Pre-Contract Audit of Chinese Supplier: Legal Compliance + QC Walkthrough

Don’t run a pre-contract audit as one big trip and one big report. Run it as a fifteen-working-day project with two lanes that move in parallel. Legal due diligence on one lane. QC walkthrough on the other. They sync twice. They feed the contract draft once. If you treat it as a single line — fly to the factory, walk the floor, sign on the way home — you will keep losing six-figure orders to suppliers who passed the wrong test on the wrong day.

The buyers who lose the most money on Chinese sourcing are not the ones who skipped the audit. They are the ones who sequenced it wrong. Most of them paid for the QC inspection first, were impressed by what they saw, and only then ran a half-hearted entity check. By the time the legal lane returned a problem — a trading shell instead of the manufacturer, a trademark sitting on a competitor’s name, a court-default flag in Zhejiang — the deposit was already on its way and the production slot was already booked.

This guide is a project plan, not a checklist. It is the calendar, the team structure, and the synchronisation rules we use at XILINK when a buyer in Almaty, Tashkent, or Aktobe asks us to run a pre-contract audit on their behalf before they wire a deposit on a six-figure order. It is also, as far as I know, the only public write-up that treats legal and QC as two parallel streams reporting into a single contract write-back step. If you are doing this work yourself, the timetable below is what to copy. If you are paying an agent, the timetable below is what to demand they show you on day zero.

Factory audit in China — what we check on-site, day rate from $60, report in 48 hours.

The Pre-Contract Audit Is a Two-Lane Project, Not a Site Visit

A pre-contract audit answers two completely different questions about a Chinese counterparty, and the questions cannot be answered by the same person on the same day. The first question is structural: is the entity on the contract the same entity that owns the factory, the trademark, the export licence, and the bank account that will receive your wire? The second question is operational: can that entity, at that physical address, on that production line, with that staff, build what your specification says, to the tolerance the contract demands, in the timeframe written into the schedule?

These are different evidence chains. The first one lives in registries — gsxt.gov.cn, the China National Intellectual Property Administration, China Judgements Online, the Sinosure credit file, the General Administration of Customs declarant database. The second one lives on a shop floor — calibration stickers, mill test reports, weld macros, finished-goods serialisation, ISO 9001 audit logs. The first one is paperwork. The second one is concrete dust on your boots.

The trap is that the two questions sound related, and so buyers and budget owners assume they can be answered together. They cannot. A trademark lawyer cannot tell you whether a hydraulic press is being assembled correctly. A QC engineer cannot tell you whether the legal owner of that factory has been quietly delisted from the AEO registry. Even the people who can do both — and there are some — should not, because doing them in parallel takes fifteen working days and doing them sequentially takes twenty-eight.

We split every pre-contract audit into a Legal lane and a QC lane on day zero. Different teams, different deliverables, different reporting lines. They synchronise twice — once at the end of week one, once on day twelve — and the only point at which their findings combine is the contract write-back session on day fourteen. Below is exactly how that timetable looks in practice.

Day-by-Day Timetable: 15 Working Days from Brief to Signature

Most importers we work with want to know what fifteen working days actually contain when we say a pre-contract audit takes “about three weeks.” Here is the calendar in detail. Every line item is something we deliver against, on a date, to a named owner. If you cannot fit your own audit into a structure like this, you do not have one. You have a hope.

Days 0 – 1: Brief and Lane Split

On day zero, the buyer’s project lead, our XILINK project manager, the lead lawyer for the Legal lane, and the lead engineer for the QC lane sit in a single call. Total time: ninety minutes. Deliverables out of that call:

• A signed scope-of-work that lists the entity name, the USCC, the product family, the destination country, the contract value range, and the delivery deadline.
• A risk-tier rating: standard, elevated, or critical. Critical-tier audits add an extra three days to the QC lane and a Sinosure-plus-Coface cross-check on the Legal lane. We tag a contract as critical if the value is over USD 250,000, the destination is a regulated industry (medical, pressure vessels, lifting equipment), or the supplier has a registered capital below CNY 5 million.
• A division of who answers what: Legal owns entity, IP, customs, tax, court history, sanctions screening, banking. QC owns line, materials, processes, equipment, ISO and EAC traceability, sample testing.
• A communication channel: a single shared spreadsheet on the buyer’s drive, with two tabs — Legal lane and QC lane — both visible to everyone but only writable by the relevant team.

