12of120
monthSold top decile across API result rows.
12 of 120 first-page rows accounted for 85.97% of the sample's reported monthSold total
The 12 rows in the top monthSold decile accounted for 85.97% of the sample total. The first API result row was not the monthSold leader in 11 of 12 selected searches. This measures returned fields and order, not verified sales or a ranking formula.
Liam Cai · Founder, SupplymoYiwu, ChinaWhat we measured
Rows inside and outside the monthSold top decile among 120 API result rows
monthSold is platform-reported and unaudited. This is not a measure of verified sales and does not reveal the ranking algorithm.
monthSold is platform-reported and unaudited. The study cannot call it verified sales, orders, units, revenue or market share.
Study reading
The returned monthSold total was concentrated
All 120 rows returned monthSold. The field summed to 25043070 across the sample, with a median of 16640.5. The maximum returned value was 8876681 and the minimum was 41.
The 12 rows in the top decile accounted for 85.97% of that total. The single largest row accounted for 35.45%. These are arithmetic properties of the returned field, not audited revenue or completed orders.
Within each selected search, the median leader share of keyword-level monthSold was 43.6%. The sample is non-random and first-page only, so that concentration cannot be generalized to all 1688 products or categories.
Study reading
API result order was not monthSold order
The association between API result position and within-keyword monthSold rank was 0.0828. In 11 of 12 searches, the first returned row was not the row with the largest monthSold value.
That difference does not explain why the API ordered the rows. Relevance, promotion, personalization and other inputs were not observed, and a marketplace is under no obligation to sort by monthSold.
The study also cannot establish what a rendered website, logged-in session, other API client or later request returned. Its claim stops at one read-only API snapshot.
Study reading
How a buyer can use the result
Keep monthSold attached to its source field and collection date. Use it as one platform-reported clue, not a sales certificate or forecast for the exact SKU under review.
Compare candidate offers on the same product specification, MOQ, quantity and packaging. A high returned value does not confirm that the activity belongs to the buyer's target variant or that the proposed order terms remain available.
Before payment, verify the product, supplier, quote, evidence and destination-market requirements. Popularity-like fields can prioritize review but cannot replace those checks.
Use this data
Everything here is free to reuse with attribution under CC BY 4.0. If you need a cut of the data we have not published, ask and we will run it.
Quote it, or take the whole file
One line, free to quote with attribution
In a Supplymo non-random sample of 120 first-page API result rows across 12 selected English searches on 1688, the 12 rows in the top monthSold decile accounted for 85.97% of the sample's platform-reported monthSold total; the first returned row was not the keyword's monthSold leader in 11 of 12 searches.
How to cite this study
Liam Cai. "Concentration in API-returned 1688 monthSold values." Supplymo, 8 September 2026, https://supplymo.com/research/1688-sales-concentration.
The retained source contains no offer or seller identity. The 120 rows therefore cannot be verified as 120 distinct offers or de-duplicated by shop. 8 stated limits travel inside the file. Free to reuse with attribution under CC BY 4.0.
Method
How the sample was collected
Everything here is reproducible from the evidence file and the sampling script.
Source
com.alibaba.fenxiao.crossborder:product.search.keywordQuery (read-only)
Sampling frame
12 English keyword searches on the cross-border distribution channel, first results page only, 10 platform-ranked rows per keyword. The same draw as the seller-score study, so the two describe the same shelf.
Collection window
Single day, 2026-08-04 in China
Write operations
None. No order, payment, enquiry, messaging or logistics endpoint was called.
Check the figures yourself
The retained source contains no offer or seller identity. The 120 rows therefore cannot be verified as 120 distinct offers or de-duplicated by shop.
- Rows
- 120
- Collected
- 2026-08-04
- Stated limits
- 8
Readings tested before publication+
Is the rank position real, or just the order rows happened to be stored in?
Method. The sampling script appends rows in the order the search API returned them, and a row is only skipped when its detail call fails, which would shift later positions. Count the detail failures in this draw.
