Vinted algorithm 2026: how listing ranking really works

Search for "Vinted algorithm" on Google and you'll find dozens of articles claiming to know the "5 secret factors" of ranking, the exact time to relist and the number of relists beyond which you'd be penalised. Almost none of them cite a source.

This guide tackles the question the other way round. It starts from what Vinted officially publishes about its recommender systems, clearly separates what's documented from what comes from sellers' observations, then turns it into an action plan. It focuses on how ranking works, rather than on day-to-day execution (time slots, prices, photos, titles, responsiveness), which comes up in the action plan at the end.

A word on method: nobody outside Vinted knows the details of its internal algorithm, not tool developers (ResellTrack included), not sellers with thousands of sales. Anything beyond the official documentation is inference.

The Vinted algorithm in brief

The short answer, if you only have thirty seconds:

• There isn't one algorithm but several ranking contexts: search, the news feed and item recommendations. Each weighs signals differently.
• Vinted officially documents its criteria in its help page on recommended content: filters, promoted content, item attractiveness (upload date, interactions, price, condition) and personalisation.
• The exact weighting isn't public. No third-party tool can claim to know it.
• Bumps and Showcase are the only two official, paid ways to buy visibility.
• The most powerful lever sits before the algorithm: listing items for which demand already exists.

What Vinted officially documents

Since the Digital Services Act (DSA) came into force, large online platforms have to explain the main parameters of their recommender systems in their terms. Vinted does this in a dedicated help page, "Recommended content on Vinted". It's the only first-hand source on the subject, and it's more precise than most people think.

The parameters presented as most important

Vinted lists four families of main criteria:

1. Filters: those applied by the buyer (size, price, brand, condition…) and those applied by the system, for example whether the item is available in the buyer's country. An item outside the filters simply doesn't appear.
2. Promoted content: new listings and the paid promotion services, Bumps and Showcase.
3. Item attractiveness: upload date, number of interactions, price and condition.
4. Personalisation: the member's past interactions on the platform.

The full list of signals used

The same page details the data its systems can take into account:

• the time since the listing was uploaded;
• the number of interactions (favourites, clicks);
• the buyer's and seller's location, and preferred language;
• the price, characteristics (category, brand, size…) and condition of the item;
• listing frequency;
• seller type (private member or Pro Seller);
• the item description;
• image resemblance;
• the member's history and the list of members they follow;
• measures taken against illegal content.

Two items on this list are rarely mentioned elsewhere. Listing frequency confirms that being an active, regular seller is a signal taken into account, without saying in which direction or with what weight. Image resemblance explains why the "similar items" under a listing aren't limited to the same brand: the photo itself is used to match items.

What Vinted doesn't say

The page gives no weighting, no freshness window, no seller rating threshold, no rule on relisting. It also states that each member can change their personalisation settings or opt out of recommendations. In other words, two buyers typing the same query don't necessarily see the same ranking. Any "position" you think you observe is relative to your own account.

The ranking signals, one by one

For each signal, here's what's documented and what comes from the seller community's observations.

Search relevance: title, brand, category, size

Documented: the item's characteristics, its description and the buyer's filters are among the parameters. On Vinted, much of the searching goes through structured filters (category, brand, size, condition, colour) rather than free text.

Direct consequence: a badly filled field shuts you out of filtered results, however good your listing is. A Ralph Lauren shirt filed under "Tops and t-shirts" instead of "Shirts", or with no size, vanishes for every buyer who filters. This first signal is binary: you're in the results or you're not.

Observed by the community: titles built like search queries (brand, type, detail, size) seem to perform better in text search. That's consistent with how a search engine works, but Vinted doesn't publish a precise rule on the weight of the title versus the description.

Freshness

Documented: the time since upload is a parameter, and "new listings" are explicitly cited as promoted content. It's the best-established signal on the whole list.

Observed: how long this advantage lasts (a few hours, one or two days) varies between sources and has never been confirmed by Vinted. It probably depends on listing volume in the category: in a very busy category, a new listing gets pushed down faster.

Engagement

Documented: the number of interactions (favourites, clicks) is part of item attractiveness.