Day one is when each lane builds its own day-by-day work plan inside the master timetable. By the end of day one the buyer has on screen the same picture both teams are working from.

Days 2 – 5: Legal Lane Sprint, QC Lane Site Booking

On the Legal lane, days two through five are pure registry work. Pulled in this window:

• Full business licence (营业执照) verification on gsxt.gov.cn against the contract entity, including the Unified Social Credit Code, the registered name, the registered address, the legal representative, and the scope of business.
• Foreign trade operator registration on mofcom.gov.cn and customs declarant registration on customs.gov.cn for both the contract entity and, if different, the actual exporter named on the proforma invoice.
• VAT general-taxpayer status, latest quarterly tax return, and confirmation that the entity is not on the tax irregularity registry (税务异常名录).
• Trademark search at the China National Intellectual Property Administration (CNIPA) on every brand, model, and graphical mark that will appear on the goods, on the packaging, and on the EAC certificate. Patent search on any technology the supplier claims as proprietary.
• China Judgements Online history (wenshu.court.gov.cn) for the entity as both plaintiff and defendant, with a five-year window. We flag every contract dispute, every IP dispute, every tax case, and every enforcement order.
• Sinosure preliminary credit pull. If the entity is uninsurable or the cover offered is below 40% of the contract value, that fact alone usually changes the deal structure.

The QC lane in days two through five does not enter the factory yet. It does pre-work that the buyer almost never sees as audit work but which determines whether the visit is worth doing:

• A factory pre-questionnaire — about 60 questions on equipment list, calibration schedule, ISO and EAC scope, sub-contractor map, raw-material suppliers, recent third-party audit reports, top three customers in CIS / EAEU.
• A documentary review of any prior third-party reports the supplier has on file — SGS, BV, QIMA, TÜV. We are not looking for a stamp. We are looking for the line items the previous auditor did not test.
• An equipment matrix — what the supplier claims they have versus what the brochure pictures show versus what the satellite image of the registered address shows. Three out of every ten suppliers cannot pass this step, because the registered address turns out to be a residential building or a 200-square-metre rented unit on the third floor of a wholesale market.
• A specification gap analysis on the buyer’s drawings, comparing the buyer’s tolerance band against industry-standard manufacturing capability for that product class. If the buyer’s drawing demands a tolerance the supplier’s machine class cannot achieve, the contract has to specify how that gap is closed before it is signed.
• A formal site-visit booking with a two-day window, designed so the supplier knows we are coming but does not know on which of the two days. The whole point is that the line is real on both dates.

Day 5: First Sync — Go / No-Go Checkpoint

End of day five, the two lane leads and the buyer’s project lead meet for thirty minutes. Three possible outcomes:

1. Green. Legal lane has cleared registries with no critical flags. QC lane has cleared documentary pre-work and the visit is scheduled. Both lanes proceed.
2. Amber. Legal lane found a flag that needs resolution before the QC lane spends money on a flight (for example, a trademark held by a third party, or a court default that the supplier had not disclosed). The QC visit is held until the flag is closed.
3. Red. Legal lane found something fatal — a revoked entity, a fraudulent business licence, a court-blacklisted legal representative, or a supplier name that does not exist in any public registry. The audit stops here. Total spend at this point is roughly USD 200 to USD 400 on Legal and a few hours of QC documentary work. Cheap insurance against the alternative.

In the 47 audits we ran across 2024 and the first three quarters of 2025, day-five outcomes broke down 64% green, 23% amber, 13% red. That 13% red rate is the entire reason the Legal lane runs first. Stopping a deal at day five costs three hundred dollars. Stopping it at day fifteen costs twenty thousand and a missed delivery window.

