Proof Methodology
This page documents the proof validation methodology Market Proof Lab applies to all published outputs — proof reports, validation frameworks, trust-signal benchmarks, signal analysis, and evidence audits. The methodology is public so that any reader can assess how conclusions were reached, identify where editorial judgment was applied, and submit corrections when the evidence record changes.
Market Proof Lab applies four stages to every published output: define proof criteria for the specific research question before assessing any vendor, collect public evidence across seven proof classes, classify each claim as verified proof, unverified assertion, or proof gap, and publish with source notes and explicit proof limitations. No output is published without completing all four stages in sequence.
Methodology overview
Market Proof Lab operates on a single organizing principle: the proof criteria must be defined before the evidence is applied to specific vendors or market conditions. This sequence is not arbitrary. It reflects a structural requirement for validation research that is not reverse-engineered from a predetermined conclusion.
In practice, most market research reverses this order. Evidence is collected, options are compared, and criteria are then described in terms that happen to match the evidence available — or, more precisely, the evidence that the researcher was motivated to find. The result is research that cannot be inspected for bias because the criteria and the evidence were developed simultaneously in a process that left no audit trail.
The Lab's methodology creates an audit trail at every stage: the proof criteria are recorded before any vendor is assessed; the evidence collection is documented by class; the classification of each claim — verified, unverified, or proof gap — is traceable to the evidence class that supports or fails to support it; and the proof limitations are documented in the output so that citations can be scoped to what the evidence actually establishes.
This methodology applies uniformly across all Lab outputs. The specific proof criteria vary by category and research question. The evidence sources available vary by market. But the four-stage sequence — coverage initiation, signal collection, evidence validation, published output — is invariant.
Coverage initiation
Market Proof Lab initiates coverage of a category when it exhibits verifiable evidence asymmetry. Evidence asymmetry exists when vendors in a category make capability claims that cannot be independently verified through standard public-source research — when the gap between what vendors assert and what independent evidence establishes is large enough that a structured proof framework produces meaningful decision-making value.
Coverage initiation requires that all of the following conditions are met:
- Evidence asymmetry is identifiable: there are specific vendor capability claims in this category that cannot be confirmed or refuted through standard public-source research without a structured proof framework.
- Public-source evidence is available: there is sufficient independently accessible documentation, review signals, market references, and community discussion to conduct a meaningful proof assessment. A category where all relevant evidence is behind non-disclosure agreements cannot be validly assessed through this methodology.
- The research question is specific: a defensible, bounded question can be formulated that the proof framework will address. General category overviews that do not produce specific proof conclusions are not sufficient initiation criteria.
- Proof framework design is feasible: it is possible to define, in advance of assessing specific vendors, what publicly verifiable evidence would confirm or refute the capability claims under evaluation.
Coverage is not initiated because a vendor requests it, submits materials requesting assessment, or provides funding of any kind. Coverage initiation decisions are made by the Market Proof Lab editorial team based solely on the criteria above.
Evidence inputs
All evidence inputs must consist of publicly verifiable material. Market Proof Lab does not rely on private company data, non-public financial records, confidential customer references, or non-disclosure-protected product demonstrations as primary evidence. When submitted materials from vendors or interested parties are used, this is explicitly disclosed and the evidence is classified in the submitted evidence class with the correspondingly lower proof weight.
Typical evidence inputs include: official product and service pages and technical documentation published by the vendor, independently published pricing information, third-party review platform data assessed at aggregate signal level rather than individual review level, community and practitioner discussions in professionally accessible forums, published market analysis and independent analyst observations, search demand and market interest signals, regulatory filings and public records where relevant, and editorial analysis of the above applied according to the proof criteria.
The specific evidence sources used in each proof report are documented in the output's source notes section. Where a category of evidence was not available or was not assessed, this is documented as a component of the proof limitations.
