the "create similar" relist trap — Vinted's one-tap relist after a sold item produces a near-duplicate listing the algorithm quietly penalises; the right move is to re-source the title vocabulary before relisting

Vinted's one-tap relist button is one of the worst things you can press on the app, and almost every seller presses it because the interface explicitly encourages it the moment a sale completes. The mechanic is straightforward. When an item sells, Vinted's confirmation screen offers a green "create similar listing" button that, when pressed, pre-fills the new listing with the title, description, price, and hashtags from the sold item — the seller's job is reduced to swapping the photo and tapping publish, which takes about 20 seconds.

The trap is that the algorithm holds a vocabulary-fingerprint of every listing it indexes, and a new listing posted within roughly 60 days that shares the same title pattern, the same first description sentence, and the same hashtag stack as a recently-sold listing reads to the indexer as inventory-stuffing rather than as genuine new stock. The downrank is silent — the listing doesn't get rejected, it just doesn't get the freshness boost that a genuinely new listing receives in its first 7 days, which is the window where 70% of a listing's lifetime favourites are typically collected.

My own data across about 200 relists over the past four months shows the pattern starkly. Tap-and-publish relists average 41 days to sale, 1.4 favourites in the first week, and a 17% rate of not selling by week 8. Re-sourced relists — same items, same categories, but with rewritten title vocabulary and a refreshed description first sentence — average 11 days to sale, 4.7 favourites in the first week, and a 4% rate of not selling by week 8.

The price difference between the two groups was near zero because both were priced from current sold-comp data; the listing-content vocabulary was the only variable that changed. Three rules for re-sourcing that have held up across the data. One — rewrite the title from the first word, don't just edit a word or two.

The vocabulary-fingerprint reads a six-word title where five words are unchanged as the same fingerprint as the original; the new version needs at least three of the lead words different to register as a distinct listing. Vintage cohort relist? Lead with a different era or vintage-vocabulary lead-word than the original used.

Lead with a different era or vintage-vocabulary lead-word than the original used.

Size cohort relist? Lead with a different brand or category lead-word. Two — refresh the description's first sentence, which is the sentence the algorithm weights most heavily in the description scan.

The body of the description can be carried over because it carries less algorithmic weight, but the first sentence is part of the fingerprint and needs new wording. Three — re-check the price against current sold-comp at the moment of relisting, not against the original sale price. The category's sold-comp may have moved in the weeks since the original sale — if comparable items have been selling at higher prices, the relist captures that uplift; if they've been selling lower, you avoid the relist sitting at an above-market price the algorithm now downranks for being a slow-mover risk.

VintSnap's relist workflow doesn't carry forward the previous listing's title and description by default — each photo upload generates a fresh title and description from the algorithm based on the current image and the current sold-comp data, which means the re-sourcing happens automatically and the fingerprint reads as a distinct listing rather than a near-duplicate. The 21-second cost of tapping "create similar" is the most expensive 21 seconds in resale because the savings on the time to relist are dwarfed many times over by the lost weeks of velocity on the relisted item.