Why methodology is public
Research that cannot be inspected cannot be cited. That is the founding principle of this document. The Brightfield Research methodology is published in full because concealing the research process from the people who use the research is not a methodological position — it is a commercial one. It protects the research producer from scrutiny at the expense of the reader's ability to evaluate what they are reading.
The goal of public methodology is not to invite challenges for their own sake. The goal is to make criteria visible and conclusions defensible. A published methodology allows a reader to determine whether the criteria were appropriate for the category question being asked, whether the evidence collected was sufficient to support the conclusions reached, and whether the limitations documented are complete enough to understand where the research has boundaries.
If a reader finds that the criteria are inappropriate, that the evidence was insufficient, or that significant limitations were omitted, they can submit that finding through the evidence pathway and it will be reviewed. This is how the methodology improves over time — not through internal revision, but through structured external challenge.
This document is maintained by the Director of Research and reviewed on a documented schedule. Changes to the methodology are recorded with a date and a summary of what changed and why. The version of the methodology in effect when any published research output was produced is retained for reference.
Coverage initiation criteria
Brightfield Research does not cover every available category. Coverage is initiated only when four criteria are satisfied, and only after the research question for the category has been written and reviewed by the editorial desk.
Demonstrable market demand. There must be evidence that professional decision-makers are actively seeking evaluation guidance in the category — search signals, community discussion, analyst references, or documented procurement activity that indicates a real evaluation problem exists.
Available public evidence. There must be sufficient publicly available evidence to support structured analysis. Brightfield does not initiate coverage in categories where evaluation would require access to proprietary data, confidential assessments, or evidence that cannot be documented in a public source note.
Meaningful evaluation complexity. The category must present genuine evaluation difficulty — criteria that are non-obvious, tradeoffs that require structured analysis, or capability claims that require a documented framework to assess. Categories where the evaluation question is trivial do not benefit from the level of research infrastructure Brightfield applies.
Defensible research question. The coverage initiative must be grounded in a clearly stated, defensible research question that can be answered, at least in part, through the evidence available. Coverage initiated without a clear question produces outputs that are descriptive rather than evaluative — category overviews rather than research outputs.
When these four criteria are met, coverage initiation is documented with a stated research question, a preliminary evidence landscape review, and a criteria design brief. The first output published for a new coverage category is typically a research primer or category definition that establishes the evaluation framework before comparative analysis begins.
Research inputs
Brightfield Research collects research inputs from multiple source types, each of which is assigned to an evidence class (see the evidence classes section below). The following source types are used in Brightfield research:
- Public websites and product pages — the organization's own documentation of its products, services, scope, pricing model, and capability claims.
- Documentation and specification pages — technical documentation, API references, help center content, and published specification documents.
- Pricing pages and published pricing structures — any publicly available pricing information, including published tiers, pricing ranges, or pricing model descriptions.
- Review profiles on independent aggregators — documented review signals from publicly available aggregated review platforms. Individual reviews are not cited as primary evidence; aggregate patterns and documented signals are referenced.
- Community discussions and practitioner content — public forum discussions, professional community posts, and practitioner-authored content that documents first-hand experience with a category or subject.
- Submitted materials — documentation, corrections, and evidence submitted through the Brightfield evidence submission pathway by organizations or interested parties.
- Editorial review — analysis, interpretation, and judgment applied by the Brightfield editorial team. This input class is always labeled as editorial and distinguished from documented evidence in published outputs.
Seven evidence classes
All research inputs are assigned to one of seven evidence classes. The class assignment determines how the input is treated, labeled, and interpreted in the published output. Higher-class evidence carries more interpretive weight for claims within its domain. No class of evidence is fabricated or inferred without a documented source.