Result. All 120 of 120 attempted rows retained their query-local sequence, so no position shifted after collection. Rank position here is exactly the first-page result order the API returned.
Conclusion. Result order is used as returned by this API client. It does not establish what a rendered website or a different client returned.
Does a missing monthSold mean zero sales?
Method. Record whether the field was returned on every row and publish the return rate next to every dependent figure.
Result. The field was returned on 120 of 120 rows, so no missing-versus-zero judgement was needed in this draw. The minimum observed value is 41, not zero.
Conclusion. All figures use rows as returned. Had the field been absent anywhere, those rows would be excluded rather than read as zero.
Could one mega-listing be a data error rather than a real outlier?
Method. Inspect the largest value against its keyword context rather than assuming.
Result. The largest row returned 8,876,681 under monthSold, equal to 35.4% of the sample total. The study has no independent transaction record with which to authenticate it.
Conclusion. Kept exactly as returned and explicitly unaudited; overall and per-keyword shares let readers see the outlier's influence without treating it as verified sales.
Limits
The 8 limits of this sample
All 8 stated limitations, in the order the evidence file lists them. A number travels further than the caveat attached to it.
- 1This is not a random sample. Rows come from the first page of a keyword search, which the platform ranks. The finding is about that first page specifically, and deeper pages may show a different relationship.
- 2monthSold is a platform-reported API field. We did not audit how it is computed, whether the unit is pieces or orders, how returns are treated, or whether the value maps to completed transactions.
- 3A ranking algorithm is under no obligation to sort by monthSold, and a low correlation is not evidence of a defect. This study compares returned order with one returned field and does not infer ranking inputs.
- 4The result order is what the cross-border keyword API returned to one client on one day. It does not establish what a rendered website, logged-in session, different client, domestic interface or later request returned.
Show the other 4 limits+
- 5Single day of collection. Sales figures and rankings both move, and no trend is claimed from one observation.
- 6The 12 keywords are ordinary consumer goods in English. Concentration in niche, regulated or made-to-order categories may look different and is not represented here.
- 7The rows are listings, not distinct suppliers. Seller identity is deliberately not collected, so the sample cannot be de-duplicated by shop and one seller may appear more than once, which could overstate or understate concentration.
- 8Nothing here says a platform or seller has done anything wrong. The result order and monthSold field serve different functions, and the study does not score product quality, seller reliability or buyer demand.
Research boundary
What this study cannot tell you
- monthSold is platform-reported and unaudited. The study cannot call it verified sales, orders, units, revenue or market share.
- The study compares one API result order with one returned field. It does not reveal ranking inputs, prove a defect or establish rendered-page order.
- The 120 rows cannot be verified as 120 distinct offers or suppliers because the retained source contains no offer or seller identity.
What this means for a buyer
A sample of this kind is a starting point for questions, not a verdict on any one supplier. Check the exact product and supplier you are about to pay.
Supplier verification
A free check before you commit. We run supplier checks from Yiwu, and we do not promise a lowest price.
Check a supplier →Questions a reader should ask about this
What is the one-line finding?
In a Supplymo non-random sample of 120 first-page API result rows across 12 selected English searches on 1688, the 12 rows in the top monthSold decile accounted for 85.97% of the sample's platform-reported monthSold total; the first returned row was not the keyword's monthSold leader in 11 of 12 searches. monthSold is an unaudited API field, not verified transactions or market share.
Was the first returned row usually the monthSold leader?
No. In 11 of 12 selected searches, the first API result row was not the row with the highest returned monthSold value.
Are these verified sales figures?
No. The official schema labels monthSold as 30-day sales, but Supplymo did not audit its unit, formula, denominator, update timing, return treatment or relationship to completed transactions.
Does the study explain the 1688 ranking algorithm?
No. It records an association of 0.0828 between API result order and within-keyword monthSold rank. Ranking inputs were not observed.
Were these 120 different suppliers or offers?
That cannot be established from the retained source. Offer and seller identities were removed, so the rows cannot be verified as 120 distinct offers or de-duplicated by shop.