The weak question
Does it look similar?
Broad resemblance is easily copied and often shared by genuine later editions, replicas and altered examples. It may justify further investigation, but rarely carries a conclusion by itself.
Authentication Methodologies Documentation
Comparative authentication asks whether the significant characteristics of a submitted object fit the documented range of genuine production better than the available alternative explanations. It is not a test of general resemblance. Counterfeits, reproductions, later issues, replacement parts and altered genuine objects can all look convincing. A defensible comparison therefore identifies reliable references, selects like-for-like criteria, records similarities and contradictions, distinguishes normal variation from meaningful anomaly, and explains how strongly each finding affects the claim.
The documentation is part of the method. A statement such as "compared with known originals and found authentic" preserves a verdict but loses the intellectual route behind it. A useful record shows which originals were used, why they are trusted, what was examined, what differed, how variation was handled, what remained inaccessible, and why the conclusion does not exceed the evidence.
Collector scenario
A convincing figure with the wrong mould geometry
A collector compares a painted figure with one online image and sees the same colours, copyright wording and general proportions. The item appears authentic. A broader reference group, however, shows that genuine paint placement varies widely while the mould geometry, underside marks, joint construction and scale remain stable. The questioned figure matches the paint scheme but falls outside the genuine mould family.
The lesson is not that visual comparison failed. The comparison was too narrow, the comparator was weak, and an easily copied feature was given more weight than production evidence. A stronger record would preserve the reference group, distinguish variable paint from stable geometry, and explain why one contradiction outweighs several superficial matches.
Orientation
Comparison is structured correspondence, not resemblance
The object under examination may be called the questioned, submitted, subject or candidate object. The material against which it is assessed may be called references, comparators, exemplars, controls or a reference population. The language matters less than the discipline applied to the relationship between them.
The useful question
Does the object display the combination of characteristics expected from the claimed maker, date, issue, factory and production process?
Any difference must then be tested against genuine variation, wear, storage, restoration, replacement, alteration and false manufacture.
Why a population is usually stronger than a match
Genuine collectibles are rarely perfectly uniform. Tool wear, hand finishing, multiple factories, revised dies or moulds, regional production, printing registration, material shrinkage, signing circumstances, ageing and conservation can all produce legitimate variation. Comparison with one convenient example may therefore turn difference into false suspicion or similarity into false confidence.
A reference population allows the examiner to identify stable characteristics, variable characteristics and measured ranges. The aim is not to make the submitted object identical to a chosen specimen. It is to decide whether the object falls within a defensible genuine range and whether any anomaly is better explained by another origin.
Reference discipline
The comparator hierarchy
The quality of a conclusion cannot exceed the quality of the references supporting it. Comparator selection should be explicit, ranked and open to challenge.
Tier 1
Primary, independently verified references
Factory-retained specimens, museum objects with secure accession histories, artist or manufacturer archive examples, witnessed signatures, controlled excavations and objects documented from creation usually provide the strongest foundations.
Tier 2
Well-established authenticated objects
Examples repeatedly studied by recognised specialists, supported by strong provenance, or published in authoritative references can define reliable characteristics and genuine variation.
Tier 3
Market-certified or historically attributed objects
Reputable third-party certification, major auction history or established catalogue attribution can be useful, but prior opinion should be examined rather than automatically converted into primary evidence.
Tier 4
Claimed genuine examples
Dealer listings, collector photographs, forums, uncited databases and unidentified auction images may help generate hypotheses. They are weak foundations for a definitive conclusion unless their status can be independently established.
Tier 5
Other questioned objects
Comparing one uncertain object with another can create circular authentication. Similarity between them may show a shared source, a shared reproduction method or a copied error rather than genuineness.
Like-for-like is not optional
A reference should match the claim as closely as the authentication question requires. Relevant variables may include maker, factory, date, edition, country, production method, mould, die, material, size, format, signing period, distribution region, packaging version and state of completion.
A genuine 1980s reissue may be unsuitable as the sole comparator for a 1960s first issue. A late-life autograph may poorly represent an early-career signature. A coin from another die pairing may carry different legitimate markers. The report should explain not only what was used, but why it is genuinely comparable.
Methods
Principal comparative methodologies
Most serious assessments combine several methods. Each method answers a different question, carries different limitations and requires its own documentation.
Direct side-by-side comparison
Comparative question
How does the submitted object differ from a reliable genuine example when examined under equivalent conditions?
Best used for
Useful for dimensions, proportions, layout, typography, colour, texture, marks, construction, component placement and packaging details.