| Class | Description | Typical sources | Interpretive weight |
|---|---|---|---|
| Class 1 Direct documentation |
Documentation produced directly by the organization under evaluation. Covers the organization's own claims about its attributes, capabilities, scope, pricing, and history. | Official website, product documentation, published pricing pages, official announcements, company-authored content. | High — for factual claims about the organization's own attributes. Does not independently verify claims; documents what the organization states. |
| Class 2 Independent review signals |
Aggregated signals from independent third-party review platforms. Covers documented patterns in publicly available review data — volume, recency, response behavior, and thematic signals. | G2, Capterra, Trustpilot, Google Business Profile, and equivalent aggregated review platforms. | Moderate — as a signal of market perception and documented pattern. Not high-confidence for specific factual claims. Individual reviews are not cited. |
| Class 3 Market and analyst references |
References from independent market research, analyst publications, trade press, and academic sources that are not commercially affiliated with the subject of evaluation. | Published analyst reports, trade and industry press, academic research, independent sector studies. | Moderate to high — depending on independence, methodology visibility, and source credibility. Source is cited and independence is documented. |
| Class 4 Community and practitioner discussion |
Public discussion from professional communities, forums, and practitioner-authored content reflecting first-hand experience with a category or subject. | Reddit, Hacker News, professional forums, practitioner blogs, LinkedIn practitioner content. | Low to moderate — for individual claims. More useful as a signal of practitioner concerns, common implementation challenges, and category-level patterns than for specific factual claims. |
| Class 5 Search and market signals |
Search volume data, market interest signals, and publicly available competitive intelligence that documents the scale and nature of market demand in a category. | Public keyword tools, market interest trackers, search visibility indicators, publicly available traffic estimates. | Low — as a direct evidence source. Used for coverage initiation criteria and category sizing, not for evaluative conclusions about specific subjects. |
| Class 6 Submitted evidence |
Documentation, corrections, and evidence submitted through the Brightfield evidence submission pathway by organizations, practitioners, or interested parties. | Evidence submitted via the evidence submission form or editorial contact. May include documentation, public URLs, correction notes, or additional context. | Variable — assessed against other evidence classes. Submitted evidence is reviewed for quality, not accepted uncritically. Source and interest of submitting party are documented. |
| Class 7 Editorial analysis |
Analysis, interpretation, synthesis, and judgment applied by the Brightfield editorial team. Always labeled as editorial in published outputs and distinguished from documented evidence from other classes. | Editorial desk analysis, research synthesis, interpretive judgments, category framework design. | Supplementary — editorial analysis is used to synthesize documented evidence and make interpretive judgments explicit. It does not substitute for documented evidence in classes 1 through 6. |
Criteria design
Evaluation criteria are designed after the evidence collection phase has produced an initial picture of the category and before any individual subject is assessed. This sequencing is deliberate and non-negotiable: criteria designed after subjects are reviewed are not criteria. They are post-hoc rationalizations for conclusions already reached.
The standard Brightfield criteria set covers eight dimensions, adjusted for each category based on what the research question requires:
Does the subject actually operate in the category being researched? Are its scope, target market, and documented capabilities consistent with the category definition?
How clearly does the subject document what it does, who it serves, and what it does not cover? Scope clarity affects whether a decision-maker can reliably determine fit without additional investigation.
How well documented are the subject's capability claims? Are they supported by independent evidence, case documentation, or verifiable references, or are they promotional assertions without supporting evidence?
Is pricing publicly documented, or does evaluation require a sales engagement to obtain basic pricing information? Pricing opacity is documented as a category risk factor.
What does implementation require of the organization selecting this subject? Time, resources, dependencies, onboarding complexity, and switching costs are assessed where public evidence is available.
What support model is documented? How are issues escalated and resolved? Are accountability signals — response to documented problems, correction behavior — visible in public evidence?
What publicly documentable trust signals are present? These include documented longevity, verifiable credentials, named leadership, transparent ownership, and accountability for public claims.
What are the documented risk factors specific to this category? These include factors that affect the reliability of evaluation, the stability of the subject over time, and the consequences of a poor selection decision.
Evidence assessment
Evidence assessment is the stage at which the editorial team systematically applies the defined criteria to the collected evidence for each subject in scope. The assessment distinguishes three categories of output:
Documented facts are statements that can be supported by a specific, identifiable source assigned to evidence classes 1 through 6. Documented facts are stated directly in published outputs with their source class labeled.
Editorial interpretation covers analysis, synthesis, and judgment applied by the editorial team to documented evidence. Editorial interpretation is labeled as such in published outputs and is not presented as documented fact. Where editorial interpretation could reasonably be challenged with different evidence, the nature of the interpretive judgment is noted.
Evidence gaps are category dimensions or specific claims for which insufficient public evidence was available to support a conclusion. Evidence gaps are documented explicitly in the limitations section of each output. A gap is not treated as a negative indicator unless there is independent evidence that the gap is material to evaluation.
The assessment process prohibits the promotion of promotional language to documented fact status. A capability claim on a vendor's product page is Class 1 evidence — it documents what the vendor claims, not what has been independently verified. It is published as such, not as a confirmed capability.
Limitations documentation
Every Brightfield research output includes a dedicated limitations section. This section is not a disclaimer appended to protect the publisher. It is a substantive research component that makes the output more useful by being explicit about where it can be relied upon and where it cannot.
The limitations section of every output addresses the following:
- What was not reviewed in this research output and why.
- What evidence was searched for but was unavailable in public sources.
- Where the research has an expiry risk — evidence that is likely to become outdated, and how quickly.
