping-line flexibility | 15% | Listed line count, route-choice breadth, destination restrictions, and whether buyers are prompted to compare weight, volume, tracking, and final quote. |
| Storage and consolidation | 10% | Free and maximum storage windows, ability to wait for multiple spreadsheet items, and risk of rushing parcel submission. |
| Support and app access | 10% | Support score, app availability, chat availability, and whether unclear item details can be resolved before shipping. |
| Fee visibility | 10% | Whether service-fee rows are verified, listed, or unverified, plus reminders to compare final landed cost instead of headline fee claims. |
| Community and history | 5% | Public familiarity, account-continuity value, and whether an established workflow can outweigh a stronger spreadsheet-first score for a specific buyer. |
Score weight total: 100%. These weights explain editorial ranking logic; they are not official platform ratings and they do not replace a live checkout check.
Verified, listed, and unverified definitions
- Verified: The local dataset marks the fee row as verified for comparison use, but buyers still need to confirm the live checkout total.
- Listed: The fee value is listed in the local dataset or page content, but it needs current platform-side confirmation before a buyer relies on it.
- Unverified: The site should not treat the fee claim as dependable until a visible source or live account check confirms it.
Data sources
- agentData.txt local comparison rows parsed into the public data asset.
- Visible Mulebuy comparison tables and guide sections reviewed during each content pass.
- The generated Mulebuy agent comparison data page and JSON file.
- Editorial policy, correction policy, affiliate disclosure, and reader-visible live-check caveats.
Source confidence definitions
- High: Stable site-owned fields, generated data assets, or policy pages that can be checked directly in this repository.
- Medium-high: Local comparison data with clear source boundaries and repeated page-level reminders to verify live checkout details.
- Medium: Useful for shortlisting, but the buyer must confirm routes, payment, support, fees, coupons, and storage in the current platform account.
- Medium-low: Useful only as a starting point when the agent, route, fee claim, or support workflow can change quickly or lacks public verification.
Update frequency
Priority comparison pages should be reviewed monthly, and sooner when a major fee, route, storage, support, payment, or platform-access change is visible. Material changes update the page copy, update log, review date, and sitemap lastmod.
Downgrade and recommendation-change rules
- Downgrade an agent when fee visibility weakens, QC-photo access changes, route breadth drops, storage windows shrink, support becomes unclear, or live checkout no longer matches the comparison assumptions.
- Change the recommendation when another agent gives a stronger verified workflow for the same seller link, destination, parcel constraints, and buyer account context.
- Do not promote a page from improve to keep until the page has transparent scoring, dated evidence, source confidence, update log, and links to methodology or data assets.
Affiliate-bias handling
This guide hub may earn referral credit from some links, but affiliate availability is not a scoring dimension. Pages must keep alternatives visible, show when competitors may fit better, disclose referral links near CTAs, and link to the affiliate disclosure, editorial policy, and correction policy.
Methodology and data asset links: