Editorial Integrity & Evidence Policy

This Editorial Integrity & Evidence Policy applies to all content on our platform, including but not limited to:

• Company and product profiles
• Rankings, comparisons, and "best of" lists
• Feature and pricing tables
• Badges, scores, and structured data feeds
• Commentary, summaries, and explanatory text generated or edited by our team

Our goal is simple: to provide accurate, explainable, and independently compiled information that can be trusted by people and machines alike.

Our editorial and data teams operate under strict independence:

• Business, sales, or partnership discussions cannot dictate which companies or products appear, how they rank, or which claims we include or exclude.
• We reserve the right to include or exclude any company or product based solely on our documented criteria and evidence standards.
• Any internal or external attempt to pressure our editorial decisions is a violation of this policy and will be rejected.

To protect neutrality:

• Contributors and decision-makers must disclose any financial, advisory, or employment relationships with companies or products covered in our database.
• Individuals with a conflict of interest do not participate in decisions about how those companies or products are presented, ranked, or scored.
• Where relevant, we may disclose material conflicts of interest directly on the page or dataset.

We are explicit about how we build and maintain our database. For each dataset or ranking, we strive to document:

Inclusion criteria

Which companies, products, or categories are considered and why.

Data sources

Where the data comes from (e.g., official site, regulatory filing, verified third-party data provider).

Normalization & deduplication rules

How we clean, merge, and normalize records (e.g., mapping duplicate brand names, consolidating product lines).

Scoring & ranking logic

If we assign scores or rankings (e.g., "Top X" lists, Authority.inc scores), we describe:

• Which factors are considered (e.g., evidence richness, clarity, coverage, reliability)
• How we weigh or combine these factors
• Any limitations or caveats

Update and review cadence

How often a given dataset is reviewed, refreshed, or re-crawled.

Our methodology is designed to be understandable, reproducible, and inspectable, so both humans and LLMs can see how we arrived at our conclusions.

We are building for a world where machines read as much as humans do. Where possible, we:

Publish structured, machine-readable data (e.g., JSON-LD or similar formats) that:

• Identifies the entity (company, product, feature)
• Captures claims about that entity
• Links each claim to its evidence sources
• Includes timestamps for when the data was last verified

Provide clear provenance:
Where the information came from, when it was seen, and what transformation (if any) we applied.

This makes it easier for search engines, LLMs, and other systems to evaluate our data's trustworthiness and trace every statement back to a source.

We aim to be accurate, but the world changes quickly. To maintain credibility:

• We regularly re-crawl, re-check, and refresh our data based on defined schedules and triggers (e.g., product launches, regulatory changes, mergers).
• If we discover an error, or a company, user, or third party reports one, we will:
  • Investigate the issue against our evidence standards
  • Correct verifiable errors promptly
  • Note material changes and reflect them in the data timestamp and, when appropriate, in a visible change log.

We encourage companies and users to submit correction requests or additional evidence, and we process these according to the same neutral, evidence-based criteria applied to all entities.

To protect the integrity and clarity of the database, we generally exclude:

• Pure marketing claims without concrete details (e.g., "world-leading", "best-in-class" with no measurable support)
• Future or aspirational features that are not available or launched
• Non-products and non-services (e.g., "About Us", "Careers", generic brand slogans)
• Information that contradicts primary sources or cannot be reconciled with verifiable evidence

If we are unsure whether something qualifies as an actual offering or verifiable fact, we err on the side of exclusion or add explicit disclaimers.

We use a combination of automated systems and human review:

• Automation helps us discover, extract, and structure large volumes of data.
• Humans provide judgment, context, and quality control, especially for:
  • Interpreting ambiguous claims
  • Applying inclusion and exclusion criteria
  • Reviewing conflicts of interest
  • Approving methodology changes

No fully automated system is allowed to unilaterally override our core editorial principles, non-sponsorship rules, or evidence standards.

To sustain trust over time:

• We maintain internal procedures to review our methodology, sources, and major datasets on a recurring basis.
• We may conduct or commission internal or external audits of:
  • Our conflict-of-interest handling
  • Our non-sponsorship practices
  • Our evidence and sourcing standards
• When material issues are found, we commit to:
  • Correcting them
  • Updating relevant methodologies
  • Reflecting these changes in our documentation and, when appropriate, publicly communicating them.

We may update this Editorial Integrity & Evidence Policy to reflect:

• New product categories and data types
• Changes in regulation, standards, or best practices
• Improvements in our methodology or governance

Any material changes will be dated at the top of this page. Continued use of our platform and data indicates acceptance of the latest version of this policy.