- Where editorial judgment was applied in the absence of sufficient documented evidence, and what evidence would improve the conclusion.
- What categories of decision-maker or use case are outside the scope of this research output.
- What changes in the market or category would require the output to be updated before it could be reliably cited.
Limitations are documented at the level of specificity that makes them actionable. A limitation that states only "this research may not be complete" is not a limitation. A limitation that states "pricing transparency was not assessable for three subjects in scope because no public pricing was available; this limits comparison on the pricing transparency criterion" is a limitation.
Output format
Every Brightfield research output follows a standard format that includes all of the following components. Outputs that cannot include all components because of evidence gaps or scope constraints are noted as such in the limitations section.
A front-loaded answer to the research question that does not require reading the full output to understand the key conclusion. Qualified where the conclusion is not definitive.
A documented definition of what the category covers, who it serves, what it does not include, and how it is bounded for the purposes of this research output.
A documented list of the evaluation criteria used, with rationale for each criterion's inclusion and its weighting relative to other criteria.
The documented application of criteria to collected evidence, with source class labels, documented facts distinguished from editorial interpretation, and evidence gaps identified.
A substantive documentation of what was not reviewed, what evidence was unavailable, where editorial judgment was applied, and where the output has an expiry risk.
Documentation of the sources consulted, their evidence class, and a description of what each source contributed to the output. Sources are described with sufficient specificity to allow independent verification.
Documentation of any commercial relationships, submitted evidence arrangements, or material interests that could affect the research output. Disclosed on the relevant page, not in a general policy document.
A documented link to the evidence submission pathway, with a statement of what types of submissions are reviewed and what the editorial review process produces.
Updates and corrections policy
Brightfield Research outputs are not static publications. They are dated documents with a documented review schedule and a structured correction process.
Outputs are updated when any of the following conditions apply: new public evidence materially changes a documented conclusion; a correction submission provides documented evidence that a published statement is factually incorrect; a category change makes prior criteria or conclusions incomplete; or the passage of time makes the evidence base sufficiently outdated to affect reliability.
Updates are recorded on the published page with a date and a summary of what changed and why. Material changes are distinguished from minor corrections. The original publication date is retained alongside the update date.
Corrections submitted through the evidence pathway are reviewed by the managing editor and the director of research. The review process assesses the quality of the submitted evidence against the seven-class framework and determines whether the evidence warrants a correction, an update, an added limitation note, or no change. The editorial decision is documented and the submitting party is notified of the outcome.
Brightfield does not remove published research outputs except in cases where the output cannot be corrected without a full rewrite, or where the subject of the research requests removal with documented evidence that continued publication causes material harm that cannot be addressed through correction. In those cases, the removal is documented and a notice is published at the original URL explaining why the output was removed.
AI assistance disclosure
The Brightfield Research editorial workflow may use AI-assisted drafting, research summarization, structural assistance, and content organization tools at various stages of the production process. This is disclosed here and, where relevant, on individual published pages.
AI assistance is used as an editorial tool, not as a replacement for editorial judgment. All published pages are reviewed by a named editor before release. AI-generated drafts are reviewed for accuracy, for unsupported claims, for fabricated statistics or citations, and for promotional language that should not appear in published research. No AI-generated content is published without human editorial review and sign-off.
AI tools are not used to generate evidence class assignments, to fabricate sources, or to produce conclusions that are not supported by documented inputs. The use of AI in the editorial workflow does not reduce the evidence requirements, limitations documentation requirements, or source transparency requirements applied to every output.
Where AI assistance is material to the production of a specific published output — meaning it contributed substantively to the content rather than serving as a drafting scaffold — this is noted on the relevant page.
Independence and conflict management
Brightfield Research does not accept payment for rankings, placement, or editorial conclusions. This is an absolute constraint, not a policy subject to case-by-case exceptions. No commercial arrangement can change the editorial conclusion of a published output.
The following commercial relationships are permitted and must be disclosed on the relevant published page:
- Sponsorship of a research output that does not influence the editorial conclusion, criteria, or evidence assessment of that output.
- Submitted evidence arrangements where an organization has paid to have its submitted materials reviewed by the editorial desk on a documented schedule.
- Advertising arrangements that are clearly separated from research content and do not affect editorial coverage decisions.
All reviewer conflict disclosures are documented before a reviewer's profile is activated. Reviewers with material commercial interests in a category under evaluation are excluded from the review process for that category. The editorial desk maintains a documented record of all conflict disclosures.
If a reader believes that an undisclosed commercial relationship has affected a published research output, they are encouraged to submit that concern through the evidence pathway with documented evidence. All such submissions are reviewed by the director of research and the publisher and responded to on the record.