Editorial Policy

Last updated: 2026-07-31

Our Editorial Philosophy

Foremy’s reviews are built around a single principle: a rating is only useful if it was earned through actual, structured testing rather than assumption or vendor marketing material. This policy explains the methodology behind the Foremy Trust Score, our standardized rating system used across all eight coverage categories.

Separation of Editorial and Revenue

Writers and testers who score products do not have input into, or visibility of, affiliate commission rates for the products they are testing. Business development and editorial teams operate independently, and no individual or partner may purchase a specific score or ranking. See our Affiliate Disclosure for how we fund the Site.

The Foremy Trust Score Framework

Each category is scored out of 100 across five weighted pillars, tailored to what actually matters for that product type:

VPN Reviews

Security & Encryption (25%) — protocol support, kill switch reliability, independent audit history, logging policy verification.
Speed & Performance (25%) — measured throughput across multiple server locations and times of day.
Server Network (15%) — country/city coverage and consistency of connection quality.
Value for Money (20%) — pricing relative to feature set and simultaneous-device allowance.
Support & Usability (15%) — app design, setup friction, and responsiveness of customer support.

Game Reviews

Gameplay & Design (30%), Technical Performance (20%) (frame rate, stability, load times), Graphics & Sound (15%), Value & Longevity (20%), Community & Support (15%) (patch cadence, multiplayer stability).

AI Tool Reviews

Output Quality & Accuracy (30%), Ease of Use (20%), Feature Depth (20%), Pricing & Value (15%), Support & Update Frequency (15%).

Software Reviews

Core Features (25%), Usability (20%), Security & Privacy Practices (20%), Pricing & Licensing (20%), Support & Documentation (15%).

Testing Process

Research phase — reviewing vendor documentation, prior versions, and known issues.
Hands-on testing period — a minimum multi-day testing window using the product under realistic conditions (e.g., real streaming/torrenting tasks for VPNs, full playthrough segments for games, real prompt workloads for AI tools, real project use for software).
Scoring — testers score each pillar independently against the published rubric before a final weighted score is calculated.
Peer check — a second team member reviews the scoring rationale before publication.
Re-testing — major reviews are revisited when a vendor ships a significant update, and a “Last verified” date is displayed on the article.

Corrections Policy

When we get something wrong, we fix it and note it. Minor factual corrections (pricing, specs) are updated silently with an updated “Last verified” date; substantive corrections that change a conclusion or score are accompanied by a visible correction note at the bottom of the article describing what changed and why.