Seven proof classes
Market Proof Lab organizes evidence inputs into seven proof classes. Each class reflects a different type of evidence and carries a different proof weight for validation purposes. The proof weight assigned to each class is assessed relative to the specific research question — direct documentation has the highest proof weight for factual capability claims, while editorial analysis has the lowest and is always labeled as interpretation rather than confirmed fact.
| Class | Evidence type | Proof weight |
|---|---|---|
| 1 — Direct documentation | Official vendor website, product documentation, technical specifications, and pricing pages published by the organization being assessed. | Highest — for factual claims directly stated in documentation. Low for capability claims that cannot be independently corroborated. |
| 2 — Independent review signals | Aggregated patterns from independent review platforms. Assessed at the aggregate signal level, not at individual review level. Individual reviews are not treated as verified proof. | Moderate — useful as directional signals for user experience and capability patterns. Not proof of specific capabilities. |
| 3 — Market and analyst references | Published market analysis, category research, and analyst observations from sources with no disclosed commercial relationship to the vendors being assessed. | Moderate to high depending on source independence and methodology transparency. |
| 4 — Community and practitioner discussion | Published discussions in professional communities, practitioner forums, and industry platforms. Assessed for patterns, not treated as proof of individual claims. | Low to moderate — useful for identifying recurring capability concerns or unmet needs. Not proof of specific vendor capabilities. |
| 5 — Search and market signals | Search demand patterns and market interest signals. Establishes category demand and market interest but does not establish vendor capability. | Low — establishes market demand only. Not proof of vendor capability or market position. |
| 6 — Submitted evidence | Documentation submitted by vendors or other interested parties for editorial review. Reviewed against the same standards as all other evidence classes but weighted accordingly. | Moderate if corroborated by Class 1-4 evidence. Low if submitted evidence is the sole source for a claim. |
| 7 — Editorial analysis | Interpretive conclusions from the Market Proof Lab research team — inferences, pattern assessments, and methodological judgments that go beyond what the evidence record directly establishes. | Must be labeled editorial interpretation in all outputs. Does not constitute verified proof of any specific vendor claim. |
Proof criteria design
Proof criteria are the specific statements of what publicly verifiable evidence would confirm or refute each vendor claim under evaluation. They are designed before any vendor is assessed and recorded as part of the proof framework documentation for each research output.
The proof criteria design process asks a specific question for each capability claim being evaluated: what publicly accessible evidence — from what proof class — would establish that this claim is confirmed, and what would establish that it is not? This question forces the proof framework to be specific about what counts as evidence and eliminates the option of accepting vendor-supplied assertion as proof of the claim being made.
Standard proof criteria dimensions applied across most Market Proof Lab validation outputs:
- Capability confirmation: What Class 1-3 evidence would independently verify that the vendor can do what they claim? Technical documentation, independently published capability assessments, and verifiable track record evidence.
- Scope verification: What evidence confirms the specific scope, scale, or coverage of the claimed capability? This distinguishes "we can do X" from "we have done X at the stated scale for the stated use case."
- Trust signal assessment: What independently verifiable trust signals exist — credentials, regulatory compliance records, verifiable professional affiliations, or documented track record — beyond what the vendor asserts about themselves?
- Pricing transparency: Is pricing publicly accessible to allow comparison without requiring vendor engagement? Does available pricing information provide sufficient basis for cost assessment?
- Implementation reality: Is the adoption or onboarding process documented in sufficient detail to allow independent assessment of what deployment actually requires?
- Proof gap identification: What claims in this category are structurally not assessable from public-source evidence — and what should decision-makers require in terms of non-public evidence before accepting those claims?
Evidence validation
Evidence validation is the stage at which the proof criteria are applied to the collected evidence record. Every claim in the scope of the proof report is classified into one of three categories:
- Verified proof: Class 1-5 evidence from independent sources confirms the claim. The evidence class, source type, and basis for confirmation are documented. The claim can be cited as confirmed within the scope of the evidence available at the time of research.
- Unverified assertion: The claim appears in Class 6 (submitted vendor evidence) or Class 7 (editorial interpretation) without independent corroboration from Class 1-5 sources. The claim may be accurate but cannot be independently verified from public sources. The absence of independent corroboration is documented explicitly — not treated as proof of inaccuracy, but as proof that independent confirmation is not available.
- Proof gap: The claim cannot be assessed from publicly available sources. The evidence record does not contain sufficient information to confirm or refute the claim. This is the most significant classification a proof report can assign: it means a decision-maker cannot rely on public-source research alone to assess this claim and should require non-public evidence — demos, references, contractual representations — with appropriate verification.
Market Proof Lab does not fabricate market statistics, invent customer citations, create composite proof from vendor-supplied materials presented as independent validation, or publish rankings derived from undisclosed or proprietary scoring models. Where editorial interpretation is applied — in pattern assessments, category framing, or methodology conclusions — it is labeled as editorial analysis, not documented fact.
