Since the topic is Chinese factories on 1688, an SEA supply-chain sourcing operator, and doesn't require any repo code changes, I'll write the blog post draft directly.
Why supplier financial health matters more than a low price tag
I have watched a shop owner sign a six-month contract with a 1688 factory quoting 15% below every competitor, only to find the factory's doors closed three weeks before Tet restocking. No warning message, no refund on the deposit, just a WeChat account that stopped replying. The shop lost its best-selling SKU during the exact week it needed 400 units shipped to keep its Shopee ranking.
That is the risk nobody puts in a spreadsheet. Price risk is easy to see: you compare quotes, you negotiate, you know exactly what you are paying. Financial health risk is invisible until it is not. A factory can quote you a great price and still be three months behind on rent, burning through a loan, or quietly restructuring debt under a new shop name. None of that shows up in a product listing.
Small suppliers on 1688 are more exposed to cash flow shocks than factories with their own brand. A branded manufacturer usually has multiple sales channels, export contracts, and a bank willing to extend credit during a slow month. A small workshop selling only through 1688 often depends on deposits from buyers like you to fund the next production run. When order volume dips, even briefly, that workshop has no buffer. It either delays your order, demands more cash upfront, or shuts down.
The real cost of a supply cut is never just the missing units. It is the orders you cannot fulfill during 11.11 or Tet, the ranking drop on Shopee or TikTok Shop when your listing goes out of stock for two weeks, and the repeat customers who buy from a competitor instead and do not come back. If you have read our post on risks of running out of stock during peak season, you already know this compounds fast: one missed restock during peak can cost more than a year of margin from a "cheap" supplier.
Early warning signs a 1688 supplier is in financial trouble
None of these signs alone means a factory is failing. Together, two or three of them in the same month is a pattern worth acting on.
Response times slow down and delivery dates keep slipping. A factory that used to reply within two hours now takes a day. Promised ship dates move from 7 days to 12, then 18, with a new excuse each time (raw material delay, worker shortage, "quality check"). One slip is normal. Three in a row on the same order is a cash flow symptom, not a logistics one.
Deposit requirements jump without explanation. If a supplier who normally asked for 30% deposit suddenly wants 50%, or insists on 100% payment before production starts, treat that as a liquidity signal. Factories that are financially stable rarely need to front-load risk onto the buyer. Ones running low on working capital do.
New SKUs and sample offerings shrink over 2-3 months. Check the shop's product page history if you can, or just note what you see over repeat visits. A factory adding new molds, new colorways, new packaging options is investing in the business. A factory that has not added anything new in a quarter may be conserving cash rather than growing.
New reviews (评价) mention slower shipping or lower quality than before. Read the most recent 10-15 reviews, not the total count. A workshop with 4,000 reviews from two years ago and five recent complaints about late shipping has changed since those older reviews were written.
The bank account for payment changes and does not match the registered company name. This is the clearest red flag on the list. A legitimate factory keeps payment tied to its registered business entity. A workshop asking you to wire money to a personal account, or a company name that does not match the one on their business license, is often signaling a restructuring, a change of ownership, or an attempt to collect payment before insolvency.
Before you commit a deposit to any of these, run the check described in how to vet a 1688 factory before you pay a deposit. It pairs well with the framework below.
The 4-layer framework: cash flow, scale, transaction history, production capacity
This is the part most sourcing guides skip. "Check the reviews" is not a framework, it is a single data point. Here is a structure you can run through in about 15 minutes per supplier, using only what is visible on the 1688 shop page.
Layer 1: Cash flow signals
Look at transaction frequency and average order value on the supplier's 1688 storefront (usually visible under their transaction or sales history tab). A factory doing steady, frequent transactions across many buyers is healthier than one with a handful of large orders and long gaps between them. Watch for month-to-month swings: a supplier that did 200 transactions in March and 40 in June, with no seasonal explanation, is worth a direct question before you commit.
Layer 2: Scale
Check 成立时间 (year established), staff count if listed, and factory floor area if disclosed. A supplier operating for 8 years with 50+ staff and a stated 3,000 square meter facility carries a different risk profile than one established 14 months ago with no staff count listed. Scale does not guarantee stability, but it usually means more than one order is keeping the lights on.
