Why this page scores sheets instead of crowning one
A "best KakoBuy spreadsheet" is a moving target. A list that was excellent in August can lose a third of its rows by October, because sellers delist items, rename sizes and restock under new IDs. A ranking of named sheets would be wrong within weeks.
So we give you a method. It takes about ten minutes per sheet, works on any list from any source, and ends in one number out of nine. You never need to know who made the sheet. You only need to open it, look at six columns and click ten links.
The nine points build on our downloadable row check CSV. That file judges rows one by one; the score below judges the sheet that holds them.
Score the sheet you already have instead of hunting for a winner.
The 9 points, in the order to check them
Each point is worth one mark, except the dead-row sample, which can earn a half. Run them top to bottom: the first three take seconds, the dead-row sample takes the longest.
| # | Point | Full mark if | Zero if |
|---|---|---|---|
| 1 | Original marketplace link kept | Every row shows a Taobao, Weidian or 1688 address or item ID, not only a KakoBuy link | Rows hold only a wrapped link |
| 2 | Last-updated date | A date within the last 90 days is visible on the sheet | No date, or a date from 2025 |
| 3 | QC photos linked | Most rows link warehouse photos of that item | No photos, or seller pictures only |
| 4 | CNY prices | Prices are in yuan, the currency the marketplace uses | Only converted dollars with no date |
| 5 | Weight column | A gram weight or range on each row | No weights anywhere |
| 6 | Category structure | Tabs or filters by type, with consistent column names | One unsorted pile |
| 7 | Dead-row rate in a 10-row sample | Two or fewer dead (a half mark for three or four) | Five or more dead |
| 8 | No agent-only short links | Links can be read, or lead to a marketplace page | Most links are shortener codes |
| 9 | No branded-copy marketing | No tabs or ratings built around fashion labels or copy-quality tiers | The sheet is organised around them |
Scores from 8 to 9 mean keep the sheet and still test each row before paying. From 6 to 7.5, use it for ideas only and rebuild links yourself. Below 6, skip it. Those bands are our judgment, not an industry standard.
Nine marks, one for each failure that has cost real money; 8 or more is a keeper.
What each point protects you from
Points 1 and 8 (links). An agent-only link is a marketplace address wrapped inside another site's domain. When that agent closes, the wrapper dies and the item number can vanish with it. Several agents closed or went offline in 2026, which we cover on the 2025 sheet recheck page. A sheet that keeps the original address survives all of that.
Point 2 (date). Without a date you cannot tell a six-day-old list from a six-month-old one. The date is also the only way to judge prices and customs notes. US duty-free entry ended in 2025, and the EU charges a flat €3 duty per tariff category on low-value parcels from 1 July 2026, so any cost note older than that is incomplete.
Point 3 (QC photos). KakoBuy says arrival inspection covers look, model, style, colour, size and big-area damage, and it does not publish how many free photos you get. Linked photos from earlier buyers show what the batch looked like. Our page on sheets with QC explains how to read them, including the images KakoBuy flagged from 2 to 8 May 2026.
Points 4 and 5 (CNY and weight). KakoBuy shows prices in yuan with a converted estimate, and it bills shipping by weight, so the yuan price and gram weight are the two numbers that fix a landed cost. A sheet with dollar prices and no weights leaves you to guess both.
Point 9 (branded-copy marketing). KakoBuy's terms prohibit goods that infringe trademarks, and customs can seize such goods. A sheet built around that pitch exposes you to seizure. We score it zero and do not review such sheets.
Each point maps to a specific loss: a lost row, a wrong cost or a seized parcel.
A worked score on a hypothetical sheet
Here is a made-up sheet, "Sheet A", scored the way you would score a real one. It does not exist; the numbers are for illustration, and the sums were run in Python.
1 Original link: all rows show marketplace links → 1
2 Date: header says nothing, last edit unknown → 0
3 QC photos: most rows link warehouse photos → 1
4 CNY prices: yuan shown → 1
5 Weight: no weight column → 0
6 Categories: tabs by type → 1
7 Dead rows: 3 of 10 sampled links dead → 0.5
8 Short links: every link readable → 1
9 Branded-copy marketing: two tabs named after fashion labels → 0
Total: 1 + 0 + 1 + 1 + 0 + 1 + 0.5 + 1 + 0 = 5.5 of 9 → below our 6-point line
A second made-up sheet, "Sheet B", is dated, keeps its links and has 1 dead row in 10, but has no QC links and no weight column. It lands at 7 of 9, which is usable for ideas with a rebuild of each link. The table puts both side by side.
| Point | Sheet A | Sheet B |
|---|---|---|
| 1 Original link | 1 | 1 |
| 2 Last-updated date | 0 | 1 |
| 3 QC photos | 1 | 0 |
| 4 CNY prices | 1 | 1 |
| 5 Weight column | 0 | 0 |
| 6 Categories | 1 | 1 |
| 7 Dead rows in 10 | 0.5 | 1 |
| 8 No short links | 1 | 1 |
| 9 No branded-copy marketing | 0 | 1 |
| Total | 5.5 | 7 |
A high score still does not price the row. Take a ¥150 item (about $21, an example price) at 1,000 g to the UK. KakoBuy's estimator quoted $16.29 on "UK HG dedicated line PT" on 12 October 2026 at a weight of 1,000 g and no dimensions. Add the typical ¥8 declaration fee, about $1.12:
$21.00 item + $16.29 shipping + $1.12 declaration = $38.41
Shipping is 42% of the landed total, which is why a missing weight column costs you real money.