Days 6 – 10: Legal Deepening, QC Site Visit

On a green or recovered-amber project, days six through ten are when the bulk of the engineering work happens. The Legal lane drops into deeper checks:

• Sinosure full credit report and Coface or D&B if the contract value justifies the extra USD 200.
• Bank account verification — the bank named on the proforma invoice must be a real corporate account at a real Chinese bank, in the same legal name as the contract entity. We have caught two cases in 2025 alone where the SWIFT instructions on the invoice routed to a Hong Kong personal account in a different name. That is not a translation error. That is the supplier hoping the buyer will not check.
• Sanctions and PEP screening. For buyers in jurisdictions with their own regimes (Russian buyers under EU and US sanctions, Kazakh buyers under EAEU controls), every entity in the supplier chain — manufacturer, exporter, freight forwarder, paying bank — gets screened against the relevant sanctions lists.
• Affiliate mapping. We pull the list of related parties from Tianyancha or Qichacha — same legal representative, same registered address, same shareholders — and run the same court and tax checks against them. A clean entity with three sister entities, all of which are on court-default lists, is itself a flag.

On the QC lane, days six through ten are the actual factory visit and the lab work:

• Day one on site: walk the line in production. Not a curated tour — we choose where we walk, we ask which order is being made on each machine, we check the work-in-process tags against the production schedule.
• Day two on site: pull samples. Random samples from the finished-goods warehouse for dimensional check, hardness test, weld macro, electrical test (depending on product class). Samples are sealed and shipped to an independent lab if the supplier’s in-house QC equipment is the only thing available — because an in-house lab on the supplier’s payroll cannot, by definition, audit the supplier.
• Days three to five: lab results, photographic documentation, draft QC report. The report has to include not only what was tested but what was not tested and why, so the contract write-back step on day fourteen knows what to bake into the warranty clauses.

Day 12: Second Sync — Cross-Lane Comparison

This is the meeting that almost no buyer runs and almost every audit failure can be traced back to. Day twelve, two hours, both lane leads in the room with the buyer. The agenda is exactly one thing: compare findings.

The reason this meeting matters is that some of the worst risks are invisible to either lane on its own. A supplier whose business licence is clean and whose factory floor is excellent can still be uninsurable, can still have a trademark conflict that will arrest the goods at destination, can still have a sub-supplier on a sanctions list. A supplier whose Sinosure cover is unusually generous can still be assembling at a sub-contracted location three hundred kilometres away.

The Legal lead presents the consolidated entity profile and flags. The QC lead presents the consolidated capability profile and findings. The buyer’s project lead asks one question: do these two pictures describe the same company? In about 8% of our cases the answer is no, and that is exactly the case where everything else looked fine until day fifteen.

Days 13 – 14: Contract Write-Back

This is the step almost no buyer treats as part of the audit, and it is the step that converts the audit’s findings into a contract that can actually be enforced. Every flag that the Legal lane raised, every weakness the QC lane identified, gets translated into a contract clause. We call it write-back because the contract is being rewritten with the audit’s findings written into it.

A few examples of what write-back looks like in practice:

• Legal lane finds the registered capital is paid-in only to 30%. Write-back: payment terms shift to 30/40/30 instead of 50/50, with the final 30% released against third-party PSI.
• QC lane finds the supplier’s in-house weld macro process is on the borderline of the spec tolerance. Write-back: a clause requiring weld macros from every batch, archived for two years, with replacement-on-failure language.
• Legal lane finds the trademark on the goods is held by an unrelated party. Write-back: either remove the trademark from the goods entirely, or get the right party signed onto the contract as a co-seller, or kill the deal.
• QC lane finds that the EAC certificate’s serial-number range is narrower than the contract quantity. Write-back: a clause that the supplier is responsible for issuing an extension of the certificate before shipment, at the supplier’s cost.

Write-back is what turns the audit into something the buyer’s lawyer can actually use in arbitration. Without it, the audit is just an opinion. With it, the audit is a structured set of contractual obligations the supplier is now bound to.

Day 15: Final Report and Decision

The output is a single document, around 25 to 40 pages, with a one-page executive summary and an explicit recommendation: sign as drafted, sign with the write-back clauses included, renegotiate, or walk. The buyer signs or does not sign. If they do, the contract that gets executed is the post-write-back version, not the pre-audit version.

