The Supplier Evidence File: What a Due-Diligence Brand Will Ask Your Factory For

There is a version of this subject that gets written about a lot: the regulation, its thresholds, its phases, its penalties. It is a reasonable subject for a law firm and a poor one for a factory, because a factory is not regulated by it. What a factory experiences is a customer asking for documents.

That is the honest framing and it is worth stating plainly, because it changes what you do. A directive with an implementation date three years out does not create urgency. A purchase order clause that says supply is conditional on the supplier maintaining and evidencing certain standards creates urgency today — and those clauses already exist, because large brands began rewriting supplier terms as soon as the scope became clear.

So the practical question is not whether you are in scope. It is: if a buyer asked you this week for the file behind your claims, how much of it could you produce, and how long would it take? That is what the eight items below are for.

Item 1 — Certifications, With Their Scope

What good looks like: a current certificate for each standard you claim, naming the site it covers, together with the scope document that says what site activities it applies to. If your buyer asks about a specific product, the certificate should identify the article or the product class rather than only the legal entity.

What is usually missing: the scope. Factories commonly hold the certificate and not the scope annex, and the annex is where the limitations live — which processes are covered, which buildings, which product categories. A certificate presented without its scope invites the buyer to assume it covers everything, and that assumption is a liability for both parties.

How long it takes to assemble: an afternoon, if the certificates are current. Weeks, if any of them has lapsed, because re-issuance is not in your control. This is the single most common reason a supplier onboarding stalls, and the cure is a renewal calendar rather than a scramble.

Item 2 — A Traceability Trail That Names Sites

What good looks like: for a given order, a written chain from finished garment back through the factory, the finishing or dyeing step, the fabric mill, and where the fibre came from — with a named site at each stage rather than a country.

What is usually missing: the middle of the chain. Most factories can name the mill they bought from and the customer they sold to. Far fewer can name the dyehouse or finisher that the mill used, which is frequently a different site and sometimes a different province. That gap is exactly what a traceability request is designed to find.

What to do about it: build the chain map once, per material platform, rather than per order. If your cotton poplin always comes through the same two mills and the same finisher, that is one chain to document and keep current. Documenting it per order is why factories find this request unmanageable.

Item 3 — Social Compliance Audits and, More Importantly, the Corrective Actions

What good looks like: the audit report, and a corrective action plan with dates and evidence of closure for each finding. A buyer assessing a supplier is looking less at whether findings existed and more at whether they were addressed on a schedule.

What is usually missing: the closure evidence. Factories keep the audit and not the follow-up, so a finding from two years ago appears unresolved on paper when in practice it was fixed the following month. The fix is administrative and it is worth doing: a one-page corrective action log per audit, kept current.

Why it matters commercially: an unresolved finding reads identically to a hidden one, because the buyer cannot tell the difference. The paperwork you did not keep is indistinguishable from the work you did not do.

Related evidence that belongs with this item: wage and hour records, working-hours systems if you operate one, an age-verification process for hiring, and a documented grievance mechanism. Each of these is a process question rather than a document question, and buyers ask process questions second.

Item 4 — Subcontracting Disclosure

What good looks like: a written statement of which production steps are performed in-house, which are subcontracted, and to which named sites — with a commitment to notify before any change. Luxury operations such as hand beading are frequently subcontracted, and that is entirely normal; the failure is in not saying so.

What is usually missing: the disclosure itself. Subcontracting is often treated as a commercial secret or as a sign of weakness. It is neither. Buyers who discover an undisclosed subcontractor after the fact treat it as a trust failure rather than a capacity issue, and the commercial consequence is disproportionate to the actual risk.

The practical line: declare it in advance, name the site, and be able to evidence the same baseline standards there. A declared, audited subcontractor is a strength — it demonstrates real capacity. An undeclared one is the thing that ends relationships.

Item 5 — Capacity and Load Data

What good looks like: how many lines you operate, standard output, current utilisation, peak-season constraints, and how you sequence new business against committed orders. Not a claim to be able to handle anything — an honest picture of what fits and when.

Why a buyer asks: the question underneath capacity data is whether work will be subcontracted without warning when the schedule tightens. A supplier that can say how much it can take, and when, is answering the subcontracting question before it is asked.

What is usually missing: the peak-season picture. Off-season capacity figures make every factory look capable. The useful statement is what happens in the weeks when everyone is producing at once, and being straight about it costs less than being discovered.

Item 6 — Environmental Permits and Records

What good looks like: current permits for wastewater discharge and emissions where those apply, records of any treatment system, and a statement about chemical handling and storage. For a cut-and-sew operation the list is short; for anything involving wet processing it is longer.