Layer 3: Transaction history
Look at on-time delivery rate, dispute rate (纠纷), and 诚信通 (Chengxintong) trust score. These numbers exist specifically so buyers do not have to guess. A supplier with a dispute rate above 5% or an on-time rate below 90% has an operational or financial problem worth asking about directly, even if the price is good.
Layer 4: Production capacity
Confirm whether you are dealing with an actual factory (工厂) or a trading intermediary (贸易商) reselling from a factory you never see. Ask directly, and cross-check against the photos and video on their storefront. A real factory can usually absorb a sudden order increase (say, doubling your usual 500-unit order to 1,000 for a promo) within a reasonable lead time extension. A trading company reselling someone else's capacity often cannot, because they do not control the production line.
Scoring it
Score each layer 1 to 5 based on what you find:
- 1-2: red flag, missing or concerning data
- 3: acceptable, no strong signal either way
- 4-5: strong, verifiable evidence of stability
Add the four scores for a total out of 20. A supplier scoring 16+ is a reasonable candidate for a larger or longer commitment. A supplier scoring under 10 should stay on small trial orders regardless of how attractive the unit price looks. This single number is also what makes comparing five candidate suppliers for the same SKU actually workable, instead of relying on gut feeling from a WeChat chat. For a longer walkthrough of applying this before signing anything, see our supplier evaluation process before contract signing.
How to read public 1688 data to check a factory's financial health
Most of what you need is already on the shop page, you just have to know where to look.
公司实力 (company strength) section. This shows registered capital, years since establishment, and any certifications on file. Registered capital under RMB 500,000 for a supplier claiming to run a large-scale factory is worth a second look. It is not proof of fraud, but it is inconsistent with the story they are telling you.
30-day and 90-day transaction history. Compare the two. A supplier whose 90-day volume is much higher than what the last 30 days would suggest (i.e., recent activity has clearly dropped off) may be in a slow patch. If the shop shows year-over-year data, compare the same month last year against this year. A 40% drop in transactions during what should be a comparable period deserves a direct question, not an assumption.
诚信通 (Chengxintong) level and duration. The level itself matters less than how long the supplier has held it. A supplier that just reached a higher Chengxintong tier last month has less of a track record at that tier than one who has held it for three years. Duration is the signal, not the badge.
Recently renamed shops or swapped logos with the same product catalog. This is one of the more reliable signs of financial distress. If a shop changed its name and logo in the last few months but is still selling the exact same products, with the exact same photos, that pattern often indicates the underlying company restructured, possibly to shed debt or exit a dispute under the old name. Cross-check the registered company name against what you paid on your last invoice. For more detail on reading this kind of pattern in reviews and shop history together, our guide on reading supplier reviews on 1688 correctly goes deeper.
Cross-checking through channels outside 1688
The 1688 storefront tells you what the supplier wants you to see. These four checks pull from sources they do not control.
Qichacha or Tianyancha. If you or someone on your team can access these Chinese business registries, search the registered company name (not the shop name, the actual legal entity) for active lawsuits, unpaid debt judgments, or a recent change in legal representative. A company with three active debt disputes filed in the last six months is telling you something the 1688 page never will.
Ask your freight forwarder. Forwarders who have moved cargo from that factory before can often tell you, informally, whether shipping volume from that address has dropped over the last 3-6 months. This is one of the most underused checks available to small importers, and it costs nothing but a message to someone you already work with.
Ask other shop owners who have sourced from the same supplier. Cross-border sourcing communities (Facebook groups, Telegram channels for SEA importers) often have someone who has dealt with the exact factory you are vetting. A direct question there beats guessing from reviews alone.
Request a recent factory video and compare it to older photos on the storefront. Ask for a walkthrough video dated within the last two weeks. Compare staff count, machine activity, and inventory levels against older photos on the shop page. A factory floor that looks noticeably emptier than the photos from a year ago is a visual version of the same signal you are getting from the transaction data.
Case study: a supply cut during peak season and the lesson from it
A shop selling insulated tumblers on Shopee had one supplier for its top SKU, a 500ml stainless steel bottle that made up 35% of monthly revenue. The relationship had run smoothly for a year: on-time delivery, consistent quality, reasonable pricing. No reason to look twice.
Two months before 11.11, the supplier's response time slid from same-day to 2-3 days. The shop noticed but attributed it to a busy season across all their suppliers. One month before 11.11, the supplier asked for 60% deposit instead of the usual 30%, citing "material cost increases." The shop paid it, wanting to lock in stock ahead of the promotion.