Two Teams, Two Skill Sets: Why You Cannot Run This With One Person

The biggest cost saving buyers think they can find on a pre-contract audit is to use one person for both lanes. This does not work, and it does not work for a reason that is not about cost: it is about evidence quality.

A trademark dispute and a weld defect cannot be diagnosed by the same person, not because the person is not smart enough, but because the registries and the shop floor are different evidence environments. The Chinese-language entity registries require a specific kind of legal Chinese literacy — being able to read 注销 versus 吊销 versus 暂停 in a status field is not an English-translation skill, it is a procedural one. The shop floor requires hands-on judgement about whether a calibration sticker is current, whether a pressure-test rig is plumbed correctly, whether a weld macro is actually being read from the correct micrometre.

In practice the Legal lane on a normal engagement consumes 12 to 20 hours of paralegal time, four to six hours of senior legal review, and one to two hours of IP-counsel time on trademark and patent searches. Total cost in our model is roughly USD 600 to USD 900. The QC lane on the same audit consumes one to two days of an engineer on site, eight to twelve hours of report preparation, and any independent lab fees. Total cost roughly USD 700 to USD 1,400. The two lanes together come in under USD 2,000 for a contract that protects six figures.

Splitting the cost out by hours and skill makes another point: if a single agent quotes you USD 600 for “a full audit,” they are doing one of the two lanes, not both. Usually they are doing the QC walkthrough, sending you photos of a factory, and skipping the legal lane entirely. That is exactly the structure that lost a Shymkent contractor USD 240,000 to a Shenzhen trading shell in early 2025 — an event we covered in detail in our 12-point pre-contract counterparty checklist, which is the document layer of the same approach. The shop-floor layer of the same approach is in our chinese supplier checklist from our audit team. Both of those guides go deep on what each lane checks. This guide is about how the two lanes work together inside a project plan.

To make the work split concrete, here is the breakdown we use internally on a typical USD 200,000 industrial-equipment audit. Same supplier, same destination, same product. Just the time and cost split between the two lanes.

Activity	Lane	Hours	Done where	Cost band
Entity verification on gsxt.gov.cn	Legal	2	Remote	USD 30
Foreign trade and customs registration	Legal	2	Remote	USD 30
VAT and tax irregularity check	Legal	1	Remote	USD 20
Trademark and patent search at CNIPA	Legal	4	Remote	USD 200
Court history at China Judgements Online	Legal	3	Remote	USD 60
Sinosure and Coface credit pulls	Legal	2	Remote	USD 200
Sanctions and PEP screening	Legal	2	Remote	USD 80
Affiliate mapping and cross-checks	Legal	4	Remote	USD 80
Bank account verification	Legal	2	Remote	USD 40
Legal lane total		22	Remote	USD 740
Pre-visit questionnaire and documentary review	QC	4	Remote	USD 120
Equipment and address matrix	QC	3	Remote	USD 90
Specification gap analysis	QC	4	Remote	USD 120
Site visit, day 1 — line walk	QC	8	On site	USD 280
Site visit, day 2 — sample pull	QC	8	On site	USD 280
Independent lab tests	QC	–	Lab	USD 250
Photographic documentation and report	QC	8	Remote	USD 240
Travel and lodging	QC	–	–	USD 200
QC lane total		35 + lab	Mixed	USD 1,580

The Legal lane is mostly remote; the QC lane is mostly on site. The two lanes overlap for roughly six working days in the middle of the project. A buyer trying to compress this into a single trip is paying the QC lane cost without the Legal lane cost — and getting half the audit.

In-House Versus Outsourced: When to Run This Yourself

Buyers ask whether they can run this work themselves rather than pay an agent. The honest answer is: parts of it, yes; the whole thing, almost never. The split looks like this.

