SUPPLIER SCORECARD / 01
The scorecard is only as honest as its weakest row.
This page is about one row. It does not supply a template, and it does not score performance; only you have that data.
Keep the source date and your review date beside each company check. A record date alone does not tell you whether its details are correct. See the dated research examples.
Start with your order history
Choose measures that fit your orders. Four examples are:
Quality: defects and returns. Delivery: arrival against agreed dates. Cost: quoted versus billed amounts. Service: response time and resolution of problems.
These measures describe past orders. For a new supplier, label missing history clearly. A sample or trial order may add evidence, but it does not prove how a later batch will perform.
Keep company checks outside the weighted total
Company checks ask different questions: does the name match one legal entity, what status is recorded and who is being paid? They do not measure delivery or quality.
Use supplier documents to find the details to check. Compare them with a named source and note its date. A licence image or badge alone cannot resolve every identity question.
Keep the source and any unresolved gap beside the entry so a reviewer can see what supports it.
What belongs in that row for a Chinese supplier
Each item below resolves to a fact with a source and a query date, not to an impression:
Match the registered name and Unified Social Credit Code. An English name can help find candidates; it does not confirm legal identity. Read how the company registration check works.
Read the recorded status and its date. A working website does not establish that the company is active. Check what the exact status means before using it.
Read the registered scope. Manufacturing wording is a clue, not proof that the company can make your product. Read scope wording in context.
Review listings and changes. For penalties, read the reason, date and outcome. For a name change, confirm the link to the entity on your documents.
Check the payee separately. Resolve a different name or new bank instruction before payment. Matching names alone do not prove account ownership. Read the beneficiary checks.
Record availability is not supplier performance
The saved results hold presence and counts. Full source responses are not retained, and the company-level query file is not published.
We queried a commercial platform for 45 company codes selected from the China-listed trailer frame on 21–22 August 2026. The table shows which record types returned data. Availability does not measure supplier performance or provide a basis for score weights.
| Candidate row | Returned a record | Use it as |
|---|---|---|
| Shareholders / annual reports | 44 of 45: 98% | Record the returned data; investigate gaps. |
| Registry snapshot (an image, not fields) | 44 of 45: 98% | Attach it as evidence; it cannot be scored. |
| Change history | 41 of 45: 91% | Read what changed and when. |
| Import/export credit record | 41 of 45: 91% | Check the exporter for your shipment. |
| Tender participation | 27 of 45: 60% | Context, not a quality measure. |
| Outbound investments | 18 of 45: 40% | Context for group structure only. |
| Branches | 8 of 45: 18% | Context; a blank leaves site questions open. |
| Operating anomalies / serious violations | 0 of 45: 0% | Record the source result; do not infer an all-clear. |
No anomaly or serious-violation records were returned. That describes this source on these dates. It does not prove that every company had no adverse record, and it should not award full marks.
Tender and investment records may help explain a company's activities. Their presence or absence does not measure whether it can make your product.
This is not a representative sample of Chinese suppliers. The snapshot row counts returned images, not checked status fields. A missing response needs follow-up, not a score. Read the related 45-code record check.
Scoring it without inventing precision
Do not convert “could not be read” into a number. When a source is unreachable or a field is not returned, the honest entry is unknown. Scoring it zero punishes a supplier for our access problem; scoring it full marks manufactures a clean result out of nothing.
Distinguish “no record found” from “no record exists”. Commercial platforms republish official disclosures with a lag. A platform showing nothing on a query date is evidence about that platform on that date, and nothing stronger.
Date every entry. Keep the source date and your review date. Recheck when details change or before a larger commitment; historical changes do not set one expiry period for every record.
Resolve identity gaps separately. Do not let good delivery or price scores cancel an unresolved entity match. Record the question, who will check it and what evidence is needed.
What this desk does not provide
We supply dated records for your file. You decide how to use them alongside your own order history.
No scorecard template or software. We supply company records for your file.
No performance rating. We have no visibility of your defect rates, your delivery dates or your prices, and any number we attached to them would be fabricated.