What is usually missing: the currency of the documents. Permits expire and are renewed on a cycle that is invisible until somebody asks. As with Item 1, a renewal calendar solves it and a scramble does not.

A note on scope: a cut-and-sew factory's environmental footprint is concentrated in its fabric supply chain rather than its own floor. Saying that clearly is better than over-claiming about your own operations, because the buyer's next question will be about the mill.

Item 7 — Quality System, Not Quality Claims

What good looks like: a documented inspection regime that exists whether or not a buyer is watching — what is checked, at which point in production, by whom, against what tolerance, with what record. This is the item most often underestimated by factories and most valued by buyers.

Why it sits in an evidence file at all: because due diligence is not only about human rights and environment. A buyer's supply chain risk includes the risk of goods arriving wrong, which is a commercial and reputational exposure. A factory with a real inspection system is a lower-risk supplier than one with a better story.

What is usually missing: the records. Most factories inspect. Fewer keep inspection records that can be produced a year later, against a specific order, showing what was found and what was done. The inspection happened; the evidence of it did not survive.

Item 8 — Policies That Exist as Documents

What good looks like: a written code of conduct, an anti-harassment and grievance policy, a chemical management policy where relevant, and a named person responsible for each. Short documents with a signature and a date beat long documents with neither.

What is usually missing: the owner. A policy without a name against it is a policy that will not be maintained, and buyers have learned to test exactly this by asking who is responsible for it.

The realistic effort: this is the cheapest item on the list and the one most often left until last. A set of one-page policies with named owners can be assembled in a week, and it materially changes how an evidence file reads.

How the Request Actually Arrives

Nobody sends a factory a letter saying due diligence now applies to you. The requirement appears in four ordinary shapes, and most factories meet all four within a year without connecting them.

1. A self-assessment questionnaire. A form, usually long, asking about policies, records, subcontracting and permits. It is the first thing to arrive and the easiest to under-prepare for, because it looks administrative.
2. A supplier code of conduct to sign. A commitment document with annexes. Signing is quick; being able to evidence what you signed is the work.
3. Onboarding documents for a new customer. Requested as part of setting up the account, before any order, and often handled by whoever manages new accounts rather than by whoever holds the records.
4. An annual re-declaration. The same questions again, twelve months later, with a shorter deadline. This is the one that catches factories out, because the first year's scramble is forgotten by the second.

The pattern to notice is that the request is persistent and repeating. A file assembled as a one-off project will be out of date by the second request, and a file that is out of date is worse than a file that is thin, because it makes a claim the factory cannot currently support. The maintenance is the discipline; the first assembly is just the entry cost.

There is also a version of this that arrives through a customer's own customer. A brand supplying a retailer will pass the retailer's requirements down, and those may be stricter or differently shaped than the legislation. It is worth asking a new buyer which standard they are being held to, because it determines what you will be asked for next.

Where Factories Get Stuck

Five recurring blockers. None of them is technical, and all of them are administrative.

• The records exist but are not findable. Certificates in a folder on someone's desktop, audits in an email thread, inspection records in a paper file in the QC office. The evidence is real and effectively unavailable, which from the buyer's side is the same as absent.
• One person holds the whole picture. The person who knows where everything is becomes a single point of failure, and their departure or promotion converts a manageable request into an archaeology project.
• Certificates have lapsed quietly. Nothing notifies you. A certificate expires and the first anyone knows is a buyer's audit of the file, at which point lead time is not yours to control.
• The chain map was never written down. The factory knows its mills by relationship rather than by record, and the finisher is a phone number rather than a named site.
• Subcontracting is undisclosed by habit. Not for any bad reason — frequently because it has always been done that way and nobody thought it was the buyer's business.

The common thread is that a factory can be entirely compliant in practice and still unable to evidence it. That distinction is uncomfortable but it is the whole game: the evidence file is not a measure of how good a factory is, it is a measure of how well a factory can be audited. Both matter, and they are not the same skill.

Turning the File Into a Sales Argument

Read from the other side of the table, this list is a competitive position rather than a compliance burden, and the factories that worked that out early are using it.

• A fast, complete response shortens the onboarding cycle. New customer setup is a notorious bottleneck. A factory that returns a complete file in a week gets orders earlier than one that takes a month.
• Disclosed subcontracting reads as capacity, not as risk. A brand choosing between two factories of similar quality will prefer the one that can tell it where everything is made.
• Corrective action closure evidence signals a working system. Buyers have learned that the closure record predicts behaviour far better than the audit score does.
• Traceability supports premium positioning. A brand making a fibre or origin claim needs a supplier who can document it, and will pay for that capability.
• It reduces the cost of every subsequent audit. The second, third and tenth audits reuse the same file rather than rebuilding it.