Observed: engagement works as a loop. A listing that picks up favourites in its first hours stays visible longer, which brings it more favourites. Hence the importance of that first exposure: price, first photo and title do most of the work in those early hours.

Price and condition

Documented: price and condition are among the attractiveness criteria.

What you can infer: Vinted doesn't say that a low price is favoured. However, a price out of line with the market produces fewer clicks and favourites, and so indirectly drags down another documented signal. The right benchmark isn't the price listed by other sellers but the price similar items actually sell for.

Seller profile

Documented: location, seller type and listing frequency are cited. So is whether the buyer follows your wardrobe.

Not documented: Vinted doesn't explicitly list seller rating or response time among its recommendation parameters. These strongly affect conversion (a buyer hesitates when facing a poorly rated profile), but their direct effect on ranking is a hypothesis. Be wary of articles announcing a precise rating threshold below which "the algorithm penalises you".

News feed and search: two different logics

Mixing the two up is the most common mistake. Vinted states that the news feed uses recommendation and re-ranking systems, while search results are ranked by relevance and interactions.

• Search: triggered by a query or filters. Dominant signals: filter match, relevance, freshness, interactions.
• News feed: triggered by opening the app. Dominant signals: buyer history, followed wardrobes, similarity with their favourites, newness.
• Similar items: triggered by viewing a listing. Dominant signals: item characteristics, image resemblance, price.

What this changes for you:

• In search, you compete with every listing matching the query. Accurate fields and price positioning matter most.
• In the feed, you're shown to buyers who have already shown interest in this type of item. The more active buyers a niche has, the more feeds your listing can appear in.
• In similar items, the photo works for you: a sharp, representative first image brings you closer to the listings the buyer is already looking at.

If the buyer changes the sort order (by price or by date), the relevance logic no longer applies.

Bumps and Showcase: the two official paid levers

Vinted offers two promotion services, cited in its documentation as promoted content.

Bumps

• A Bump promotes a single listing, which appears higher in the news feed and search results.
• It runs for 3 or 7 consecutive days, or until the item sells.
• Reach can be national or international, as you choose.
• The price varies with the item's price, the duration and the reach.
• The listing must comply with the listing rules. You can lower the price during a Bump, but not raise it.

Showcase (formerly Wardrobe Spotlight)

• It promotes up to 15 listings from your wardrobe, selected according to buyers' preferences.
• It runs for 7 consecutive days.
• It requires at least 5 items listed.
• The price depends on the number of listings and their value, and is shown at checkout.

When these services make sense

A Bump or Showcase buys exposure, not demand. If nobody is looking for the item, putting it at the top of a list nobody reads changes nothing. If the item is badly priced, you're paying to show a bad price to more people. Both services make sense on items that are already well positioned, in a niche with active demand, and with a margin that absorbs the cost. On a low-value item, the maths rarely works out.

Myths about the Vinted algorithm

"You have to relist every day"

Freshness is a documented signal, so relisting a listing gives it a temporary advantage again. But relisting your whole stock daily causes three problems: your own listings compete with each other in the same results, you lose the favourites and interaction history you'd built up, and you hide the real problem if the item isn't selling because of price or demand. Nothing in the official documentation points to an optimal frequency. The "ideal" frequencies you'll read elsewhere are community conventions.

For tools that automate relisting, read our analysis of Vinted bots and their risks before diving in: automation is restricted by the platform's Terms and Conditions.

"My account is shadowbanned"

Vinted doesn't document any "shadowban" mechanism. What does exist are moderation measures against content that breaks the rules, cited in the parameters. In the vast majority of cases, a drop in views has another explanation: seasonality, a saturated category, a price above the market, a badly filled field that shuts you out of filters, or simply the end of the freshness effect. Before concluding you've been hit by an invisible penalty, check those points.

"There's a magic time to list"

Traffic peaks differ between categories, and listing time makes up for neither a bad price nor missing demand. Timing amplifies, it doesn't cause.

"More photos means better ranking"

The number of photos doesn't appear in the official parameters. Photos act on clicks, favourites and image resemblance: quality beats quantity.

"Third-party tools know the algorithm"

No external tool has access to Vinted's ranking code. A tool measures observable outcomes (sales, prices, volumes), not the internal mechanics.

