Factory Checks: Distinguishing a Factory Claim From a Documented Production Control
Published: Read Time: 12 minutes

Factory Checks: Distinguishing a Factory Claim From a Documented Production Control

A practical pre-sourcing method for separating factory claims from documented, traceable production controls and planning a distinct Factory checks task.

Executive introduction. A factory statement such as “every batch is inspected” may be relevant to pre-sourcing research, but it is not proof that inspection occurs. A procedure offers stronger support because it defines what should happen, yet even a polished procedure does not establish that personnel consistently follow it.

A documented production control connects an applicable requirement to a current procedure, responsible roles, traceable implementation records, results, and action when requirements are not met. The purpose of pre-sourcing research is not to approve the supplier from a document review. It is to identify which claim matters, evaluate the evidence level, and plan a distinct Factory checks task.

This article provides a practical method for doing that. It draws on audit-planning and evidence-evaluation concepts in ISO 19011, attribute-sampling concepts in ISO 2859-1, and product-safety-oriented manufacturing practices described in the CPSC Manufacturing Best Practices.

Decision rule: claim, document, or verified control

Purpose at the pre-sourcing stage

Treat factory statements as research inputs, not proof. For a production or product-safety claim that could affect the sourcing decision, determine whether the available evidence includes:

  • A defined requirement, specification, or procedure.
  • Assigned responsibility for performing and reviewing the control.
  • Controlled and current documentation.
  • Records showing that the procedure was performed.
  • Monitoring, measurement, or inspection results.
  • A process for handling nonconformities.
  • Corrective-action and follow-up evidence where applicable.

Use the result to plan a separate Factory checks verification task. Do not use this initial classification as a final supplier-approval, compliance, safety, or capability decision.

Three evidence levels

Evidence level Meaning Typical finding Pre-sourcing treatment
Claim only The factory describes a capability or practice but provides no suitable supporting evidence. “Every batch is inspected.” Record the statement as unverified; request defined controls and implementation records.
Documented arrangement A procedure, plan, specification, instruction, or form defines what should happen. An inspection procedure identifies checkpoints and responsibilities. Confirm that the document is current, applicable, controlled, and supported by records.
Documented production control Defined requirements are connected to responsibilities, implementation records, results, and follow-up. The applicable procedure is linked to completed inspection records and disposition of nonconforming output. Include implementation, traceability, and exception-handling checks in the Factory checks task.

These levels are a practical classification for planning verification. They do not represent certification grades or establish conformity with an ISO standard.

The core distinction

A polished policy is not, by itself, evidence that a control operates. It might describe intentions without defining the method, acceptance criteria, frequency, responsible role, or records required.

A completed record is also insufficient when its context cannot be identified. The verifier should be able to determine which product, material, batch, lot, line, date, procedure, result, and reviewer the record concerns, as applicable.

Equipment, personnel, certifications, photographs, and sample reports can provide context. None automatically demonstrates an operating production control. Stronger evidence forms a traceable chain:

Claim → applicable requirement → controlled procedure → responsible role → implementation record → result → nonconformity response → follow-up

Questions that test the evidence chain

  • Applicability: Does the control apply to the product, process, site, and production stage under review?
  • Definition: Does it state what is checked, how it is checked, and what result is acceptable?
  • Responsibility: Is a role assigned to perform, review, release, escalate, or act on the control?
  • Document control: Can the factory identify the current version, approval, or effective status?
  • Implementation: Are there records showing that the defined activity occurred?
  • Traceability: Can records be linked to the relevant material, product, batch, lot, line, or production period?
  • Nonconformity handling: Do records show what happens when requirements are not met?
  • Follow-up: Is corrective action reviewed or otherwise followed through where applicable?
  • Consistency: Do interviews, documents, records, and production-floor observations agree?

Plan the distinct Factory checks verification task

1. Define a narrow verification objective

Write the task to test a specific control rather than instructing someone to “check the factory.”

Determine whether the factory’s stated control for [production stage or product characteristic] is defined, current, implemented, traceable, and followed when results do not meet requirements.

Do not complete the bracket with an unsupported assumption. Fill it only after identifying a factory claim that matters to the sourcing decision.

The scope should also state the relevant site, product or product family, process, production stage, and examination period. A narrow scope makes it easier to distinguish missing evidence from evidence that simply falls outside the task.

