A supplier claim is a starting point, not a finding
Claims are necessary. A supplier needs to describe its products, processes, capacity, customers and experience before a buyer can decide what to examine.
The problem begins when a statement is treated as proof merely because it appears in several places.
For example:
- a company website may repeat a sales team's description;
- a marketplace profile may reproduce information submitted by the supplier;
- a brochure may contain photographs from a genuine facility without showing which legal entity operates it;
- a certificate may be authentic but cover a different site, product category or activity;
- a patent may exist without demonstrating current production capability;
- a customer logo may show a relationship, but not its scope, timing or commercial significance.
Repetition does not automatically create independence. Three sources derived from the same supplier statement are still one underlying claim.
The first task is therefore not to decide whether the supplier sounds convincing. It is to define each material claim precisely enough that evidence can be matched to it.
“We have strong manufacturing capability” is too broad to test. A decision-ready version would identify the legal entity, production site, product, process, required tolerance, expected volume and relevant period.
Start with the decision, not a generic checklist
The evidence standard should follow the decision.
A preliminary longlist does not require the same depth as a final supplier appointment. A buyer comparing ten possible partners may begin with legal identity, operating status, stated business scope, relevant product evidence and obvious contradictions. A buyer preparing to place a material order may also need current certification scope, technical records, sample or trial results, site-level checks, contract review and specialist quality assurance.
This is why a single universal “verified supplier” label can be misleading. Verification is always verification of something, for a defined purpose and within a defined scope.
A useful screening brief should state:
- the decision the research will support;
- the capability that matters;
- the minimum evidence required at the current stage;
- the issues that require direct supplier confirmation; and
- the conditions that would trigger deeper technical, legal or on-site review.
This risk-based approach also prevents unnecessary work. Not every supplier needs the same level of investigation at the same time. Research effort should concentrate on the claims and risks that could change the decision.
How to verify a Chinese supplier's capability: four evidence layers
1. Legal identity and operating context
The first layer establishes which organisation is being assessed.
Useful questions include:
- What is the company's registered Chinese name?
- What is its Unified Social Credit Code?
- What is its current registration status?
- Which entity signs contracts, receives payment or holds relevant licences?
- Is the production or service activity carried out by the same entity, a subsidiary, an affiliate or an unrelated supplier?
Chinese registration information can establish important identity and corporate facts. It cannot, by itself, prove that a company has the people, equipment, quality controls or experience needed for the buyer's requirement.
The registered business scope should therefore be read as one contextual signal, not as conclusive proof of actual production. A manufacturing-related scope may be consistent with the supplier's claim; it does not demonstrate current output, capacity or product-level competence.
2. Supplier-reported capability
The second layer records what the supplier says.
This may include product range, manufacturing processes, capacity, research and development, customer sectors, export experience, quality systems and delivery performance.
Supplier-provided information is not automatically unreliable. It should simply be labelled according to its source. Clear labelling allows the buyer to distinguish a useful response from an independently supported conclusion.
Where possible, convert broad descriptions into specific questions:
- Which site performs the relevant process?
- Which products with comparable technical requirements have been delivered?
- Which part of the work is performed internally and which is subcontracted?
- Which legal entity owns or operates the equipment?
- What period does the stated capacity refer to?
- What evidence can be shared without breaching customer confidentiality?
3. Source-backed evidence
The third layer connects each important claim to evidence with an identifiable source, date and scope.
Depending on the question, this may include:
- official corporate and ownership records;
- certification documents and their stated scope;
- product-specific test reports;
- patent and intellectual-property records;
- environmental, regulatory or industry records where relevant;
- dated project, customer or delivery evidence;
- technical documentation;
- production records supplied under appropriate confidentiality controls; and
- observations or test results from qualified third parties.
Each item should be assessed for what it actually proves.
An ISO 9001 certificate, for example, concerns a quality management system. The organisation, physical site and activities covered by the certification scope matter. The certificate should not be treated as proof that every product meets the buyer's technical requirements or that the supplier can deliver a particular volume on a particular schedule.
