The cloth is woven from this chapter's 4 exhibits: each weft is one exhibit's own data, normalised; each warp group is a section. The probe reads the exhibit's own value where it crosses. Procedural visualisation, not a measurement.
EX 60
EX 61
EX 63
The chapter as a draft: 7 sections, 3 exhibit wefts, one shuttle.
Lead exhibit · 60Decision mapillustrativeCritical failures remain visible at the release gate
01 / 06
Read-out
Method and source
Illustrative GFT decision map. X runs from routine evidence administration (0) to operational process change (10). Y runs from limited release consequence (0) to decisive protection or release consequence (10). Positions are categorical design aids, not measured risk, legal findings or supplier scores. Sources support the issue categories, not the coordinates.
Unit: illustrative decision coordinates from 0 to 10. Status: illustrative. Sources: S09, S16, S17, S33. Illustrative figures are teaching examples, not measured factory results.
The chapter's warpBar: share of the reading · square: an exhibit
GFT would approve the compliance roadmap only when it can show two operating results: people are protected from harm, and each product can be answered for at its release boundary. A larger document library, a busy reporting calendar or a high average score cannot establish either result. The roadmap therefore begins with a dated decision about the sales route, legal roles and current product duties. It then tests a bounded evidence pack, measures whether workers confirm remedy and whether batches retain the right evidence, and changes scale only after review. Future rules enter through a controlled watchlist rather than an invented deadline.
This is a proposed operating system, not a claim that GFT has implemented or certified an export factory. It uses the Cameroon textile programme as a design reference for feasibility, performance management and digital strategy. Field performance, buyer acceptance, legal conclusions and implementation costs remain to be established. The recommendations are dated as of 30 October 2026 and are not legal, tax or investment advice. Qualified reviewers must confirm the actual entity, product, destination, contractual terms and current law before a shipment or investment decision.[S02][S35]
The management rule is deliberately strict. Administrative completeness may be averaged across a defined requirement population, but a critical failure may not be hidden inside that average. An unresolved worker-protection issue, an unreviewed safety concern or missing evidence for an applicable chemical restriction remains a hold even when every reporting field is complete. The amended CSDDD itself connects corporate due diligence with complaints, monitoring and public communication, while EU product-safety guidance requires action when a business identifies an unsafe product.[S02][S33] The roadmap joins those duties to daily production decisions without treating them as interchangeable legal regimes.
14.13 min · 8 paragraphs · read with a related exhibit
Start by approving the route, roles and current-law register
The first approval is not a technology choice. It is a written scope determination naming the contracting entities, production sites, product family, intended use, sales route, consignee, importer and other relevant economic operators. The reviewer should also identify the destination countries and the party making each public claim. A factory cannot select the correct evidence population while these questions remain open. Nor can it distinguish a direct legal duty from a buyer request merely by reading the buyer's questionnaire.
The route decision should produce a role map rather than a generic statement of EU readiness. For each obligation or request, record the duty holder, the operational contributor, the evidence reviewer, the payer and the person authorised to act. Legal advice, investment approval, product release and publication authority need separate decision rights. One sponsor may coordinate the register, but that sponsor should not silently acquire every approval. Roles are named by function here because no GFT staffing or client organisation is assumed.