Strength
Subtle differences are easier to recognise when the objects or images are considered together rather than recalled from memory.
Limitation
One example may represent an unusual batch, a repaired object or only one legitimate version. It cannot define the entire genuine range by itself.
Document
- The exact comparator and why it is regarded as genuine.
- Whether it is the same date, factory, issue, version and production type.
- Whether comparison was physical, photographic or database-based.
- Any differences in lighting, scale, angle, holder or access.
Diagnostic-feature comparison
Comparative question
Are specific genuine or counterfeit diagnostics present, absent or altered?
Best used for
Useful for die markers, mould seams, printing patterns, letterforms, tooling sequences, security features, factory codes and recurring counterfeit defects.
Strength
A well-established diagnostic may carry far more weight than broad resemblance, especially when it is linked to a particular production process or counterfeit family.
Limitation
A feature may be common rather than necessary, supportive rather than exclusive, or visible on only part of the genuine population.
Document
- The expected genuine form and the observed form.
- Whether the feature is necessary, common, supportive or exclusionary.
- The reference source establishing its significance.
- The weight assigned and any alternative explanation.
Pattern and population comparison
Comparative question
Does the object fall within the documented range of genuine variation?
Best used for
Useful for measured ranges, signature habits, serial sequences, print registration, component relationships, material combinations and recurring production variation.
Strength
A reference population reveals what remains stable and what legitimately varies. It reduces the risk of mistaking difference from one specimen for evidence of falsity.
Limitation
The population may be small, biased, poorly documented or already contaminated by misidentified examples.
Document
- The number, date range and status of references examined.
- Which characteristics were constant and which varied.
- Measured ranges or recurring patterns where available.
- Where the questioned object sits within or outside the range.
Typological comparison
Comparative question
Is the object compatible with the precise type, issue, edition or production family it claims to represent?
Best used for
Useful for coins, books, toys, ceramics, watches, prints, cards, stamps, militaria and other categories with documented variants or issue structures.
Strength
It tests chronological and categorical possibility before finer judgements are attempted.
Limitation
A broad label such as 'vintage figure' or 'first edition' may conceal the factory, mould, region, state or packaging revision on which authentication depends.
Document
- The claim at the narrowest relevant level.
- The type definition and source used.
- Features expected for that precise issue or family.
- Any known exceptions, transitional forms or unresolved variants.
Production-method comparison
Comparative question
Does the physical evidence correspond with how genuine examples were made?
Best used for
Useful for striking, casting, moulding, printing, carving, machining, stitching, plating, painting, glazing, binding and assembly.
Strength
Manufacture leaves evidence that may be harder to imitate consistently than surface appearance.
Limitation
Wear, restoration and later components can obscure original manufacture, while sophisticated reproductions may copy some production traces.
Document
- Observed manufacturing evidence.
- The historically expected process.
- The comparison between them.
- Why the observation supports, weakens or does not affect the claim.
Dimensional and metrological comparison
Comparative question
Do measurements fall within a relevant and defensible expected range?
Best used for
Useful for height, width, diameter, thickness, weight, density, spacing, alignment, thread count, card stock and component tolerances.
Strength
Measurements create repeatable evidence and can expose errors that general visual inspection misses.
Limitation
Measurements are not self-interpreting. Wear, shrinkage, trimming, deformation, moisture, restoration, instrument error or undocumented variants may explain deviation.
Document
- Instrument, unit and exact measurement location.
- Repeated readings and calibration status where relevant.
- Expected range, source and tolerance.
- Material or environmental factors affecting interpretation.
Material comparison
Comparative question
Are the object's materials compatible with genuine production and the claimed date?
Best used for
May involve visual examination, microscopy, ultraviolet examination, spectroscopy, X-ray fluorescence, fibre, paper, ink, adhesive, alloy, pigment or polymer analysis.
Strength
An anachronistic or incompatible material can be highly significant and may direct attention to later repair, replacement or false manufacture.
Limitation
Period-compatible material does not prove period manufacture. Old stock and salvaged components can be reused, and testing may apply only to the sampled area.
Document
- Sampling location and whether the method was destructive.
- Equipment, method, raw result and reference set.
- Detection limits, uncertainty and interpretation.
- Whether the result concerns the whole object or one component.
Image-overlay and proportional comparison
Comparative question
Do aligned images reveal correspondence or divergence in geometry, placement or registration?
Best used for
Useful for outlines, die details, mould geometry, signatures, labels, design elements, printing registration and component spacing.