Similarly, patent records can help examine whether a company or related entity has filed in relevant technical areas. A patent count alone does not establish commercialisation, ownership of all technology used, freedom to operate or the ability to manufacture the buyer's product.
4. Decision-specific confirmation
The fourth layer tests whether the evidence is sufficient for the next decision.
Some questions can be answered through desktop research. Others require direct supplier clarification, document review, samples, reference checks, site observation, an audit, laboratory testing or specialist legal and technical advice.
The purpose of research is not to make these boundaries disappear. It is to show the buyer which conclusions are supported now and which require a different form of verification.
Supplier capability assessment: a six-part claim-to-evidence test
For every material claim, apply six checks.
Scope
Does the evidence refer to the correct legal entity, site, product, process and time period?
Source relationship
Is the source a government record, an independently issued document, a customer or third-party source, or the supplier's own statement? Independence should be assessed, not assumed from the number of websites repeating the information.
Specificity
Does the evidence address the capability in the brief, or only a broad adjacent capability?
Currency
Is the information recent enough for the decision? A historical project can show experience but may not demonstrate current personnel, equipment or available capacity.
Consistency
Do company records, addresses, certification documents, product materials and direct answers point to the same operating reality? Contradictions are not always disqualifying, but they should be resolved rather than silently averaged away.
Decision threshold
Is the evidence sufficient for the current stage, or is a deeper check required before commitment?
Common mistakes in supplier screening
Treating company existence as proof of capability
A valid legal entity answers “does this entity exist?” It does not answer “can it perform this work?”
Treating a certificate as a universal endorsement
Certification can be important, but its issuer, validity, site and scope must match the conclusion being drawn.
Counting patents, customers or exports without context
Counts can create an impression of substance while hiding relevance, ownership, timing and product fit.
Rewarding the supplier with the most public information
Larger or more internationally active companies often disclose more. More information may improve research confidence, but it is not automatically evidence of better fit.
Converting missing evidence into a negative fact
“Not found” is not the same as “does not exist.” The correct result may be an unresolved question and a specific next verification step.
Asking every supplier the same questions without defining the decision
Consistency matters in comparison, but a generic questionnaire can produce large amounts of information that do not resolve the material risks.
What a decision-ready output should show
A useful supplier screening report does not need to declare every company simply “verified” or “unverified.” It can classify material findings more precisely:
- Supported within scope: relevant evidence supports the claim for the stated entity, site, activity and period.
- Supplier-reported: the supplier has made the statement, but independent or decision-specific support remains limited.
- Partially supported: some elements are evidenced, while important parts of the claim remain unresolved.
- Inconsistent: available sources do not align and require clarification.
- Not established: the required evidence was not available within the research scope.
- Requires specialist confirmation: the issue falls outside desktop research or needs technical, legal, financial, regulatory or on-site examination.
This structure is more useful than a polished profile because it preserves the relationship between the claim, the evidence, the limitation and the next decision.
Better screening produces better questions
Supplier research should not promise certainty that the available evidence cannot provide. Its value lies in improving the quality of the buyer's next move.
That may mean removing a candidate whose stated capability does not match the requirement. It may mean retaining a supplier while requesting a site-specific certificate, a product-level test report or clarification of which entity performs the work. It may also mean recognising that two suppliers cannot yet be compared fairly because the evidence available for each is fundamentally different.
The strongest outcome is not the longest supplier file. It is a defensible view of:
- what is known;
- what is claimed;
- what is supported;
- what remains uncertain; and
- what should happen before the buyer commits.
Research notes
- State Administration for Market Regulation, Measures for the Administration of Unified Social Credit Codes for Organisations (2025).
- International Organization for Standardization, How can you have confidence that your supplier meets ISO 9001?.
- OECD, Due Diligence Guidance for Responsible Business Conduct.
- China National Intellectual Property Administration, introduction to the Patent Search and Analysis System.
Scope note: This article describes a desktop-research framework. It is not a factory audit, product test, legal opinion, certification decision, financial audit or guarantee of supplier performance.