¶ 02
A current-law register then converts the route into dated rows. Each row needs the instrument, provision, product or entity scope, legal status, application conditions, relevant national implementation, source version and next review trigger. Adoption, publication, entry into force, transposition and application are different states. The Commission's acts register, as opened for this research, lists the July 2026 simplified reporting and voluntary value-chain standards as not in force until Official Journal publication. A roadmap should preserve that qualification rather than treating adoption as immediate application.[S04]
The same discipline applies to rules already described as applying. The Commission states that the consumer-law changes entered application on 27 September 2026, while the PPWR began applying on a phased basis from 12 August 2026. Those dates do not make every provision identical in scope or make every supplier the duty holder. The register must still identify the actual claim, packaging, role and destination.[S18][S20]
Corporate due diligence needs a separate row from product evidence. The Commission describes the amended CSDDD as a duty for very large companies within its scope, with Member States applying the principal transposition measures from 26 July 2029 and the Article 16 reporting measures for financial years starting on or after 1 January 2030. These are not universal African factory deadlines. They are review points for the buyer relationship and any entity that may itself be in scope.[S02][S35]
The adopted amending text also shows why purpose belongs in every register row. Its reporting protection for protected undertakings concerns information requested for sustainability reporting, and its CSDDD information safeguards answer a different purpose. Product-safety and chemical evidence cannot be dismissed by attaching a general reporting label to the request. The route reviewer should record why the information is needed before selecting a response or a refusal.[S35]
Approval requires a short exception log. Record provisions not checked, national transposition not verified, uncertain product classification, unavailable buyer terms and any evidence that could not be retrieved. An exclusion should state its reasoning and its reopening trigger. A register that says only compliant or not applicable hides the work that remains. A register that preserves conditional and unresolved states gives management a decision it can safely revisit.
¶ 07
Note
14.23 min · 9 paragraphs · 1 exhibit
Make worker protection and current product duties the first gate
The first operating gate protects people and the product before it optimises administration. Worker protection begins with safe recruitment and grievance routes, the ability to escalate beyond the production hierarchy, and a remedy process that can be checked with the affected person. The ILO and IOE handbook uses the forced-labour definition and its indicators to help companies identify and assess risks in practice. An indicator should start enquiry and protective action, not become an unsupported accusation or a line item deferred until the order closes.[S09]
Recruitment and remedy records should therefore sit outside the shipment deadline's control. A supervisor under pressure to dispatch should not be the only person able to close a concern about withheld documents, debt or coercion. The proposed gate asks whether the affected person is safe, whether immediate action has an independent owner, whether evidence is restricted to those who need it and whether the agreed remedy has been confirmed. The gate remains open when administrative closure is not matched by the person's experience.
¶ 02
Product evidence forms the other side of the same release decision. EU guidance states that every textile product marketed in the EU must be labelled or marked to show fibre composition, subject to the applicable scope and business-to-business provisions. The bill of materials, approved composition, language version and printed label therefore need a traceable relationship to the product being released.[S17] A claim approved for an earlier material specification should not be carried across an unreviewed change.
Chemical evidence needs its own product-specific test. The Commission explains that REACH restrictions on placing substances on the market apply to domestic production and imports, and identifies restrictions for clothing, textiles and footwear in Annex XVII entry 72. That establishes the need to examine the actual article and restriction. It does not allow a generic supplier declaration to replace the operative limits, method, sample identity or batch relationship.[S16]
Safety continues after production. EU guidance states that products sold in the EU must be safe regardless of where they were made, and that manufacturers and other businesses linked to distribution or sale must act if they find an unsafe product. The release file should therefore connect design and test evidence to complaints, corrective action, withdrawal or recall routes where applicable. A sustainability label or a completed buyer portal does not answer that product-safety question.[S33]
Exhibit 60 shows why the release decision should not become a single maturity score. The vertical dimension is the consequence of leaving an issue unresolved, not a measured legal risk. The horizontal dimension is the degree to which the issue sits in routine evidence administration or requires operational change. A duplicate schema can be inconvenient without being decisive. Unresolved chemical evidence or a worker-protection failure requires containment or protection regardless of the average completeness score.
The map is not permission to rank one person's harm against another product failure. It is a rule against compensation between unlike failures. Missing reporting information may justify clarification and rework. A label mismatch may require correction and a new approval. An unresolved safety, chemical or worker-protection condition may require a hold, urgent action or specialist escalation. Each state needs its own decision and named authority.
¶ 07
A hold should be precise about its object. Identify the affected people, requirement, input lots, batches, claims and shipments. Record what is known, what is missing, who may release the hold and what evidence would change the decision. Do not describe an internal precaution as an authority prohibition. Conversely, do not wait for an external authority before taking a protective step that the supplier can already see is necessary.
¶ 08
The gate also needs a change trigger. A new dye recipe, trim supplier, production site, subcontractor, intended use or consumer claim can change the evidence question. A responsible reviewer should determine whether the earlier evidence remains valid before the changed product enters the eligible release population. A form marked complete after the change is not evidence that the review occurred before it.