Activity	Realistic for in-house buyer	Why or why not
Business licence and USCC verification	Yes	Free, public, takes 10 minutes once you know where to look
Foreign trade and customs registration	Yes	Free and public, in Chinese — translation tools handle it
Trademark and patent searches at CNIPA	No	Requires Chinese-language IP counsel; mistakes are catastrophic
China Judgements Online	Partially	Free but the case-law density requires legal training to interpret
Sinosure and Coface	Partially	Coface report is purchasable from outside China; Sinosure requires a Chinese intermediary
Sanctions screening	Yes	Standard tools (LexisNexis, Dow Jones) work from any jurisdiction
Pre-visit questionnaire and equipment matrix	Yes	If you have an engineer on staff who can read industrial Chinese
Site walk	Partially	Possible if you can fly to China yourself, but you lose the unannounced-visit advantage
Sample pull and lab tests	No	Lab logistics are easier to coordinate from inside China
Contract write-back	Yes	This is your lawyer’s job and it is the highest-value step

The pattern is that the cheap, public-data parts of the Legal lane are realistic to do in-house. The expensive, judgement-heavy parts of both lanes are not. Most of our buyers run the entity, customs, and tax checks themselves on day zero as a free pre-screen, and only engage us for the deeper Legal work, the on-site QC, and the contract write-back. That pre-screen filters out about a third of suppliers before any money changes hands. The rest of the project is what we get paid for.

For the on-site QC step in particular, the comparison of independent inspection providers — QIMA, SGS, Bureau Veritas, freelance auditors, and agent-based engineers — is covered in detail in our 2026 china quality auditor pricing and scope guide. The takeaway from that piece is that a name-brand inspection report is not a substitute for the parallel Legal lane work; it is a single deliverable inside a much larger plan.

Case 1: The 15-Day Audit That Saved a USD 310,000 Order in Karaganda

In June 2025 a Karaganda contractor signed a non-binding letter of intent for USD 310,000 worth of pre-engineered steel building components, scheduled for shipment to a mining-services site near Zhezkazgan. The supplier was based in Hebei, had a Gold Supplier badge on Alibaba, had three years of declared exports to Russia, and had passed a previous QIMA inspection on a smaller 2024 order from a different buyer.

We ran the standard fifteen-day audit. The QC lane came back almost immaculate: the production line was real, the welds were within spec, the steel mill test reports tied back to a credible upstream mill in Tangshan, the EAC certificate was valid and resolved on the Russian Federal Service for Accreditation registry. The Legal lane, in parallel, found three problems in two days:

The contract entity name was different by two characters from the gsxt.gov.cn record — close enough to look like a typo, but legally a different company with a different USCC. The actual factory was owned by a third entity that had no contractual link to either of the first two. And the trademark on the building components, which the supplier had been using in its catalogue for two years, was registered to a Tianjin company that the supplier’s legal representative had been a director of until eighteen months earlier and had since fallen out with.

If the buyer had run only a QC visit, none of that would have come up. The factory was real. The product was good. The trap was that the goods, if shipped, would have arrived at the Russian border with a trademark held by an estranged former associate who was actively in dispute with the supplier in a Tianjin court. The most likely outcome was an enforcement action at the EAEU border on the trademark claim alone, with the goods seized and our buyer’s USD 310,000 already wired.

Day twelve sync flagged it. Day fourteen write-back rewrote the contract: the trademark was removed from the goods entirely, the contract entity was changed to the actual factory rather than the trading shell, and the payment terms moved to a 20/40/40 split with the final 40% conditional on a clean export declaration from the factory’s own customs broker, not the original entity’s. The buyer signed at the end of day fifteen on the rewritten terms. The supplier accepted the changes, which itself confirmed that the original structure had been opportunistic rather than strategic. Goods shipped on schedule, cleared the EAEU border without incident, and the building was commissioned in October.

Total audit cost: USD 1,820. Order saved: USD 310,000.

Case 2: The Audit We Stopped on Day Five — and What It Cost Not to Continue

In November 2024 a Tashkent buyer brought us a quotation for USD 95,000 of bench-top CNC equipment from a supplier in Guangdong. The buyer had already been on a video call with the supplier, had been impressed by the workshop tour, and wanted us to do a “fast” pre-shipment check before they wired the deposit.