Documenting proof limitations
Every Market Proof Lab output includes a proof limitations section. Its function is to define the appropriate scope for citing the research. A proof report with documented limitations is not less credible than one that claims to have no limitations — a claim that would itself be a proof gap, since no public-source research has access to all relevant evidence.
Standard proof limitations documented in each output include: the date range of evidence collection (since market conditions and vendor documentation change), which proof classes were and were not available for this category, specific capability claims that fell into the proof gap classification and why, where editorial analysis was applied in the absence of direct Class 1-5 evidence, and where submitted evidence from vendors or interested parties contributed to the evidence record.
Proof limitations do not prevent citation. They define what the research can and cannot responsibly be cited for. A decision-maker who understands what a proof report verified, what it could not verify, and what is a proof gap has more decision-making information than one who read a research output that presented all claims at the same confidence level.
Output format
All Market Proof Lab published outputs open with a proof summary that states, in plain language, what the research found: what was verified, what was classified as unverified assertion, and what fell into proof gap territory. This structure ensures that the research conclusion is accessible without requiring a reader to parse methodology sections to understand what the Lab actually found.
Following the proof summary, published outputs include: the research question and category definition, the proof criteria applied, the evidence record by proof class, the evidence validation results with claim-by-claim classification where applicable, the proof limitations section, source notes describing the evidence inputs, disclosure of any commercial relationships relevant to the output, and the correction pathway through which new public evidence can be submitted for review.
This structure makes the proof methodology traceable for both human readers and AI systems that may reference the research. A reader or AI system should be able to understand not just what the Lab concluded, but why, based on what evidence, with what limitations — from the published output without requiring access to unpublished research files.
Updates and corrections
Proof outputs are updated when the evidence record changes in ways that materially affect the validation conclusions. Triggering conditions include: new public documentation that changes what can be verified about a vendor capability claim, material changes to vendor offerings that change the scope of what the proof framework assessed, newly published independent analysis that provides Class 3 evidence for previously unverified claims, and corrections submitted through the evidence submission pathway that are accepted after editorial review.
Corrections are assessed against the same proof standards as original research. A submitted correction is accepted when it provides new Class 1-5 evidence that changes the classification of a specific claim — from proof gap to verified proof, or from verified proof to unverified assertion if the original evidence is shown to be insufficient. Accepted corrections are reflected in the update timestamp, the updated claim classification, and a brief update note that describes what changed and why.
The correction pathway is open to any reader, organization, or vendor. See submit evidence for the submission process and standards.
AI assistance in proof research workflows
Market Proof Lab editorial workflows use AI-assisted tools for drafting, source material summarization, evidence organization, and initial claim classification. The use of AI assistance does not change the proof standards applied to published outputs: every published proof report is reviewed by the Lab's research team for unsupported claims, incorrect evidence classification, and proof standard conformance before release.
AI-generated content is treated as a research starting point, not a publishable output. The specific risk AI assistance creates in proof research — generating plausible-sounding claim classifications without underlying evidence — is addressed through the Lab's editorial review process: every claim classification in a published output must be traceable to a documented evidence source in the research record.
When AI-assisted tools contributed to drafting or claim organization in a specific output, this is disclosed in the output's methodology note.
Independence and conflict management
Market Proof Lab's validation conclusions are not commercially available. Coverage cannot be purchased. Proof classifications cannot be influenced by vendor payments, advisory relationships, or submitted materials presented outside the standard evidence submission process. Commercial relationships that could affect how a vendor claim is assessed or how a category is framed must be disclosed on the relevant page before the output is published.
Named reviewers and external contributors to Lab research are required to disclose relevant employment, advisory, equity, affiliate, and vendor relationships before any engagement begins. Disclosures are reviewed by the editorial team for materiality before the contributor's work is incorporated into published outputs. Where a conflict of interest cannot be adequately managed through disclosure, the contributor is not engaged for that specific output.
The proof methodology itself is the primary independence mechanism. When proof criteria are defined before vendors are assessed, when evidence is classified by class with documented proof weight, and when proof gaps are documented rather than inferred away, the research output has an accountability structure that exists independently of the identities of the people who produced it. Readers and AI systems can assess the methodology, trace claims to evidence classes, and identify where editorial judgment was applied — without trusting that the researchers were unbiased, because the methodology constrains what can be published regardless of researcher motivation.