¶ 09
14.33 min · 8 paragraphs · read with a related exhibit
Prove a reusable pack on a bounded product family
The pilot should be small enough to inspect and real enough to expose the work. Select one product family, one defined sales route and a limited set of applicable requirements after the scope review. The product choice, order volume and acceptance criteria must come from the actual opportunity. This chapter does not invent them. The useful unit is not a generic factory transformation but a controlled chain from product specification to production event, batch evidence, review, release and buyer response.
The pilot pack should contain the route-and-role statement, current-law rows, product specification, applicable label and test evidence, change history, worker-protection route, corrective-action register and a controlled response index. Each item needs an owner, period, evidence location, version and review state. Missing, disputed, corrected and not-applicable entries should remain distinguishable. A blank field must not become a zero incident count or a declaration that every upstream input conforms.
¶ 02
Reuse begins with a canonical answer rather than a copied attachment. The answer should retain its definition, source boundary and effective version so that a later request can be matched to it. The Commission's VSME recommendation is intended to reduce administrative burden by making it easier for SMEs to answer sustainability-information requests from large companies and financial institutions. That supports a common response structure for its intended purpose, not a universal product file or proof that every buyer will accept the same extract.[S05]
Purpose-based triage protects the pack from request inflation. Route each question to sustainability reporting, due diligence, product safety, chemicals, commercial specification, claims or another identified purpose. Reuse only evidence with the same population, period and meaning. The adopted reporting amendment gives protected undertakings a right, subject to its conditions, to decline information beyond the voluntary standards for specified reporting requests. It does not create a blanket refusal for necessary product or investigation evidence.[S35]
A retrieval test should use a real question and a controlled clock. Record when the request becomes sufficiently clear to answer, when the reviewer receives the pack and how much active work was needed for collection, translation, checking, approval and portal entry. Keep recipient waiting time separate from supplier work. A portal acknowledgement is not buyer acceptance, and a file found quickly is not necessarily the right file.
¶ 05
Corrective action belongs inside the pilot rather than after its evaluation. If the pack lacks a batch link, assign the repair and repeat the retrieval. If a worker-related record exposes unnecessary identity information, revise access before circulating the pack again. If a product test does not cover the changed specification, hold the relevant release and establish the evidence required. Scaling should be conditional on these repeated decisions working, not on the first demonstration looking complete.
¶ 06
Note
The scale decision needs evidence of review as well as evidence of completion. A qualified legal reviewer may need to confirm an unresolved scope point. The buyer may need to confirm that a controlled extract answers its stated purpose. Operations may need to demonstrate that the release gate holds when production is busy. Worker-protection actions need appropriate confirmation and recurrence review. None of these is interchangeable with a draft pack or a successful upload.
¶ 07
Interlude · the cloth half-woven
4 exhibits carry this chapter's argument.
Woven: sections 14.0 to 14.3Still to weave: 14.4 to 14.7
Sourced legal-event sequence as of 30 October 2026. Numeric positions 0 to 6 are ordinal and equally spaced, not date calculations. Exact dates appear in labels. Q4 2027 is the Commission's indicative planned textile delegated-act adoption window, not a universal DPP application date. National implementation, product scope and economic-operator role require separate review.
14.4 · the bodyExhibit 61 above; its figures are illustrative where marked.
The watchlist should change the evidence system only when a verified source changes the applicable question. Each row needs an owner, official source, instrument, status, affected product or entity, last check, next review and implementation decision. Proposal, adopted, published, in force, applicable and guidance states should remain separate. A future requirement should not enter the release checklist merely because a vendor has added a field with the same name.