We refused to run it as a pre-shipment check and ran it as the standard fifteen-day pre-contract audit instead. The Legal lane on day three returned three findings:

The supplier entity had been registered for fourteen months. Its registered capital was CNY 100,000, paid in 0%. Its scope of business read “wholesale and import-export” with no manufacturing terms. The legal representative was on the court defaulters’ list (zxgk.court.gov.cn) for an unsatisfied judgment of CNY 1.4 million from a 2023 contract dispute with a German buyer. And the WeChat-only contact pattern matched four other “factories” we had seen in the previous twelve months, all of which had since vanished.

We stopped on day five. Total cost so far: roughly USD 380. We told the buyer not to proceed. The buyer was reluctant — they had already spent three weeks negotiating, the price was attractive, and the workshop tour video had been convincing. They asked whether we could “just do the QC visit anyway.” We declined. There is no QC outcome that fixes a court-defaulter legal representative.

Six weeks later the buyer told us they had pushed back to the supplier on price one more time, the supplier had become evasive, and the buyer had walked away on their own. In January 2025 a different buyer in Almaty, who had been negotiating with the same supplier in parallel, sent USD 87,000 deposit. The supplier disappeared within the month. The Almaty buyer never recovered the money. The court-defaulter flag we found on day three of our Tashkent audit was already public on the day the Almaty buyer sent the wire.

The point of the case is not that day five always saves a deal. It is that day five is the point at which the cost of stopping is at its lowest and the cost of continuing is at its highest. A USD 380 day-five exit is dramatically cheaper than a USD 87,000 disappearance.

How the Two Lanes Feed the Contract: Five Clauses Buyers Almost Never Write

Once you have run both lanes, you have evidence the contract draft does not yet reflect. The contract write-back step is where evidence becomes enforceable. Five clauses we recommend on every engagement, drawn directly from real findings:

Clause 1: Entity-and-bank consistency. The buyer’s payment is only valid if the receiving bank account is in the exact registered Chinese name of the contract entity, at a Chinese bank inside mainland China, with the SWIFT details matching the proforma invoice character for character. Any change requires a written addendum signed by both legal representatives. This single clause defeats the most common scam pattern: a last-minute “updated” SWIFT detail emailed shortly before payment.

Clause 2: Manufacturing-location lock. The contract names the registered factory address where the goods will be produced. Any sub-contracting requires written buyer approval, and the buyer reserves the right to inspect the actual production location with 48 hours notice. This defeats the bait-and-switch pattern where the QC visit happens at one factory and production happens at another.

Clause 3: Trademark warranty. The supplier warrants that all trademarks, trade dress, and IP markings on the goods, the packaging, and the certificates are owned by the supplier or licensed under a written agreement that survives until at least 18 months after delivery. The supplier indemnifies the buyer against any third-party trademark claim at the destination border. This defeats the trademark-conflict pattern.

Clause 4: Serial-number traceability. Every unit of finished goods carries a unique serial number that ties back to the supplier’s production logbook, the relevant batch’s mill test reports, and the EAC or destination certificate range. Any unit whose serial number falls outside the certificate range is non-conforming and rejectable at delivery. This defeats the EAC-certificate-mismatch pattern.

Clause 5: Lab-result release condition. Final payment is conditional on the buyer’s receipt of pre-shipment lab results — independently sampled, sent to an independent lab — confirming spec compliance on dimensions, materials, and any safety-critical performance. The lab is named in the contract and pre-paid by the buyer. This defeats the in-house-QC-rubber-stamp pattern.

These five clauses cost nothing to add to a contract. They cost the buyer nothing extra in audit fees. They each map to a specific finding the audit lanes either confirmed or ruled out. And they each turn a known risk into a written contractual right that an arbitrator can actually enforce.

FAQ

How long does a pre-contract audit of a Chinese supplier really take? Fifteen working days is the standard timetable when both lanes run in parallel. Sequential — Legal first, then QC — takes twenty-five to twenty-eight working days. We have done compressed audits in nine working days for buyers with hard deadlines, but the compression always costs deliverable depth somewhere; usually we drop the affiliate mapping and the second sample pull. Anything advertised as a “three-day pre-contract audit” is almost certainly a single-lane site visit relabelled.

