How review verification approaches differ
There are several approaches to review integrity, each with genuine strengths and real limitations. We believe in proof-based verification. Here is an honest comparison of the approaches, including our own limitations.
We describe approaches, not specific companies. We aim for accuracy and fairness. If anything here is inaccurate, contact us and we will correct it.
Proof-based verification
VeriBureau
Every review requires cryptographic proof of a real transaction before publication. Scoring is computed from verified data, weighted by reviewer history, and sealed in an auditable chain.
STRENGTHS
Fake reviews are structurally impractical. All records are independently verifiable. No editorial discretion over content. Published methodology.
LIMITATIONS
Requires business participation. Lower review volume. Young protocol with limited dataset. Proof Tokens are generated by the business, creating a known trust boundary.
Moderation-based platforms
Platforms that use human and AI review after publication
Reviews are published first, then evaluated by human moderators, AI detection systems, and user reporting. Suspicious reviews are flagged or removed. Businesses can invite customers to leave reviews.
STRENGTHS
Large established user bases. Extensive review coverage. Significant investment in fraud detection. Some offer business-initiated verification via integrations. Brand recognition and consumer familiarity.
LIMITATIONS
Reviews do not require proof of transaction. Moderation is reactive and imperfect. Platforms may have commercial relationships with reviewed businesses. Detection cannot scale as fast as AI-generated content.
Algorithm-based platforms
Platforms that use automated recommendation algorithms
Reviews are filtered by proprietary algorithms that determine visibility based on undisclosed criteria. The algorithm decides which reviews are shown prominently and which are suppressed.
STRENGTHS
Attempts to surface high-quality content. Reduces obvious spam. Large user base and established market position.
LIMITATIONS
Algorithm criteria are not published. Businesses report correlation between advertising spend and review visibility. No independent verification of algorithm fairness. Opacity undermines trust.
Ecosystem-integrated reviews
Platforms where reviews are tied to an existing user ecosystem
Reviews are linked to user accounts within a broader platform ecosystem. Some verification of account activity exists, but proof of specific transactions is not required for reviews of most business types.
STRENGTHS
Massive scale. Location-based signals for physical businesses. Integration with maps and search. Automated spam detection at scale.
LIMITATIONS
Any account holder can review any business without proof of transaction. Automated moderation misses sophisticated fakes. Limited recourse for businesses facing coordinated attacks. No public audit trail.
Our position
We believe proof-before-publication is architecturally stronger than detection-after-publication. We also believe that transparency about limitations is more valuable than claims of perfection.
Every approach described above represents genuine effort to solve the review integrity problem. We respect the work of every team in this space. We chose a different architecture because we believe the evidence supports it — and we publish that evidence for scrutiny.
Examine the protocol yourself
Read the methodology. Query the audit chain. Verify a review. Then decide.