Textile ecodesign and the Digital Product Passport illustrate the control. The Commission states that exact textile information requirements will be defined through relevant delegated acts and supporting technical specifications. Its indicative timeline gives Q4 2027 as planned adoption of the textile delegated act and expressly warns that implementation timelines may evolve. The correct preparation is reusable product identity and evidence mapping, not a claim that every garment must carry a completed passport by that quarter.[S11]
The forced-labour product regime has a different status and decision path. The Commission states that the prohibition applies from 14 December 2027 to covered products placed on, made available on or exported from the EU market, regardless of origin. It also states that the regulation introduces no audit or reporting obligations. The watchlist should therefore track guidance, investigation practice and the supplier's product evidence, without inventing a routine filing duty.[S10]
Textile extended producer responsibility requires destination-level work. The revised Waste Framework Directive introduces common rules under which Member States establish textile and footwear EPR schemes, with producers paying fees for products placed on the market under those schemes. The register should follow the actual national scheme, producer definition, route and fee conditions rather than assigning every African manufacturer a uniform EU registration task.[S15]
Claims need an equally careful status label. The Council describes ongoing work on separate Green Claims measures and records negotiations between the Council and Parliament. That proposal content should not be copied into the current-law register as an adopted verification regime. Current consumer-law duties still require review of the actual wording and evidence, while any future regime needs its own applicability check.[S18][S19]
Exhibit 61 orders the principal dates used in this paper without converting them into a single supplier deadline. Equal ordinal spacing is used to make the sequence legible, so distance between events carries no duration meaning. The Q4 2027 textile entry marks an indicative planned adoption window, not an application date. Every other entry still requires the scope analysis described above.[S11]
The sequence is a review aid, not a readiness countdown. GPSR started to apply on 13 December 2024 for covered non-food products sold online or offline, so current product safety cannot wait for the later sustainability milestones.[S34] The unsold-goods prohibition for large companies began on 19 July 2026, with its own exceptions and evidence conditions.[S14] These events affect different decisions and should not be merged into one circularity or product-passport task.
¶ 07 · 2 sources
GPSR started to apply on 13 December 2024 for covered non-food products sold online or offlineS34
A controlled change should produce an impact note before a schema or checklist changes. The note identifies what the new text alters, which products and roles are affected, which existing evidence remains valid, what requires legal confirmation and who authorises implementation. The reviewer then tests one affected product before scaling the change. If the source is only a proposal or an indicative plan, the note records preparation and monitoring rather than asserting a new legal duty.
¶ 08
14.53 min · 8 paragraphs · 1 exhibit
Measure outcomes that reveal the operating truth
Exhibit 628 paragraphs step it
Exhibit 62RegisterillustrativeEvery proposed metric needs a denominator and a decision
Measure
Evidence completeness
Definition
Complete applicable requirement-batch pairs / all applicable pairs
Decision
Investigate gaps; hold if critical
01 / 09
Read-out
Open the full register (9 rows)
Evidence completeness
Complete applicable requirement-batch pairs / all applicable pairs
Investigate gaps; hold if critical
Active retrieval time
Supplier work from clear request to controlled response
Find delays without rewarding wrong answers
Duplicate request hours
Repeated work on the same evidence for the same purpose
Clarify schema and test reuse
Overdue corrective actions
Open actions past their agreed review date / all open actions
Escalate owner and resource constraints
Worker-confirmed remedy
Closed cases with affected-person confirmation / all closed remedy cases
Keep unconfirmed outcomes open to review
Repeat incidents
Same defined failure recurring after case closure
Reopen cause and effectiveness review
Batch test traceability
Released batches with applicable linked test evidence / all released batches
Hold where critical evidence is absent
Prior change review
Released changed specifications reviewed beforehand / all released changes
Stop unreviewed release and repair the gate
Unresolved critical holds
Open critical product or worker holds at review
No release or closure by average score
Method and source
Illustrative GFT KPI dictionary. Sources support the control purpose and legal context, not any baseline, target or performance result. Applicable populations and periods must be fixed before calculation. Critical holds remain separate from all averages.
Unit: proposed KPI definitions, no observed values. Status: illustrative. Sources: S02, S05, S09, S16, S17, S33. Illustrative figures are teaching examples, not measured factory results.
Measurement should show whether the system changed a decision or an outcome. The numerator, denominator, population and period must therefore be visible. Evidence completeness means fully evidenced applicable requirement and batch pairs divided by all applicable requirement and batch pairs in the reviewed population. It does not mean documents uploaded divided by documents expected across an undefined list. The applicability register establishes the denominator before completeness is calculated.
