How ratings are collected, validated, and published on the RigorRank platform.
RigorRank is a professional intelligence platform for the digital asset industry. Its primary output — entity ratings — is produced by a structured, peer-governed voting mechanism designed to capture informed professional opinion while actively resisting manipulation.
This document explains, in full transparency, how that mechanism works: who is eligible to vote, how votes are structured and stored, how aggregates are computed, how cohorts are defined, and what safeguards exist against fraud, bias, and gaming. It is written for three audiences: subscribers who want to understand the methodology before participating; institutional stakeholders who need to assess the reliability of RigorRank scores for their own due diligence processes; and regulators or auditors who may review the platform’s data governance practices.
Transparency is not incidental to RigorRank’s value proposition — it is foundational to it. A rating that cannot be explained cannot be trusted. A score that cannot be audited cannot be relied upon.
Only verified, active subscribers may submit ratings on the RigorRank platform. This eligibility requirement is intentional and structural, not administrative.
The subscription requirement serves a specific integrity function: it introduces a real economic cost to voting. Free registration systems — including those that require only email verification — are susceptible to synthetic voter creation at negligible cost. By requiring a paid subscription, RigorRank ensures that any attempt to manufacture votes at scale requires a proportionate financial commitment, raising the cost of manipulation to a level that is impractical for most adversarial actors.
This is not a revenue consideration. It is a fraud-resistance design choice. Platforms that allow free voting are fundamentally more vulnerable to coordinated rating manipulation. RigorRank’s subscription gate is one of several layered defences against this risk.
A subscriber is eligible to vote when all of the following conditions are satisfied:
The following are ineligible to submit ratings under all circumstances:
RigorRank does not produce a single undifferentiated aggregate score. It produces cohort-specific scores that reflect the distinct perspectives of meaningfully different participant groups. This is one of the platform’s most important methodological distinctions from general-purpose review systems.
| Cohort | Eligible Subscribers | Typical Participants |
|---|---|---|
| Retail | retail_explorer and retail_pro subscribers | Individual investors, traders, developers, researchers, journalists |
| Institutional | institutional and institutional_pro subscribers | Asset managers, compliance officers, legal practitioners, risk professionals, regulated financial institutions |
Retail and institutional scores are computed separately and displayed separately on entity profiles. Institutional subscribers can see both cohort scores. Retail subscribers see the retail aggregate. This separation ensures that the professional standing of participants is reflected in the scores they produce and consume.
A subscriber’s cohort is determined at the moment of vote submission — not at the time of account creation, and not at the time the aggregate is computed. The cohort assigned to each vote is derived from a fresh database read of the subscriber’s current account tier at submission time. Session values are never used for cohort assignment.
The cohort value is stored permanently with the vote record as a snapshot. If a subscriber upgrades from a retail plan to an institutional plan after having submitted votes as a retail subscriber, their historical retail votes remain attributed to the retail cohort. Future votes submitted under the institutional plan are attributed to the institutional cohort.
Within each primary cohort, subscribers are further characterized by three professional profile dimensions collected during the onboarding survey. These vectors — professional role, organization type, and primary use of the platform — enable future analysis of sub-cohort opinion patterns without fragmenting the primary aggregate. They are stored as a snapshot with each vote record.
Each rating consists of six scored dimensions: one Overall score and five category-specific attribute scores. The five attribute labels are not generic — they are assigned specifically to the entity’s classification within the RigorRank taxonomy, ensuring that the attributes being evaluated are materially relevant to the entity type.
For example, a cryptocurrency exchange is evaluated on attributes including regulatory compliance, liquidity and execution quality, security and fund safety, transparency and disclosure, and customer support. A DeFi protocol is evaluated on smart contract security, liquidity depth, protocol governance, code transparency, and incident disclosure. These are not interchangeable.
All scores are submitted on a continuous scale of 0 to 10, in increments of 0.5. The Overall score is the subscriber’s holistic assessment. The five attribute scores allow granular evaluation of specific performance dimensions and enable sophisticated analysis of score composition.
RigorRank maintains a classification framework of 17 categories and over 70 subcategories covering the full spectrum of digital asset entities. Attribute sets are defined at the category or subcategory level, depending on the degree of meaningful variation within that category.
Attribute labels are resolved at render time and are never stored in the database — only the numeric scores are stored. This means attribute labels can be refined over time without invalidating historical vote records.
Ratings may be submitted through three authorized entry points. All three route through the same shared submission service, ensuring that validation logic, audit field capture, and transaction handling are identical regardless of origin.
| Entry Point | Description | Quota Impact |
|---|---|---|
| Profile Widget | Desktop rating form on the entity’s public profile page | View consumed |
| Mobile Widget | Mobile-optimized rating form within the Rankings tab on the profile page | View consumed |
| My Rankings Dashboard | Inline update form within the subscriber’s personal ratings dashboard | No view consumed |
Every rating submission requires the subscriber to confirm the following attestation before their vote is accepted:
This attestation is not merely a checkbox. It establishes the subscriber’s representation that their rating is independently formed and free from undisclosed conflicts of interest. The attestation text is versioned alongside the attribute set version.