The real lever sits before the algorithm

Reread the list of official signals. Interactions, followed wardrobes, buyer history, feed personalisation: almost everything depends on there being buyers interested in this type of item. The algorithm doesn't create demand, it distributes it.

A perfect listing in a category × brand combination few people buy will get few favourites, appear in few feeds and stay unsold. A decent listing in a high-demand combination will automatically get more interactions, and so more visibility. That's why the decision that matters most is made at the sourcing stage, not at the listing stage.

That's where sales data becomes useful. ResellTrack doesn't optimise your ranking and doesn't claim to know Vinted's algorithm. It shows you what has actually sold, with data consolidated within 24 hours:

• The instant scan of a niche gives you the data immediately, without creating a tracker or waiting for it to build up history.
• The Trending page ranks the category × brand combinations that sell best.
• The niche Score, from 0 to 100, sums up how attractive a niche is.
• The Median sold price gives you a price benchmark based on sales, not on listed prices.
• The Sales flow highlights the winning time slots and the ones to avoid for a given niche.
• The Chrome extension shows, directly on Vinted, sales of similar items, the average, minimum and maximum price, a profile's sales and catalogue statistics with the sell-through rate.

To dig deeper into upstream selection, see our method to find a profitable Vinted niche and our analysis of saturation vs opportunities: a high-demand niche that's overloaded with listings also dilutes your visibility. The tools are detailed on the ResellTrack features page.

Action plan: working with the Vinted algorithm

Here's the order to act in, from the strongest lever to the weakest.

1. Pick items for which demand exists. Before buying, check that the category × brand combination sells and at what price. An instant scan of the niche is enough; at a car boot sale, Lens finds similar items and a price estimate from a photo.
2. Fill in every structured field accurately. The most precise category possible, exact brand, size, condition, colour. That's what gets you into filtered results.
3. Write your title and description for search. Use the terms buyers actually type, without keyword stuffing.
4. Set your price from real sales. Price is an official attractiveness criterion, and it drives interactions.
5. Take care over the first photo. It decides the click in search and the matching by image resemblance.
6. List regularly rather than in bursts. Listing frequency is a parameter cited by Vinted, and spreading out your uploads multiplies your freshness windows.
7. Diagnose before relisting. Views without favourites point to price or photos; few views point to fields, title or weak demand.
8. Keep Bumps and Showcase for items that justify them. An optimised listing, active demand, enough margin.
9. Review your selection every week. If you work with an AI assistant, the ResellTrack MCP (from the Pro plan) connects Claude or another compatible assistant to the sales data, so you can query niches, brands, sellers and prices in plain English.

For a big-picture view of the business, from buying to reselling, the complete guide to reselling on Vinted covers the whole chain.

Frequently asked questions

How does the Vinted algorithm work?

Vinted ranks listings in several contexts (search, news feed, similar items). According to its official documentation, the main criteria are buyer and system filters, promoted content (new listings, Bumps, Showcase), item attractiveness (upload date, interactions, price, condition) and personalisation based on the member's history. The exact weighting isn't public.

Did the Vinted algorithm change in 2026?

Vinted doesn't publish a changelog for its ranking. The official criteria remain those on its recommended content page. On the paid services side, the former Wardrobe Spotlight is now called Showcase. Be wary of articles announcing an "algorithm update" without a source.

How do you push your Vinted listings back up?

Three options: relist the item to get the freshness effect again, buy a Bump (one listing, 3 or 7 days) or Showcase (up to 15 listings, 7 days), or improve what drives interactions, starting with price and the first photo. Fix the cause before relisting, or the listing will slide back down at the same pace.

How can you improve your visibility on Vinted for free?

Fill in every field precisely to appear in filtered results, align your prices with real sales, list regularly rather than in bursts, and above all sell items for which demand exists. An active niche automatically produces more interactions, and so more visibility.

Does shadowbanning exist on Vinted?

Vinted doesn't document any shadowban. The platform applies moderation measures to content that breaks its rules, which is different. A drop in views is most often explained by seasonality, a price above the market, a badly filled field or a saturated category.