2. Set criteria before requesting evidence

Build the criteria from:

  • Applicable buyer or product requirements, if separately established.
  • The factory’s stated procedures and acceptance criteria.
  • Relevant manufacturing practices identified in the CPSC Manufacturing Best Practices.
  • Audit-planning and evidence-evaluation principles from ISO 19011.
  • ISO 2859-1 when inspection by attributes and lot-by-lot sampling are appropriate.

Do not invent an acceptance quality limit, inspection level, sample size, acceptance number, or rejection number. If sampling is proposed, require the applicable parties to identify the plan and confirm that it is suitable for the inspection purpose.

3. Convert each claim into a testable proposition

Factory claim Verification proposition Evidence to request On-site or remote check
“We inspect incoming material.” Incoming material is checked against defined requirements before release or use. Current procedure, material specification, completed records, and status or disposition records. Trace selected records to material identification and verify how failed material is controlled.
“Operators follow work instructions.” Applicable, current instructions are available and used at the relevant operation. Approved instruction, revision information, and training or authorization records where applicable. Compare the instruction with the activity and ask the responsible role to explain key steps.
“Production is monitored.” Defined process or product characteristics are measured at stated stages or frequencies. Control plan or procedure, monitoring records, acceptance criteria, and exception records. Trace entries to the relevant production period and compare actual practice with the documented frequency.
“Every lot is inspected.” A defined lot, inspection method, sampling basis, and disposition process exist. Lot definition, inspection plan, completed results, and release or rejection records. Select identifiable lots and trace inspection results through disposition.
“Problems are corrected.” Nonconforming output is identified, controlled, investigated, and followed up under a defined process. Nonconformity procedure, records, disposition decisions, and corrective-action records where applicable. Trace a recorded exception from detection through disposition and closure.

These are verification patterns. They are not assertions that a particular factory performs the activities described.

4. Prepare a focused evidence request

Request only the material needed to test the selected claim:

  • Applicable procedure, work instruction, control plan, or inspection plan.
  • Revision, approval, or effective-status information.
  • Defined product or process requirements.
  • Assigned responsibilities.
  • Blank forms and corresponding completed records.
  • Identification needed to connect records to production.
  • Relevant nonconformity, disposition, and corrective-action records.
  • Sampling-plan information if attribute sampling is used.
  • Restrictions on viewing, copying, photographing, or retaining evidence.

Requesting a blank form together with completed examples helps distinguish the intended process from recorded implementation. Neither should be evaluated in isolation.

5. Use more than one verification method

An audit-style task should compare evidence from multiple methods rather than relying on a single interview or selected document. ISO 19011 provides guidance for structuring audit objectives, scope, criteria, evidence gathering, findings, and reporting.

Plan to use an appropriate combination of:

  • Document review.
  • Record review.
  • Interviews with relevant roles.
  • Observation of the activity or production area.
  • Traceability from a production item or lot back to records.
  • Traceability from a record forward to release or disposition.
  • Sampling where justified and appropriately planned.

Record contradictory evidence as a finding or unresolved issue. Do not resolve a contradiction by assuming that the most favorable explanation is correct.

6. Define the required output

The Factory checks task should report:

  • The exact claim tested.
  • Scope boundaries, including site, process, product, stage, and period examined.
  • Criteria used.
  • Evidence reviewed.
  • Sampling basis, if any.
  • Findings and unresolved contradictions.
  • Access or timing limitations.
  • Follow-up actions, owners, and due dates if assigned by the sourcing team.

Use one of these evidence statuses:

  • Documented and implementation evidence observed
  • Documented, but implementation not established
  • Claim only
  • Not verifiable within the task scope

A status applies only to the examined scope and evidence. It should not be extended to other sites, products, periods, or processes without additional verification.

Practical verification checklist

Task setup

Check Verification action Record in task file
Write the factory claim verbatim or accurately paraphrase it. Claim and source of claim.
Identify why the claim matters to sourcing or product safety. Decision relevance.
Define the product, process, site, and production stage in scope. Scope statement.
Convert the claim into a proposition that can be supported or contradicted. Verification objective.
Establish criteria before reviewing evidence. Criteria and source references.
Identify needed competence, access, timing, and confidentiality arrangements. Task constraints.
Keep the Factory checks task distinct from supplier self-description and sales discussions. Task owner and scheduled activity.