Every submission undergoes server-side validation before any data is written. Client-side validation is present for user experience purposes only and is never relied upon for data integrity. Server-side checks confirm that all scores fall within the permitted range, that the entity type corresponds to a recognized category, that the submission source is one of the three authorized entry points, and that the subscriber’s account is active at submission time.
Each subscriber may hold exactly one active vote for any given entity at any point in time. The platform enforces this through a database constraint that permits only one vote record in active status per subscriber-entity pair.
When a subscriber submits a revised rating, the prior active vote is not deleted or overwritten. Instead, it is demoted to superseded status in an atomic database transaction. The new vote is simultaneously inserted as the active record. Both votes exist permanently in the database.
| Status | Meaning | In Aggregate |
|---|---|---|
| active | The subscriber’s current rating for this entity | Yes |
| superseded | A prior rating replaced by a subsequent submission | No |
| suspended | Administratively flagged — excluded pending review | No |
| pending | Survey vote awaiting payment confirmation | No |
Aggregate scores are computed as the arithmetic mean of all active votes within each cohort. For a given entity and cohort, the aggregate overall score is the average of the overall scores submitted by all active subscribers in that cohort who have rated the entity. Attribute aggregates are computed independently for each of the five attribute dimensions.
Scores are rounded to one decimal place for display. The number of votes contributing to each aggregate is displayed alongside the score, providing a transparency signal about the statistical weight of the aggregate.
Aggregate scores are only displayed when a minimum number of active votes has been received for the entity within a cohort. This threshold prevents individual ratings from being displayed as representative aggregates and reduces the risk that a small number of coordinated submissions can establish a misleading early-stage score.
Every aggregate score displayed on the platform is fully reproducible from the underlying vote records. An administrator or auditor with access to the vote data can reconstruct any displayed aggregate at any point in time by querying the active vote records for the relevant entity and cohort.
RigorRank employs multiple independent layers of fraud resistance. No single mechanism is relied upon in isolation. The effectiveness of the system derives from the combination of barriers, not from any individual control.
The requirement for a paid subscription introduces a real cost to voting. The cost of manufacturing a statistically significant number of synthetic votes — sufficient to materially move an aggregate — is directly proportional to the subscription cost multiplied by the number of votes required.
The attestation requirement creates a legal representation by the subscriber that their vote is independent and unconflicted. A subscriber who submits ratings at the direction of a third party with a financial interest in the outcome is making a false representation on a professional platform. This exposure functions as a deterrent independent of technical controls.
Every vote record captures the subscriber’s user identifier, the timestamp of submission, the IP address, the user agent string, the account tier at submission, the attribute set version, the submission source, and the cohort vectors from the subscriber’s onboarding survey. This combination enables post-hoc forensic analysis of voting patterns, including the detection of coordinated submissions from common IP ranges, rapid successive votes across many entities, and statistical anomalies in score distributions.
Cohort assignment is derived from the database at submission time, never from a client-provided value or a session variable. A subscriber cannot misrepresent their subscription tier by manipulating a client-side request.
The platform maintains the ability to administratively suspend individual vote records where fraud, manipulation, or breach of the attestation is detected or suspected. Suspended votes are excluded from aggregates but preserved in the audit record.
The five attribute labels associated with each entity category are subject to periodic review and refinement. When attribute labels change, the change is recorded by incrementing the attribute set version number. Each vote record stores the attribute set version in effect at the time of submission.
| Version | Effective Date | Notes |
|---|---|---|
| v1 | March 2026 | Initial attribute set. 17 categories, 70+ subcategories. |
Ratings published on RigorRank represent the aggregated opinions of verified subscribers. They do not constitute investment advice, legal advice, compliance recommendations, or any form of regulated financial analysis. RigorRank is a data and intelligence platform, not a regulated financial services provider.
RigorRank does not independently verify the accuracy, completeness, or reliability of any individual subscriber rating. Aggregate scores reflect the collective opinion of subscribers who have confirmed the attestation in Section 5.2. RigorRank makes no representation that any aggregate score accurately reflects the objective performance, conduct, regulatory standing, or financial condition of any rated entity.
To the maximum extent permitted by applicable law, RigorRank and its affiliates, officers, employees, and agents shall not be liable for any loss, damage, or harm arising from reliance on ratings, scores, or any other content published on the platform. Users who rely on RigorRank data for investment, business, regulatory, or legal decisions do so at their own risk and are encouraged to consult qualified professionals.
Subscribers are personally responsible for the accuracy of the ratings they submit and for their compliance with the attestation requirement. A subscriber who submits a rating that does not reflect their genuine professional assessment, or who submits a rating at the direction of a third party, is in breach of the platform’s terms of service and may be subject to account suspension and civil liability.
Vote records are retained indefinitely as part of the platform’s audit infrastructure. Personal data associated with vote records is processed in accordance with RigorRank’s Privacy Policy. Subscriber identifiers used in audit analysis are not disclosed to third parties, rated entities, or other subscribers except where required by applicable law or court order.
RigorRank reserves the right to update this Voting Methodology document at any time. Material changes will be communicated to subscribers via the registered email address on their account.
RigorRank welcomes feedback from subscribers on the voting methodology. Suggestions for attribute refinements, cohort definitions, or fraud resistance mechanisms may be submitted to methodology@rigorrank.com.
Suspected fraud or manipulation of the voting system should be reported to integrity@rigorrank.com.