Sheet A scores 5.5 and Sheet B scores 7; the score ranks lists, not prices, so cost each row separately.
Running the 10-row sample properly
The dead-row test is the one people skip, and it is the one that predicts your experience. Do it in a private window so saved logins do not hide a delisted item.
- Pick rows by number, not by eye. Take rows 1, 12, 23 and so on, or use random numbers. The first rows of a sheet tend to be the freshest.
- Open the marketplace link first. If the Taobao, Weidian or 1688 page loads with a price and options, the row is alive.
- Open it through KakoBuy too. On the item page, a live row displays a yuan price, the options and a rough international shipping estimate.
- Compare price and option names. A jump of more than about 10% or a missing size option counts as a change, not a clean pass.
- Count the dead. Count a delisted item, a removed shop and an option that no longer exists as one dead row each.
Ten rows is a small sample. If three of ten fail, the true dead share could plausibly sit anywhere between about 11% and 60%, so open ten more before you judge. Our guide to very large sheets shows how to take a larger sample and remove duplicates first.
Test ten random rows through both the marketplace and KakoBuy before you trust a list.
Red flags that cap a sheet's score
Some patterns are bad enough to stop scoring and walk away. None needs a calculator.
- A request to log in or pay on the sheet's own page. Payment belongs on kakobuy.com only; see our page on official KakoBuy sites and lookalikes.
- Every link is a shortener or app deep link. You cannot recover the item once the shortener expires.
- Brand-name tabs or tier badges. They signal trademark-infringing listings and point 9 goes to zero.
- "Updated daily" with no visible dates. A claim is not a date.
When a sheet fails, the best move is to keep its good rows. Paste them into the haul tracker CSV, which has columns for the original link, the KakoBuy link, the free-storage end date and the return deadline. Where to look for fresh lists is a separate question; the Reddit thread-types page and the Telegram, Beacons and Discord page cover that.
- kakobuy-row-check.csv — 9 per-row checks with pass rules
- kakobuy-haul-tracker.csv — 23 columns for rows that passed
One red flag is a reason to look closer; two is a reason to leave.
Using the score once you have chosen a sheet
Re-score your chosen sheet monthly, because points 2 and 7 change. Keep your own copy with the original marketplace link in its own column, so one agent failure costs you a single formula rather than a rebuild. Pay per order, so no money waits unused with an agent; KakoBuy's help pages mention an account balance, but a score of 9 does not change our advice.
Prefer to skip the list hunt? Scanning the code on the contact page opens a WhatsApp chat with us. Otherwise the KakoBuy spreadsheet homepage carries fresh finds and door-to-door quotes.
Score monthly, keep original links and pay per order.
Questions buyers ask
Is there one KakoBuy spreadsheet I should follow this month?
None can be crowned honestly, because a sheet's quality is its freshness and a ranking goes stale faster than the sheet does. Instead, score any candidate out of 9. In our hypothetical example, a sheet with no date and no weight column landed on 5.5 of 9, below our 6-point line for use with checks. Anything at 8 or above is worth keeping; below 6, skip it.
How many of 10 sampled rows can be dead before a sheet fails?
Open 10 rows at random. Two or fewer dead is a full point, three or four is half a point, and five or more is zero. With a 10-row sample the margin of error is wide: 3 dead of 10 is consistent with anywhere from roughly 11% to 60% dead overall. If you land on three or four, test 10 more rows before trusting the rest.
Does a sheet with more rows score higher?
No. Row count appears nowhere in the nine points. A 600-row list with an undated header, no weights and agent-only short links scores worse than a 60-row list that keeps marketplace links and dates. More rows also means more dead rows to find. Judge by the 10-row sample, not the headline number of products.
Why does the score penalise a missing weight column?
Shipping is billed by weight, and KakoBuy's estimator quoted one US line at $38.07 for 1,000 g and $69.17 for 2,000 g on 12 October 2026. A row without a weight cannot be priced to your door. You will have to guess from a QC photo, which is exactly how total costs end up higher than expected.
Can a sheet that earns commission still score well?
Yes, scoring ignores motive and tests whether the row works. KakoBuy's own help page on its Share W2C tool says a shared purchase link automatically includes the sharer's commission link, so referral links are normal. What matters is that the original marketplace link also stays visible, so you can open the item through any link you prefer.
Can I rescue rows from a sheet that scores under 6?
Salvage rather than discard. Copy only rows that pass the link test into your own haul tracker, rebuild each one from its item ID, and drop the rest. In our hypothetical example, a 5.5 sheet still yielded 7 usable rows from a 10-row sample. Rows with only short links cannot be rebuilt, so those are the first to delete.