Someone in a Discord posts a link to a Google Sheet with four hundred rows on it. Brand, model, seller, batch, price, link, a colour-coded column for quality. It looks like a database. It is a snapshot of what one person believed on the day they typed each row, and the rows were typed on different days.
That is not a criticism of the people who build them. Community spreadsheets are one of the few genuinely collaborative artefacts in this hobby, and some are maintained with real care. It is a description of the format. A spreadsheet has no way to tell you which of its rows was last touched, and that single missing fact is what makes the rest of it dangerous to trust wholesale.
What a spreadsheet row is really made of
Pull one apart and it is four claims of very different durability, glued together in one line.
- The item id. The numeric id inside a Weidian, Taobao or 1688 link. This is the durable part. It survives seller rebrands, dead agents and reposted screenshots.
- The wrapper around it. Whatever agent link the contributor happened to paste. This is the part that rots, and it rots on someone else's schedule.
- The price. True on one day, in one currency, before shipping, before the agent's cut, and before the seller changed it.
- The judgement. "Best batch", "1:1", a green cell. This is one person's opinion of a photo, usually not a garment they held, almost never dated.
Read a row knowing which of those four you are looking at and it becomes useful again. Read it as a single fact and you will eventually pay for a listing that stopped existing in 2023.
Why the links go bad specifically
Three distinct failures, and they do not look alike when you hit them.
The agent shut down. Pandabuy links are the obvious case — an enormous number of spreadsheet rows still point at an agent that no longer takes orders. The link does not resolve, but the item id is still sitting inside the URL.
The seller pulled the listing. The link resolves to a marketplace page that says the item is gone. Sometimes the seller relisted the same product under a new id, which is why finding the seller matters more than finding the row.
The row was never a marketplace link. Some contributors paste a shortener, a Yupoo album, or a screenshot host. A shortener hides the real destination behind a redirect, so nothing can recover the item from the link text alone — you have to open it and read where it lands.
The one that catches people out
A spreadsheet link can be alive and still wrong. Sellers reuse listings. An id that pointed at a hoodie can, months later, point at a different colourway, a different size run, or a completely unrelated product from the same shop. The link works, the page loads, and the thing on it is not the thing in the photo you were sold on.
Four checks before you spend
Run these in order. The whole sequence takes a couple of minutes and it is the difference between buying an item and buying a row.
- Decode the link before you open it. Paste it into the link decoder. It pulls the platform and item id out of the URL's shape with no network call, so a link from a dead agent still gives you a working item. If it comes back as a shortener, open it in a browser and paste the address it lands on instead.
- Look at the listing, not the row. Compare the photos on the marketplace page against the photo the spreadsheet is selling you. If they do not match, the id has been reused or the row was wrong to begin with.
- Check the seller has a record. A row naming a seller nobody has written about is a row with no second opinion behind it. We hold quoted, dated comments on 124 sellers at /sellers, each linked back to the thread it came from, and we say plainly on every one of those pages where the evidence runs out.
- Check whether you are buying from a middleman. Some rows point at a reseller's store rather than the original listing. Same item id upstream, higher number. How to spot that is a short read and it pays for itself once.
What spreadsheets are still good for
They are excellent at the thing they were built for and bad at the thing they get used for.
Good: discovery. A spreadsheet will tell you that a silhouette exists in rep form at all, roughly what it goes for, and which sellers keep coming up next to it. That is a map, and a map is genuinely useful.
Bad: pricing and verdicts. A price column is a historical record with no timestamp. A quality column is an aggregate of opinions from people who mostly saw pictures. Neither should be the last thing you read before you pay.
Why this site is not a spreadsheet
Worth saying, since we are obviously in the same business. This site holds 691 items with a photo we checked and a link that resolves, and 124 seller pages built from quoted Reddit comments rather than summaries of them. Every quote carries its author, its subreddit and its date, and links to the thread so you can read what surrounds it. Nothing here scores a seller, because the evidence we hold is not the kind of thing you can honestly reduce to a number — our method explains what we do and do not claim.
None of that makes us right and a spreadsheet wrong. It makes the failure modes different, and knowing which failure mode you are exposed to is most of the skill. If you are working from a spreadsheet, at minimum date the row yourself before you trust it, and read how to buy reps if the agent step is new to you — these are replicas bought through an intermediary, and the QC photo stage is the only inspection anyone gets.