None of that requires the factory to be larger, greener or better equipped than it is. It requires the factory to be legible. Legibility is cheap and it is chronically under-supplied, which is exactly why it is worth having.

A Ninety-Day Order of Work

If none of this exists yet, the sequence matters more than the speed. Thirty days per block, and each block produces something usable on its own.

Then test it. Take a live order, assemble the full pack for it, and time how long it takes. Whatever made it slow is the next thing to fix — and whatever made it slow is almost always a filing problem rather than a missing document. Do the ninety days once and the file stops being a project; it becomes a folder that happens to be current, which is the state you want to be in before a buyer asks rather than after.

A Readiness Self-Check

Twelve questions. Answer them as a supplier would, not as a salesperson would, and the gaps become obvious.

The last question is the only one that really matters. A file that exists in principle and takes three weeks to produce is, from the buyer's side, a file that does not exist. Run the test once on a live order, time it, and fix whatever made it slow — and expect the answer to be filing rather than missing documents, because it almost always is.

The Questions Buyers Ask Second

Documents get you through the first review. What decides the outcome is usually the conversation that follows, and it is predictable. Three questions recur.

Who is responsible for this?

For each item, a name. Not a department, not a job title in the abstract — a person who can answer a follow-up question and who knows the file exists. A supplier that answers this confidently signals that the documents are maintained; one that has to think about it signals that they were assembled once, for this request, by somebody who has now moved on.

What has changed since the last review?

This is where a re-declaration earns its keep. A supplier that can name the changes — a new subcontractor, a renewed permit, a closed corrective action, a certification that lapsed and was replaced — is demonstrating a live system. A supplier that reports nothing has either a static operation or a file nobody updates, and buyers know which is more likely.

How would you know if something went wrong?

The hardest of the three and the most revealing. It asks about internal detection rather than internal compliance. A factory with a grievance mechanism, an inspection record and a corrective action log can answer it. A factory relying on the buyer to notice cannot, and that answer is remembered even when nothing is wrong.

Preparing for those three questions is a different exercise from assembling documents, and it is the one that separates a supplier who passes an audit from a supplier who is trusted. Both are worth having; the second is worth more and costs less to build.

FAQ

Are we actually in scope of the directive?

Almost certainly not directly — the thresholds are set for very large companies. But scope is the wrong question. The right one is whether your customers are in scope, because they will pass the requirement to you contractually regardless. Expect the request through a supplier agreement rather than a regulator.

What does it cost to assemble this file?

Almost nothing in money and a meaningful amount in attention. Nearly every item is a document that already exists somewhere and needs to be found, dated, scoped and filed so it can be produced on request. The expense is in the first pass; after that it is maintenance.

Can we simply decline to provide some of it?

You can, and buyers increasingly treat a declined item as a finding. The two items worth protecting commercially — price breakdown and customer names — are not on this list, so declining the rest tends to cost more trust than it protects value.

We are a small factory. Is this proportionate?

The file scales with your operation. A cut-and-sew workshop with two subcontractors has a short chain to map and a limited permit list. The shape of the answer is the same at any size; only the number of pages changes.

Where should we start if none of this exists?

With the document inventory. Collect every certificate, permit and audit report you hold into one folder and record two things against each: what it covers, and when it expires. That single exercise usually reveals a lapsed certificate and produces a renewal calendar, and both of those matter more than anything else on the list.

Only one customer is asking. Do we build the file for them?

Build it once and use it for everybody. The items are near-identical across buyers, so a file assembled for one request answers the next five with light reformatting. A file assembled only for the customer who asked gets rebuilt from scratch next time, and that is the version that costs money.

What does the buyer actually do with it?

Files it as evidence for their own reporting and due-diligence obligations, and uses it to decide whether you are a supplier they can keep. The second use is the one that matters commercially: this file is now part of how suppliers are ranked. A complete file does not win an order on its own, but an incomplete one quietly removes you from the shortlist before the conversation about capability ever starts.

Working With Luxudress

Luxudress is the factory-direct front end for womenswear production across production facilities in Guangzhou and Dongguan, covering development, sampling, material sourcing, manufacturing, inspection and export packing. Our compliance position is ISO 9001 · OEKO-TEX STANDARD 100 · amfori BSCI-audited · SGS-inspected.

On a due-diligence request we will give you the certificates with their scope documents rather than the certificate alone, name the sites in the chain rather than the countries, declare which operations are subcontracted and where, and produce inspection records against a specific order rather than a general description of our process. Where an item is not available yet, we will say so and say when.

If you are assembling an evidence file for a customer, send the request you have received. You can also see how we run inspection, read how to verify a certificate, or check the questions worth asking any factory.

Request a quote and we will return a costed plan including the documentation it assumes.