No single supplier score. A composite number would hide the thing that matters, which item failed, and against which source on which date.
We verify records and report them with their source, retrieval date and limits. We do not certify that a supplier is safe, approved, or good to work with. Those are conclusions the record cannot support.
Filling the row for a supplier you are looking at now
Start with the supplier's registered Chinese name and code from its business licence. Use them to match the entity before requesting further records.
Check the supplier details you have
The free page explains its available query and document-check modes. It does not supply a supplier score. For a written record report, check the scope and theme packs in the menu.
Research examples: what record dates can tell you
| Velocity measure | Result |
|---|---|
| Companies with at least one registered change | 264 of 264 (100%) |
| Amended within the last 12 months | 145 of 264: 54.9% |
| Amended within the last 24 months | 207 of 264: 78.4% |
| Median lifetime changes on record | 39 |
These figures describe the selected list and its recorded history. They do not set a record expiry date or predict your supplier. The published summary supplies counts and dates; it does not publish all company-level logs. Read the study method for its collection limits.
Record the source date beside the fact you use. Of 106 NHTSA vPIC manufacturer records we read on 28 August 2026, 16.0% had not been updated in over eight years. That date does not prove the details are wrong or still correct. Check the facts that matter to your order. The 106-record field and date study →
Retested on 6 September 2026 against the refreshed list of 111. The number of records older than eight years is still 17, and they are the same 17. The share now reads 15.3% because the denominator grew. The LastUpdated dates for all 106 shared IDs were unchanged between the two snapshots. A percentage can fall while its numerator holds, so read the count beside it. Median record age is 2 years and the oldest is 11. The 111-record file carries every id and its last-updated date.
The two date fields add context: 108 of 111 had a last-update date later than their submission date. Across the file the median gap between the two dates is 147 days, and the median time since that last edit is 733 days. Of the 68 records that are at least two years old, 56 had a last-update date more than two years old. These gaps do not explain why records were updated or left unchanged.
What that second reading cannot show. SubmittedOn and LastUpdated are the regulator's own fields, and this study does not establish what triggers an update. The dates alone do not show which details changed or whether a review occurred. The frame is 111 Chinese trailer manufacturers, so it describes that list and no wider population.
PRIMARY SOURCES / 02
Official sources for this page
The disclosure rule was rechecked on 6 September 2026. Other links are background sources; consult the applicable text for your transaction.
- ISO 9001:2015, Quality management systems: RequirementsA quality-management standard. Consult the applicable requirements for your own purchasing process.
- State Council: Provisional Regulation on Enterprise Information Publicity (2024 revision)Article 8 sets the annual reporting duty. Article 9 distinguishes public information from financial fields the company may choose not to publish.
These sources give background on quality management and company disclosures. They do not validate the score weights in your supplier model.
COMMON QUESTIONS
Questions readers ask at this point
- Should identity checks be part of the weighted score?
- No. Keep them in a separate row with a source and a query date. A high performance total must never hide an unresolved entity or payee question.
- How should I score a record that could not be read?
- As unknown. Scoring it zero punishes the supplier for an access problem. Scoring it full marks manufactures a clean result.
- How fast does the registered record change?
- In the selected 264-company list, 145 had a recorded change within twelve months of 8 August 2026. That history does not set an expiry date for every record. Keep dates beside the facts and recheck what your decision depends on.
How we checked
Fill rates come from dated checks of chosen company codes through one paid data source. A hit is a returned record; it may be incomplete or mislinked. Record ages use the LastUpdated date in each public vPIC record and the query date. They do not measure company age. Last date marked for the checks: 6 September 2026. Use the date shown with each finding. A page update does not mean all checks were run again.
Use the source links. Check the dates and limits stated for each finding.
Start with the free name-comparison tool if you need to compare the parties in your documents. It does not verify bank-account ownership.
If you want these records pulled for your own supplier: select the relevant categories in the report menu, packs from $26.55. Delivery follows the window on your order confirmation. First paid order: unhappy for any reason, tell us within 14 days of delivery and it is refunded in full.