Strength
It can make small positional differences visible and preserve an annotated comparison for later review.
Limitation
Perspective, focal length, camera angle, distortion, scale, cropping and digital processing can create false differences or false matches.
Document
- Original images and adjusted copies.
- Scale references and alignment points.
- Transformations, distortion correction and overlay method.
- Annotations explaining what the comparison does and does not establish.
Signature and handwriting comparison
Comparative question
Does the questioned writing fit reliable contemporaneous exemplars and the signer's natural variation?
Best used for
Useful for letter formation, proportions, spacing, slant, pressure, rhythm, speed, line quality, connecting strokes and placement.
Strength
A group of secure exemplars can reveal recurring habits while accommodating genuine variation.
Limitation
Brief, smudged or slowly written signatures may be weakly diagnostic. Exact duplication may indicate tracing, printing or automated production rather than genuineness.
Document
- Number, date range and source of exemplars.
- Whether originals or reproductions were used.
- Relevant natural variation and unexplained divergences.
- Whether the conclusion concerns the signature, the object or both.
Provenance and documentary comparison
Comparative question
Do ownership records and accompanying documents withstand independent comparison?
Best used for
Useful for catalogues, invoices, stock books, inventories, labels, correspondence, archival photographs, certificates and registration records.
Strength
It can connect an object to a documented place, person, sale or collection and expose chronological or administrative impossibility.
Limitation
A copied statement is not verified provenance, and an authentic document does not automatically prove that it belongs to the submitted object.
Document
- Each provenance link as verified, supported, plausible, reported, inferred, contradicted or unknown.
- The authenticity of the document itself.
- The evidence connecting the document to the exact object.
- The authority and competence of any issuer or prior authenticator.
Catalogue, registry and census comparison
Comparative question
Does an authoritative reference describe this exact object or only a broadly similar type?
Best used for
Useful for catalogue numbers, known examples, dimensions, inscriptions, images, edition records, provenance and exhibition histories.
Strength
A specialist reference can place an object within a documented corpus and reveal prior scholarship or known variants.
Limitation
Catalogues can contain errors, duplicates, omissions and outdated attributions. Being 'in the book' is not self-proving.
Document
- Full title, author or issuing body, edition and date.
- Volume, page and catalogue number.
- Whether the entry and image were inspected directly.
- Later supplements, corrections or changed opinions.
Negative comparison
Comparative question
What expected feature is absent, or what incompatible feature is present?
Best used for
Useful for missing marks, wrong edges, anachronistic materials, impossible serials, incorrect codes, misplaced design elements and incompatible construction.
Strength
A mandatory absent feature or a decisive contradiction may outweigh many weak similarities.
Limitation
Absence is persuasive only when the feature is genuinely mandatory and could not reasonably have been worn away, trimmed, obscured, repainted or omitted in a legitimate variant.
Document
- Why the feature is expected.
- How consistently it appears across reliable genuine examples.
- Whether it could be inaccessible, removed or obscured.
- Whether the finding is exclusionary, concerning or unresolved.
Judgement
Classify findings before weighting them
A disciplined report does not treat every observation as equally useful. It first records whether a finding supports, conflicts with, fails to discriminate, or remains unresolved.
Positive
Supports the claim
- Expected genuine diagnostic present.
- Construction consistent with documented production.
- Measurements within a relevant known range.
- A provenance link independently verified.
Negative
Conflicts with the claim
- Anachronistic material or impossible process.
- Known counterfeit marker or incorrect factory code.
- Serial information incompatible with the proposed issue.
- A tightly controlled feature outside the genuine range.
Neutral
Does not discriminate
- General stylistic resemblance.
- Common material used by both genuine and false examples.
- Ageing that may be natural or artificial.
- A broad measurement shared across multiple types.
Unresolved
Requires further work
- Unreadable or inaccessible mark.
- Unidentified adhesive or component.
- Unexplained dimensional variance.
- Comparator status or provenance link not yet verified.
Evidence is weighted, not counted
Ten weak similarities do not necessarily outweigh one decisive contradiction. Correct colour may be limited evidence if it is easily copied. A correct alloy may be supportive but reproducible. A modern pigment in an allegedly untouched early surface may be strong negative evidence. An impossible factory code may be exclusionary. A missing accessory may concern completeness rather than authenticity.
Useful weight descriptions include decisive, strong, moderate, limited, contextual, neutral and unresolved. The report should explain why a feature deserves its weight, rather than presenting the category as an unexplained score.
