Two independent products
Fan50 maintains two strictly separate products: editorial service ratings and measured expert accuracy. They're never combined into a single score. A high editorial score doesn't imply good measured accuracy, and vice versa.
Editorial scoring rubric
Providers are scored on five dimensions using anchored 1 to 5 scales. The final score is a weighted mean displayed to one decimal.
| Dimension | Weight |
|---|---|
| Advice usefulness | 30% |
| Tools and usability | 25% |
| Value | 20% |
| Coverage and timeliness | 15% |
| Transparency | 10% |
- Evidence requirement: an editor must attach evidence to every dimension before publication. Missing required tests block scoring; they don't cause silent weight redistribution.
- Plan specificity: published reviews reference the exact plan tested. Material plan changes require re-review.
- Free sources: free resources receive a separate rubric and aren't silently ranked against paid software on tool breadth.
- AI assistance: AI may draft or flag changes but can't publish or assign final scores. All scores require reviewer sign-off.
Freshness standards
A 90-day editorial review reminder and a yearly price-check reminder are set internally. These are freshness targets, not automatic proof that content is wrong. Prices checked more than a year ago are shown as the last verified price, with the current price marked unverified. Displayed “verified” dates reflect when staff confirmed specific facts.
Accuracy methodology
OIThe Fan50 Order Index, weekly lock, evaluation universe and season gates are documented in full.Read the accuracy methodology