Retrieval time should measure active supplier work from a sufficiently clear request to a controlled response, with recipient waiting time shown separately. Duplicate request hours should capture repeated collection, translation, checking or entry for the same evidence and purpose. These measures can test the administrative objective of a reusable pack. The VSME burden-reduction objective supports measuring that work, but it does not supply a factory baseline or guarantee a saving.[S05]
Worker measures need a stronger test than case closure. Worker-confirmed remedy is the share of administratively closed remedy cases in which the affected person, through a safe and appropriate route, confirms the agreed outcome. Repeat incidents should be counted for the same defined failure after closure. A falling complaint count may indicate improvement, fear, reduced access or a changing workforce, so it cannot stand alone as evidence of prevention. The ILO and IOE focus on identifying and assessing forced-labour risks supports enquiry and response rather than a simple grievance-volume target.[S09]
Product measures should retain the batch and change boundary. Traceable test evidence is the share of released batches linked to evidence covering the applicable question, sample and method. Prior change review is the share of released changed specifications with documented review before release. Withdrawn claims and unresolved critical holds should remain visible as counts with reasons, not converted into a favourable composite. Product-safety and chemical duties make those release boundaries consequential.[S16][S33]
Exhibit 62 defines the proposed measurement set. It deliberately supplies no baseline or target. A target set before observation could reward incomplete populations, rapid but wrong answers, premature case closure or release despite a critical gap. The first pilot should establish whether the measure can be calculated consistently and whether its decision rule protects the person and product.
The management review should read the measures together and preserve their contradictions. Faster retrieval beside more repeat incidents is not success. Higher evidence completeness beside an unresolved chemical hold is not release readiness. More worker-confirmed remedies may show that concerns are being addressed rather than that harm is increasing. The reviewer needs the underlying cases, population changes and evidence quality before interpreting direction.
Targets should be set only after the baseline has survived a definitions check. Confirm who records the event, whether sites and subcontractors use the same status meanings, whether the period is complete and whether restricted worker information can be reviewed safely. Where a denominator is too small or the case mix changes materially, report the cases rather than manufacturing a trend. A missing observation remains missing; it is not zero harm or full compliance.
14.62 min · 6 paragraphs · 1 exhibit
Integrate the new controls into the existing management cadence
The controls should enter the management routines that already direct production, people, quality and decisions. For this roadmap, GFT labels the cadence set up, launch, stabilise, scale and optimise, aligned with the programme's performance-management and digital-design logic. These are proposed planning stages, not a report that a client site has completed them. The existing programme remains a feasibility, performance-management system and digital-strategy reference rather than an implemented compliance service.
Exhibit 63 makes the dependency visible. Equal intervals separate decision states; they are not weeks, months, budget releases or promised completion dates. A later stage begins only when the preceding evidence gate has an authorised outcome. Legal change monitoring continues throughout, even though the diagram shows the main management sequence.
Exhibit 63ScheduleillustrativeThe roadmap advances through evidence gates rather than calendar promises
01 / 05
Read-out
Method and source
Illustrative GFT programme sequence. Equal numeric intervals from 0 to 5 separate decision stages and do not estimate implementation duration, resources or cost. Set up approves scope; launch applies current worker and product gates; stabilise repeats the bounded pilot; scale requires acceptance evidence; optimise uses controlled change and outcome review. No live implementation is claimed.
Unit: relative gate positions, not elapsed time. Status: illustrative. Illustrative figures are teaching examples, not measured factory results.
Set up finishes when the route, roles, requirement population and unresolved legal questions have named reviewers. Launch tests current worker-protection and product gates before an EU shipment commitment. Stabilise requires repeated retrieval, controlled correction and effective action on the bounded product family. Scale begins only after the specified buyer, legal, operational and worker-protection reviews are complete. Optimise then uses observed outcomes and verified legal change to revise the system.