Myth
More matches mean a stronger authentication
A long list of broad, easily copied similarities can create an illusion of rigour while contributing little discrimination.
Reality
Diagnostic significance matters more than volume
One well-established contradiction may outweigh many cosmetic matches. Authentication asks which explanation best fits the important evidence, not which side has the longest list.
Variation
Normal difference versus disqualifying anomaly
This is often the hardest comparative judgement. A difference becomes meaningful only when its relationship to genuine production is understood.
Sources of normal variation
- Hand finishing, signing circumstances and individual workmanship.
- Tool, mould or die wear and production tolerances.
- Printing registration, material shrinkage and multiple subcontractors.
- Factory changes, revised components and regional production.
- Ageing, use, storage, damage, conservation and restoration.
When an anomaly gains weight
- It is structurally incompatible with the genuine production method.
- It is chronologically impossible or uses an anachronistic material.
- It repeats across known counterfeits or false production families.
- It is absent from a substantial reliable genuine population.
- It affects a tightly controlled feature and is supported by independent contradictions.
Similarity does not prove common origin
Two objects may share features because one copied the other, both copied a third source, both came from the same modern reproduction process, a counterfeit mould was taken from a genuine object, components were transferred, or images were manipulated.
Comparison can demonstrate correspondence. The cause of that correspondence remains an interpretation that must be supported. A repeated scratch, for example, may indicate common manufacture, but it may instead show that a mould or transfer process reproduced damage from a source object.
Convergence
Independent evidence streams strengthen one another
The strongest conclusions usually arise when genuinely independent lines of comparison point in the same direction.
Visual
Design and proportions correspond with the relevant type.
Manufacture
Moulding, striking, printing or assembly evidence matches genuine production.
Material
Tested material is compatible with the claimed production and no anachronism is found.
Documentary
Typography, factory codes, labels or catalogue details are appropriate.
Provenance
The object can be traced to an independently documented collection or transaction.
Negative testing
No known counterfeit diagnostic or decisive contradiction is observed.
Evidence is not independent merely because it appears in several places. Multiple descriptions copied from the same auction listing, photograph or certificate remain one underlying source. Apparent convergence built from duplicated information should not inflate confidence.
Documentation
Build a report another specialist can reconstruct
The purpose of documentation is not to decorate a verdict. It is to preserve enough evidence and method for another reader to understand, test and revise the reasoning.
Identify the exact object and claim
- Description, maker or claimed maker, category and claimed date.
- Dimensions, weight, materials, edition, serial and inscriptions.
- Component inventory, distinguishing marks, photographs and report number.
- The precise proposition being tested, including issue, factory, variant or component where relevant.
Define scope and examination conditions
- Whether the whole object, signature, packaging, component, supplied images or documents were examined.
- Lighting, magnification, imaging equipment, measurement tools and scale.
- Whether the object was removed from a holder or packaging.
- Inaccessible areas, restoration, restricted testing and other limitations.
Build a comparator register
- Identifier, description, source and current location where appropriate.
- Reason the comparator is considered reliable.
- Date, production and variant relevance.
- Whether examined physically, through photographs or through a publication.
- Access date, image reference and known limitations.
Preserve the reasoning
- Comparative criteria selected before conclusion.
- Observation separated from interpretation and conclusion.
- Positive, negative, neutral and unresolved findings.
- Weight assigned to each significant finding and why.
- Alternative explanations considered and rejected or retained.
Make the work reviewable
- Annotated full-object and detail photographs with orientation and scale.
- Original images preserved alongside adjusted or overlaid copies.
- Measurement tables, raw test results and cited references.
- Limitations, confidence wording, author, date and revision history.
Observation, interpretation and conclusion
Observation
A raised linear feature runs continuously around the edge.
Interpretation
Its form and position are consistent with the joining seam expected on a two-part cast or electrotype reproduction.
Claim impact
The feature is inconsistent with the claimed struck manufacture and carries strong negative weight.
Example comparison matrix
| Feature | Questioned object | Reference range | Assessment | Weight |
|---|---|---|---|---|
| Dimensions | Recorded values | Documented genuine range | Consistent | Moderate |
| Maker mark | Correct form and location | Stable across references | Consistent | Strong |
| Surface texture | Softer than references | Crisp mould texture | Inconsistent | Moderate |
| Polymer | Identified material | Known genuine material | Consistent | Strong |
| Provenance | Unsupported dealer claim | No independent record | Unverified | Limited |
Annotated images should preserve the comparison, not conceal it
Include full object views, front and back, sides and edges, marks, defects, components and comparison areas. Distinguish clearly between the questioned object and each reference. Captions should identify orientation, scale, magnification, adjustments and why the feature matters.
