Five criteria, one transparent calculation.
Each criterion receives a score from 0 to 100. We multiply each score by its weighting, add the results, divide by 20 and round to one decimal place. Community ratings are always shown separately and never change the EdgeCheck editor score.
Trust score = (Transparency + Evidence) ÷ 2
Transparency
30%Ownership, pricing, terms, conflicts and limitations are disclosed clearly.
Evidence & performance
30%Claims are supported by complete, verifiable and representative evidence.
Usability
15%The product is reliable, accessible and straightforward to use in practice.
Value
15%Price, features and realistic benefit compare fairly with the alternatives.
Support
10%Help, communication, complaints and cancellations are handled well.
A short winning run is not enough.
A tipster needs at least 90 days of tracking and 100 settled selections for a full rating. Before both thresholds are met, the rating is marked provisional and cannot exceed 3.5.
If we cannot verify a complete public results record, the overall score cannot exceed 3.0. We expect losing bets to remain in the record, use odds that customers could reasonably obtain, and avoid cherry-picked time periods. Past performance is never presented as a guarantee of future profit.
Evidence before opinion.
- 01
Scope the claims
We record the product’s promises, ownership, price, terms and intended audience before testing.
- 02
Use the product
We complete at least 15 hours of hands-on assessment and purchase access independently wherever possible.
- 03
Check the evidence
We verify functionality, support, pricing and performance records, including limitations and losing periods.
- 04
Score and challenge
The five criteria are scored independently, calculated automatically and checked against the written verdict.
- 05
Publish the workings
Every review shows its criterion scores, assessment status, latest score note and planned review date.
Scores are monitored, not frozen.
Linked tipster result graphs refresh from the published evidence sheet. A data refresh alone does not automatically change the editorial score.
Tipsters receive a formal reassessment monthly or after each additional 50 settled selections, whichever comes first.
We check material facts such as price, ownership, features, terms and availability.
Every published product is scheduled for a full retest at least once every six months.
Material product changes, credible complaints, misleading claims or evidence concerns can trigger an earlier review.
When a score changes, the review records the date and a plain-language reason. Previous status is not silently presented as a fresh assessment.
Commercial relationships do not enter the formula.
Affiliate links may earn EdgeCheck a commission, but they do not affect selection, testing, weighting or ranking. Providers may flag factual errors and supply evidence. They do not receive preview approval and cannot negotiate an opinion or score.
We correct confirmed factual errors promptly and explain material score changes. Community scores remain a distinct reader signal, with their own count displayed beside the editorial assessment.