¶ 01
The cadence should separate escalation from periodic review. A critical safety, chemical or worker concern moves immediately to the role with authority to protect or hold. Routine evidence quality, duplicate work and overdue actions enter the periodic operating review. Management examines unresolved holds, repeat incidents and handoff exceptions before considering scale. This prevents an urgent protection decision from waiting for a scheduled reporting meeting.
¶ 02
Each review needs a decision record, not only a dashboard. Note the population and period examined, the evidence challenged, the decision made, the owner, resources and next verification. Where a measure cannot be calculated, record the reason and repair the definition or source. Where a target is missed because the product mix or legal scope changed, preserve that explanation rather than adjusting the historic denominator to create apparent improvement.
¶ 03
Companion papers can supply the detailed engineering methods for factory measurement, provenance, resources, skills, circularity, data governance, footprinting and architecture. This compliance roadmap retains the duty-to-decision connection: which rule or request applies, what evidence answers it, who acts and what remains unresolved. The programme's Warp and Signal principle is practical here. Production events create the records, and measurement returns to change production rather than decorating a report.
¶ 04
14.71 min · 4 paragraphs · read with a related exhibit
Publish bounded claims and preserve the unanswered questions
External communication should use the same gate as product release. Publish the exact product, site, period, method and evidence boundary that a reviewer approved. Do not turn a pilot into a certification, a buyer review into regulator acceptance, an internal hold into an authority decision or a planned textile act into a current passport duty. Where wording is uncertain, narrow it or withdraw it. Current consumer-law application and the separate Green Claims negotiations make that status distinction especially important.[S18][S19]
The handoff register should preserve every unanswered question that could change a decision. Include unverified national implementation, final operative legal versions, product classifications, buyer-specific requirements, unavailable source records, actual operating costs and field evidence not yet collected. Give each item an owner, next action and reopening trigger. An explicit unresolved item is more useful than a confident statement that the research cannot support.
¶ 02
Management should receive a short decision pack containing the approved route, current critical gates, pilot definitions and observations, unresolved holds, worker-remedy status, legal watchlist and conditions for scale. It should separate sourced facts, derived arithmetic and illustrative scenarios. It should also record dissent where a reviewer does not accept the evidence or its interpretation. Scaling is a reversible management decision subject to new facts, not a reward for completing the previous phase.
¶ 03
GFT's recommendation is therefore to measure at source, act on harm, test the product, negotiate proportionate requests, protect personal information and review legal change through named owners. These are reasoned design recommendations as of 30 October 2026, not legal, tax or investment advice and not a compliance certification. The remaining task is to prove them in a bounded operating pilot while keeping every critical exception visible.[S04][S35]
Every knot is a source this chapter cites; select one to read its record, or filter the register by grade. Publication dates and inherited-source limits stay explicit. Read the paper's full evidence summary.
16 sources · 48 citation marks on this pageMost cited: S35, 6 times
Grade AGrade BGrade CKnot size: citations on this page
All 16 records shown
c14-S02 · Grade A
European Commission
Corporate sustainability due diligence
This Directive establishes a corporate due diligence duty for the very large companies that fall within its scope.
n.d.. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
European Parliament and Council of the European Union
Adopted sustainability reporting and due diligence amending directive text, PE-CONS 66/2/25 REV 2
Protected undertakings shall have the right to decline to provide information exceeding the information specified in the voluntary standards in response to a request made for the purpose of sustainability reporting as required by this Directive.
2026-02-24. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Move the shuttle along the rack to preview any chapter: its sections, exhibits and reading time. Draft web edition. This research is not investment, legal or tax advice; illustrative models are not measured factory results.
The paper's rack · 14 boltsBolt length: reading time
1.1
1.3
1.5
1.7
1.9
1.11
1.13
01Part I: Establish which duties reach the supplier
African suppliers should build evidence into production before promising compliance
GFT argues that a measured factory can answer changing buyer requirements without confusing documentation with the prevention of harm.
1.1The processing opportunity creates an evidence obligation
1.3The legal duty and the commercial request have different owners
1.5The programme is a design reference rather than a certified factory
1.7The strongest claim is the one the factory can reconstruct