Preserve original files alongside cropped, contrast-adjusted, aligned or overlaid versions. An overlay is useful only when another reader can see how it was produced and assess whether perspective or processing created the apparent match.
Workflow
A defensible comparative sequence
The sequence below converts comparison from an impression into an auditable authentication process.
Define the claim
Specify the maker, date, issue, factory, variant, signature, component or other proposition being tested. 'Is it real?' is usually too broad.
Record before analysis
Photograph, measure and describe the submitted object before alteration and, where possible, before absorbing a persuasive seller story or prior conclusion.
Identify comparator groups
List the relevant dates, factories, dies, moulds, editions, signing periods, packaging revisions and production families that could explain the object.
Rank comparator reliability
Separate independently verified references from prior opinions, market claims and other questioned objects.
Establish genuine variation
Determine which features are fixed, tightly controlled, commonly variable, batch-dependent or altered by age, use and restoration.
Set comparative criteria
Choose the relevant features before presenting the conclusion so the analysis does not select only convenient matches after the event.
Compare systematically
Use consistent terminology, images, scale, measurement locations and observation categories across the submitted object and references.
Investigate anomalies
Test whether each difference is explained by variation, wear, damage, restoration, replacement, altered genuine manufacture or false production.
Add independent evidence
Use provenance, archival research, material analysis or specialist examination where comparison alone cannot discriminate between explanations.
Record competing explanations
Keep genuine variant, altered genuine object, replacement component, period copy and modern counterfeit in view until the evidence supports narrowing them.
Reach a proportionate conclusion
Match the wording and confidence to the evidence. Do not convert compatibility into proof or uncertainty into false precision.
Preserve the audit trail
Retain the comparator register, source copies, images, measurements, annotations, reports, access dates and later revisions.
Failure modes
Common methodological failures
Most weak comparison reports fail through reference quality, hidden selection, category confusion or overclaim rather than through a complete absence of observation.
Reference failure
Comparing against one convenient object
The reference may be repaired, atypical, incorrectly attributed or from another production variant. One match can illustrate; it rarely defines the genuine population.
Circularity
Uncertain objects authenticating each other
Object A is accepted because it resembles B, while B is accepted because it resembles A. Repetition of a claim does not create independent evidence.
Variant error
Mixing factories, issues or periods
A genuine reissue, regional version or late-life signature may be a poor comparator for an earlier claimed form. Like-for-like selection is part of the method, not a later refinement.
Weighting error
Counting similarities instead of judging significance
Ten easily copied matches do not necessarily outweigh one chronologically impossible material or decisive counterfeit diagnostic.
Selection bias
Recording matches and hiding contradictions
A report containing only confirming features may be advocacy rather than investigation. Negative and unresolved observations belong in the record.
Overclaim
Letting one method answer the whole object
A genuine signature does not authenticate the memorabilia item, and a compatible material does not establish maker, date or provenance.
Authority shortcut
Treating a certificate as self-proving
The certificate, issuer, methodology and object-document link must each be assessed. A confident verdict without an inspectable basis remains a prior opinion.
Record loss
Failing to preserve comparison sources
Listings disappear, web pages change and databases are corrected. A conclusion can no longer be audited when its comparators and images vanish.
Application
How comparative priorities change across collectible types
The underlying method remains the same, but the most discriminating characteristics vary by category, production process and claim.
Coins and medals
Die, strike, edge and metal
- Compare die markers, letterforms, relief, edge, weight, diameter and alloy.
- Distinguish genuine die variation from casting, transfer-die or altered-date evidence.
Autographs
Exemplars and natural variation
- Compare contemporaneous secure exemplars, speed, rhythm, line quality, placement and ink.
- Do not treat mechanically identical signatures as stronger matches.
Trading cards
Stock, print and cut
- Compare dimensions, card stock, print pattern, colour separation, registration, fonts, gloss and security elements.
- Account for known reprints, trimming, recolouring and altered surfaces.
Toys and figures
Mould, polymer and assembly
- Compare mould geometry, copyright marks, plastics, joints, screws, paint, accessories and factory codes.
- Separate packaging authenticity from figure and accessory authenticity.
Books and ephemera
Edition points and manufacture
- Compare typography, paper, binding, signatures, plate position, publisher marks and dust-jacket details.