How much does a pre-contract audit cost on a six-figure order? For an order in the USD 100,000 to USD 250,000 range, the all-in cost of both lanes done properly sits between USD 1,500 and USD 2,500. Legal lane runs USD 600 to USD 900. QC lane runs USD 700 to USD 1,400 plus travel. As a percentage of contract value the audit comes in around 1% to 2.5%, which is the same rule of thumb we apply to commercial trade-credit insurance. Below USD 50,000 contract value, we recommend a Legal-only audit (USD 600 to USD 800) with the QC lane replaced by a structured remote-call review.

Can the buyer skip the QC lane and only run the Legal lane? Only on a known supplier with a multi-year history. For a first contract with a new supplier, skipping the QC lane means accepting that you cannot tell whether the entity you have just legally cleared can actually produce what your contract describes. We have seen perfectly clean entities with no production capacity to speak of. Equally, skipping the Legal lane for a “quick QC trip” is the more common and more expensive mistake — see Case 1 above.

Why do the lanes have to run in parallel rather than sequentially? Two reasons. The first is calendar — sequential adds two weeks to a project that already has a delivery deadline pressing on it. The second is information — the day-twelve sync only adds value if both lanes have substantively progressed by that date. If you run Legal first and finish before QC starts, you lose the ability to spot the cross-lane discrepancies that catch the worst risks. The whole point of the two-lane structure is that the two views check each other.

What happens if one lane finishes early or finishes late? The Legal lane is the one that more often finishes early, because registry checks are predictable. When that happens, the team uses the slack to run the affiliate-mapping and sanctions cross-checks more deeply, or to pull a second Coface report from a different angle. The QC lane more often finishes late, especially when lab results are delayed or when a sample pull reveals a defect that requires a follow-up sample. We move the day-twelve sync to whatever working day actually catches both lanes at a substantive midpoint, even if that means day thirteen or day fourteen. The fifteen-day calendar is a target, not a contract.

Who owns the audit report after delivery? The buyer owns it. We deliver one master copy to the buyer’s project lead and retain one redacted copy for our own quality records. We do not share findings with the supplier under any circumstance, even if the supplier asks. Sharing findings with the supplier defeats the purpose: an audit’s value is the asymmetry of information it gives the buyer.

Can a supplier refuse to be audited this way? Yes, and it happens. Roughly 7% of suppliers in our 2024-2025 dataset refused at the day-zero stage when they understood that both lanes were going to run. Of those refusals, nearly all came from suppliers we would have flagged on the Legal lane within the first three days anyway. A supplier confident in their own legal and operational position has no reason to refuse this kind of audit — the supplier is gaining a buyer who has done their work and will sign with conviction. Refusal at day zero is itself a finding.

How does this differ from a normal third-party inspection report? A normal third-party inspection — QIMA, SGS, BV — covers what we would call the QC lane only, and within that, mostly the production-line and finished-goods steps. It does not run entity, IP, court, or sanctions checks. It does not write back to the contract. Some of the brands offer “supplier verification” add-ons, which are partial Legal-lane work, but they are usually a fraction of a real Legal-lane sprint. Treat third-party inspection reports as one input to the QC lane, not as the audit itself.

If you are about to sign a contract with a Chinese supplier for an order over USD 50,000 and you have not run both lanes in parallel, stop. Day zero costs nothing. Day five costs less than USD 400. Day fifteen costs about 1% to 2.5% of your contract. Walking away on day five with USD 400 spent is the cheapest insurance product in cross-border trade.

If you are reading this because you have already lost money on a Chinese supplier and are trying to make sure the next one does not repeat the pattern, the fifteen-day timetable is what we run for buyers in your situation every week. We split the work between our XILINK legal team in Shenzhen and our engineering team in Jiangsu, we report to the buyer on a single shared timetable, and we deliver a contract draft that has every audit finding written into a clause an arbitrator can enforce. Send us the supplier name, the USCC if you have it, and the proforma invoice. We will tell you, within forty-eight hours, whether the audit is worth running and what it will cost.

The cost of running this audit is your attention and around USD 1,500 to USD 2,500. The cost of skipping it is the contract value plus the project delay plus the lost commissioning window. The buyers who come back to us a second time are the ones who have already paid the second cost once.

Related: our complete guide to a factory audit in China — on-site checks before you pay, verifying the factory, production line and documents.