Vinted's seller-trust score is one of the least-understood mechanics in resale because the platform never publishes a complete description of how it's calculated, but it can be reverse-engineered from observable ranking behaviour and from what Vinted does and doesn't show in the app. The score is built from at least five signals, and only two of them are fully visible to you as a seller. Signal one — average rating across all completed sales.
Visible. Sits on your seller profile, on every listing, and is the most-weighted signal because it correlates highest with buyer-side platform trust. Signal two — dispatch speed.
Visible as the "fast dispatcher" badge that appears on the profile when the median dispatch time clears a threshold (the threshold sits somewhere around 12 hours but Vinted hasn't published the exact figure). Below the threshold there's no badge but the dispatch-time data is still being collected and fed into the score. Signal three — response rate to buyer messages.
Partially visible as a percentage on the seller profile but the weighting against the score is hidden, and the response-time component is invisible — a 100% response rate over an average of 24 hours scores differently from a 100% response rate over an average of 2 hours, but only one of those is shown publicly. Signal four — dispute and return rate. Invisible.
Inferred from the seller's history of buyer-opened cases (refunds requested, item-not-as-described disputes, payment-refusal cases). Vinted aggregates this across a rolling window — typically the last 90 days — and a shop with three or more open disputes in that window is silently downranked even if the disputes ultimately resolve in the seller's favour. Signal five — active-listings-to-sold ratio.
Invisible. Inferred from shop vitality. A shop with 100 active listings and 80 sold-in-last-90-days reads to the algorithm as healthy throughput; a shop with 300 active listings and 30 sold-in-last-90-days reads as inventory bloat and a slow-mover risk and is silently downranked even on the seller's freshest listings.
Three rules for working with the invisible signals because you can't see them and can't directly optimise them. One — treat every dispute prevention as worth more than the sale itself. A single buyer-opened case in the past 90 days drags the score; the second pushes you into a meaningful downrank tier; the third effectively de-prioritises your shop on competitive searches.
Pre-emptive description detail — accurate condition, measurements, flaws photographed — is worth more than the words you put in the listing itself because it prevents the case being opened. Two — manage the active-listings count downwards when sales are slow. Counter-intuitive but true.
A shop bloated with 200 dead listings ranks worse than the same shop with 60 active fresher listings because the ratio reads as inventory-bloat. The discipline of deleting listings older than 60 days that haven't favourited is one of the highest-leverage shop-level moves you can make, and the only one that touches the active-to-sold ratio signal. Three — the response-time component of signal three is reachable even though it's hidden.
Replying to buyer messages within 1 hour during your active hours pushes the response-time component without changing the response-rate percentage you can see, and the data Vinted is collecting on response time is fed into the score regardless of whether the percentage moves. The wider rule across all five signals — the visible ones tell you what Vinted wants you to focus on publicly, the invisible ones are what actually decides your ranking. VintSnap doesn't change any of the five signals directly — none of them are listing-level metrics — but the listing quality VintSnap generates affects two of them indirectly.
The pre-emptive description detail from the algorithm-tuned generation reduces dispute rates because the listings are more accurate and complete. The faster-velocity listings improve the active-to-sold ratio by moving stock through the shop faster, which pulls the ratio toward the healthy-throughput band. The trust score isn't the listing's job to solve — it's the shop's — but the listings are the throughput that feeds the shop-level signals the score is built from.
Part of our Reseller economics series — field notes from building VintSnap.