Ten days before 11.11, the factory stopped responding entirely. No shipment, no refund, no explanation. The shop scrambled to find an alternative supplier with five days' notice, paid a rush premium of roughly 20% above normal unit cost, and still could not get full inventory in time. The listing went out of stock for six days during the highest-traffic period of the year. Shopee's algorithm deprioritized the listing afterward, and recovery took close to two months.
Running the 4-layer framework in hindsight, layer 1 (cash flow) would have flagged the slowing response times as an early signal, and layer 3 (transaction history, specifically the deposit change) would have scored this supplier a 2, not a 4, a full month before the cutoff. The lesson is not that the shop should have predicted the exact failure. It is that periodic re-scoring, not a one-time check before the first order, would have surfaced the risk with enough runway to act.
When to sign a long-term contract vs. when to test with a small order first
A framework score is only useful if you act on the number.
16/20 or higher: reasonable to discuss a longer-term contract or volume-based pricing tiers. Still start with a moderate order, not your full peak-season volume, to confirm real-world performance matches the score.
11-15/20: stick to smaller, repeat trial orders. Do not commit to a single-supplier arrangement for a SKU that drives significant revenue.
10 or below: treat as backup capacity only, not a primary source, regardless of price.
Even at a strong score, do not put all your volume behind one supplier. A supplier scoring 18/20 today can still have a bad quarter. Split your top 2-3 SKUs across at least two suppliers where the product allows it, even if one is slightly more expensive. The cost of that redundancy is almost always smaller than the cost of one missed peak-season restock.
When a supplier's score sits in the middle range but they are otherwise a good fit (unique product, good communication, reasonable price), negotiate payment terms instead of walking away outright. Ask for milestone-based deposits (30% on order confirmation, 30% at midpoint production check, 40% on shipment) rather than 100% upfront. This limits your exposure without losing access to a supplier you might want to grow with once their track record lengthens.
Re-run the framework on your active suppliers at least twice a year, and always in the 6-8 weeks before a major peak season (Tet, 11.11, 12.12). A supplier that scored well in January can look very different by September. Fifteen minutes per supplier is a small cost against the alternative.
Frequently asked questions
Is a 1688 supplier stable if their Chengxintong (诚信通) score is high?
A high Chengxintong score is a positive signal but not a complete answer. It reflects platform-tracked reliability metrics like dispute rate and response time, not the supplier's actual cash position. Use it as one input in layer 3 of the framework, not as a standalone stability check. A supplier can hold a high score while cash flow is deteriorating in the background, especially if the score reflects data from months ago.
How do I check a 1688 factory's financial health if I do not read Chinese?
Most of the data you need (transaction history, 成立时间, Chengxintong level) sits behind labeled fields you can translate with your browser's built-in translate function or a screenshot tool. You do not need fluency, you need to know which five or six fields to look for, which is what this framework is built around. For the review-reading portion specifically, a browser translation of the most recent 15-20 reviews is usually enough to catch a pattern shift.
Should a high number of reviews (评价) be the main thing I trust?
No. Review count reflects history, not current state. A shop with 5,000 reviews accumulated over three years can still be in trouble today if the last month of reviews shows a shift in complaints. Weight the most recent 20-30 reviews far more heavily than the total count, and read them for pattern changes (shipping delays, quality complaints) rather than star rating alone.
How often should I re-evaluate a supplier I am already working with?
At minimum twice a year, and always 6-8 weeks ahead of any major sales peak your business depends on. If you notice any of the early warning signs in this post (slower responses, deposit changes, shrinking SKU range), re-run the framework immediately rather than waiting for the scheduled check.
What if only one supplier fits my product and they score low on the framework?
Do not walk away entirely if the fit is genuinely unique, but change how you structure the relationship. Move to milestone-based deposits instead of large upfront payment, keep order sizes smaller than you would with a high-scoring supplier, and actively look for a second-choice backup supplier in parallel, even one you do not use yet. The goal is reducing your exposure, not eliminating the relationship.
Ordinex Scout is built for exactly this kind of supplier vetting, pulling transaction history, Chengxintong data, and shop signals into one view instead of ten open tabs. It is in private beta right now alongside Ordinex Orders, our import order tracking tool. If you are sourcing from 1688 regularly and want early access, check ordinex.cc.