Documented-control review

Check Verification question Acceptable evidence direction Warning sign
Is the control written or otherwise formally defined? Applicable procedure, plan, instruction, or specification. Verbal explanation only.
Is the document identifiable and current? Title, identifier, revision, approval, or effective status. Undated, unapproved, obsolete, or conflicting versions.
Does it define what is controlled? Named material, process, product characteristic, or production stage. Generic language with no clear application.
Does it define how the check is performed? Method, equipment, sequence, or inspection approach. “Inspect as needed” without a defined method.
Are acceptance criteria defined? Requirement, limit, classification, or other decision rule. Pass/fail decisions with no identifiable basis.
Is responsibility assigned? Defined role for performance, review, release, or escalation. No accountable role.
Are records required and retained? Defined record and completed forms linked to the activity. Blank templates offered as proof of implementation.
Is nonconforming output addressed? Identification, control, disposition, escalation, or corrective-action process. Failed results with no recorded response.

Implementation and traceability review

Check Verification question What to trace
Do completed records correspond to the current procedure? Procedure revision to record format and required fields.
Can each record be linked to actual production? Product, material, lot, batch, line, date, or another applicable identifier.
Are required entries complete and attributable? Results, date, responsible role, review, and disposition as applicable.
Does the documented frequency match available records? Required inspection frequency to chronological records.
Do recorded results support the stated decision? Measurements or classifications to acceptance criteria.
Can exceptions be traced to control and disposition? Failed result to segregation, rework, rejection, release authorization, or another recorded decision.
Is follow-up documented where corrective action is required? Issue to action, review, and closure evidence.
Does observed practice agree with documents and interviews? Floor activity to procedures, records, and staff explanations.
Are inconsistencies recorded as findings? Contradictions, missing links, and unavailable evidence.

Sampling-plan review

Use this section only when the verification task includes inspection by attributes. ISO 2859-1 provides sampling procedures for inspection by attributes, including lot-by-lot inspection concepts. A reference to the standard in a factory document does not prove that a plan was correctly selected or followed.

Check Verification question
Is the lot clearly defined?
Is the characteristic classified by attribute using an established requirement?
Is the sampling plan identified rather than described only as “random checking”?
Is the inspection level or other plan input stated?
Is the acceptance quality limit or other applicable indexing basis stated?
Are sample-size and acceptance or rejection decisions derived from the selected plan?
Is sample selection protected from convenience or selective presentation?
Are lot disposition and follow-up decisions recorded?
Has the plan’s applicability been confirmed for the inspection purpose?

Finding classification

Status Use when Required follow-up
Claim only No applicable controlled document or implementation record is available. Request defined evidence or leave the claim unverified.
Documented, not implemented A control is defined, but implementation evidence is absent, incomplete, or contradictory. Plan record tracing and observation.
Implemented evidence observed Applicable documents and traceable records align within the examined scope. Preserve scope and sampling limitations in the conclusion.
Control breakdown observed Records or observations show that the defined requirement was not followed. Record the finding and relevant disposition or corrective-action evidence.
Not verifiable Access, timing, confidentiality, production status, or task scope prevents a conclusion. State the limitation and plan separate follow-up.

Image credit

Inside a large factory, equipment and workers are present. Photo by TECNIC Bioprocess Solutions on Unsplash, used under the Unsplash License.

Use this image only as general production-environment context. Visible equipment, workers, cleanliness, organization, or activity is not evidence of a documented production control. The relevant verification question is whether observed operations can be connected to current requirements, implementation records, results, and follow-up.

Sources

Scope and limits

What this approach supports

  • Pre-sourcing research.
  • Planning a distinct Factory checks task.
  • Separating promotional or verbal claims from documented and traceable controls.
  • Identifying focused evidence requests and verification methods.
  • Recording findings without extending conclusions beyond the evidence examined.

What this approach does not establish

  • That any factory, supplier, product, or process is compliant, capable, safe, or approved.
  • That a document is authentic merely because it is provided.
  • That a control operates continuously because selected records were observed.
  • That a particular sampling plan, acceptance quality limit, inspection level, or sample size is suitable.
  • That ISO guidance or CPSC best practices replace applicable legal, regulatory, contractual, technical, or product-specific requirements.
  • That the selected image is evidence about a factory under consideration.
  • Certification, regulatory compliance, or conformity with an ISO standard.

Final next move

Select one sourcing-critical factory claim and enter it at the top of the checklist. Convert it into a testable proposition, define the applicable criteria and scope, and schedule a separate Factory checks task that traces the claim from the current procedure through production records, results, nonconformity disposition, and follow-up.

Sourcing information earns its value when it is verified, compared and turned into a decision.