- Distinguish a genuine later issue from the claimed first issue.
Ceramics and glass
Form, body and finishing
- Compare form, body composition, glaze, foot, pontil, firing marks, seams, decoration and factory marks.
- Allow for hand finishing and kiln variation while testing chronological compatibility.
Watches and mechanisms
Component chronology
- Compare case, movement, calibre, serials, dial, hands, finishing and hallmarks.
- Identify service replacements and assembled objects rather than forcing a whole-object verdict.
Prints and photographs
Process, margins and image state
- Compare paper, dimensions, printing method, dot pattern, cropping, margins, inscriptions and edition marks.
- Test whether a later reproduction copied an original image or damage pattern.
Collector judgement
How to read a comparison-based authentication report
- What exact claim was tested, and does it match the claim being made in the sale, insurance or collection record?
- What was compared, and were the references the same issue, factory, variant, production period or signing period?
- Why are the references trusted? Are they independently documented or merely commonly accepted?
- How many references were used, and was the range of genuine variation considered?
- Were differences and negative findings recorded as carefully as similarities?
- Which features are genuinely diagnostic, and which could be copied easily?
- Are observation, interpretation and conclusion distinguishable?
- Can another specialist inspect the images, measurements, sources and reasoning?
- Does the conclusion authenticate the whole object or only a signature, component, document, period or production family?
- Does wording such as "consistent with" remain appropriately distinct from "proven authentic"?
Comparison can
- Demonstrate consistency or identify contradiction.
- Associate an object with a production family.
- Detect known counterfeit characteristics.
- Place an object inside or outside a genuine range.
- Distinguish original components from later additions.
- Direct further testing and strengthen or weaken provenance claims.
Comparison cannot automatically
- Prove authorship from style alone.
- Prove provenance from resemblance.
- Prove authenticity from age-compatible material.
- Eliminate every unknown legitimate variant.
- Guarantee that a certificate or prior opinion is valid.
- Create certainty where the reference population is incomplete.
Escalation
Know when collector-level comparison has reached its limit
Routine comparison can identify obvious reprints, dimensional errors, known packaging variants and elementary counterfeit features. It can also establish that further examination is needed.
Seek a qualified specialist when
- The object is high value, historically important or likely to be disputed.
- The counterfeit is sophisticated or the genuine reference population is poorly understood.
- The item may be restored, assembled, relabelled or made from mixed-period components.
- The conclusion depends on handwriting, metallurgy, polymers, pigments, paper, ink or printing technology.
- Scientific testing, sampling or access to hidden construction is required.
- The outcome may materially affect sale, insurance, lending, publication or legal representations.
Do not clean, disassemble, sample, remove labels, lift seals or expose hidden areas merely to improve comparison unless the likely benefit, risk and authority to intervene are understood. Potentially damaging access belongs with an appropriately qualified conservator, scientist or category specialist.
Model conclusion
The submitted object was compared with six independently documented examples from the same factory, mould family and production period. Its mould geometry, underside marks, component construction and measured dimensions fall within the observed genuine range.
Paint application differs in minor respects but is compatible with the hand-applied variation recorded across the reference group. No known counterfeit diagnostic was observed. The object is therefore considered highly likely to be an authentic production example, subject to the stated limitation that no polymer analysis was undertaken.
Key takeaways
- Comparison authenticates through significant correspondence, not general resemblance.
- The reference hierarchy and like-for-like selection determine how much weight a comparison deserves.
- Normal variation must be established before difference is treated as anomaly.
- Findings should be classified and weighted; they should not simply be counted.
- Independent evidence streams strengthen confidence only when they do not derive from the same underlying source.
- A strong report preserves the object, claim, comparators, observations, limitations and reasoning route.
- The conclusion should state exactly what was authenticated and stop at the edge of the evidence.
Continue learning
Provenance as Evidence Documentation
Review how to document provenance as independent, object-specific authentication evidence.
Back to Authentication Documentation
Return to the Authentication Documentation sub-domain and its full topic list.
Source Risk
Continue to how seller, source and market context affect authentication confidence.
Related topics
Authentication Methodology
Understand the reasoning sequence that moves from first impression to defensible conclusion.
Evidence-Led Authentication Documentation
Build an authentication record around evidence, reasoning and uncertainty rather than conclusion alone.
Evaluating Expert Conclusions
Read expert conclusions backwards from verdict to question, method and unresolved limits.
Interpreting Test Results
Keep test results tied to sample, method, question and claim impact.