Steam Discovery Algorithm 2026 and How Studios Plan Launches

Steam is one of the main storefronts for PC games, but the Steam discovery algorithm and the surfaces around it changed noticeably in 2026. A June store refresh reworked the home page, a September update moved a major promotional section from hand-scheduled placements to personalized recommendations, and festival competition keeps growing. For studios and publishers, the practical question is which levers still move visibility and which ones are being overestimated. The sections below sort what Valve documents from what the industry only infers, then turn the 2026 changes into a launch plan.

What Valve says drives visibility

Steam does not run one algorithm. It runs several discovery surfaces side by side, and Valve’s Steamworks documentation names the inputs. Purchases and playtime are the strong signal, tags decide which shelves and similar-game lists a title can appear on, and supported languages influence who is shown the game.

The documentation also rules out several things studios commonly optimize for. Store page traffic, store page conversion rate, and wishlists outside the Popular Upcoming tab are not visibility inputs. Review scores matter less than many assume. As long as reviews stay at mixed or above (40% or higher), the score is not a factor in algorithmic visibility, but a game that falls below 40% is less likely to be featured.

The takeaway is that external marketing does not buy Steam visibility directly. It works only indirectly, through the customer response that Steam then rewards. Creator campaigns, festivals, and community work matter because they generate the purchases and playtime Steam reads as demand.

Tags carry more weight than most store page copy

The top 20 tags determine the tag pages and browse views a game can appear on, and the top 5 should paint a clear picture of the game, since those are also used to describe it. Steamworks requires at least five tags before launch. Tags also got attention from Valve this year. A May 2026 tag taxonomy overhaul added 17 tags, removed 28, and merged several. Any title tagged before that change is worth auditing against the current list.

The 2026 changes that matter

The June store refresh

The redesigned store home went live on June 4, 2026, after a beta that began April 1. The most consequential change for indie teams involves Popular Upcoming. The widget used to rotate through any game about to release that met a wishlist threshold relative to the other releases that week. It now shows only the most popular upcoming releases.

Historically, roughly 5,000 to 7,000 wishlists was typically enough to appear. While Valve has never published an official threshold, 2026 industry benchmarks show that a minimum baseline has moved closer to 8,000 wishlists for basic visibility features, while hitting front-page algorithmic shelves now realistically requires between 50,000 and 90,000 wishlists.

Other refresh changes shape how players find upcoming games. A personal calendar now shows new and upcoming games from the next two weeks, recommended from each user’s play history. The Discovery Queue was rebuilt into a faster, overlay-style flow, and hover trailers became the most visible new feature. The practical implication is that the capsule and the opening seconds of the trailer now work harder, since many shoppers will see only a short preview before deciding whether to click.

September and the algorithmic Discounts & Events section

On the post-launch side, Valve announced that Discounts & Events, previously made up of curated placements scheduled by partners or the Steam team, is moving to an algorithmic showcase. In testing, the model showed 10x more games to customers in a given day in that section, and Valve reported meaningful increases in store page visits, wishlists, and cart additions. Valve describes the goal as showing the right game to the right customer.

Trade press analysis points to a limitation. Collaborative filtering needs behavioral data, so a game with no purchases, wishlists, or engagement history gives the system little to work with. That makes pre-launch audience building more valuable, since early activity becomes the data that later recommendations draw on.

What this means for wishlists

Wishlists remain useful, just in a narrower way than often claimed. They are not a general ranking input, but they function as stored demand. They drive launch notifications by email and push, and they hold audience for a year of discount campaigns. On conversion, a typical game converts around 15% of wishlists to sales.

For planning, three points hold up:

• Wishlist counts remain a useful measure of owned demand, even though Valve does not use them as a ranking input.
• The 7,000 wishlist rule of thumb predates the June refresh, which may push the effective threshold higher. Plans built on a single number carry risk.
• Purchases and playtime are what Valve names as the strong signal, so the first weeks after release matter more than the pre-launch count alone.

Steam Next Fest in 2026

Next Fest remains one of the most reliable wishlist drivers, and it is also crowded. The October 2026 edition runs from October 19 at 10 AM PDT to October 26. Registration closed August 31, a demo must be live before the fest begins, and a title can take part in only one Next Fest. The 2026 editions ran February 23 to March 2, June 15 to 22, and October 19 to 26. For teams planning 2027 releases, Valve has not yet published dates beyond October 2026, though the festival has historically run three times a year.

What results look like: A gaming report estimates the median demo earns about 200 wishlists across 3,500+ entrants. At the top end, community tracking of the June 2026 edition found about 788 of roughly 4,931 entries crossed 2,000 wishlists, and the top 5% earned an estimated 7,000 to 10,000.

Momentum matters going in: An analysis of 3,000+ games across four Next Fests found that total wishlists and recent wishlist growth rate were the strongest predictors of success, which makes the fest a multiplier on existing momentum, not a replacement for it. Entry guidance varies. One guide recommends at least 2,000 wishlists, while another suggests at least 1,000 wishlists, 10+ organic wishlists per day, a stable demo, and a launch date within 12 months.

The store page works even for non-players: Per Zukowski’s research, 68 to 88% of Next Fest wishlists come from people who never downloaded the demo. Capsule, trailer, and tags do much of the converting.

A 2026 discovery plan by launch stage

Pre-launch

• Audit tags against the post-May taxonomy and order the top 5 deliberately.
• Publish the Coming Soon page 6 to 12 months before launch.
• Open trailers with gameplay in the first five seconds, which suits the new hover previews.
• Build audience in Discord and Reddit, and line up genre-matched creators ahead of the festival.
• Treat Next Fest as the last festival before release.

Launch window

• Concentrate marketing so purchases and playtime arrive together, since those are the signals Valve names.
• Plan Update Visibility Rounds deliberately, as five are granted to start, each lasting up to 30 days or a million impressions.
• Keep reviews healthy, since a fall below the 40% line reduces featuring chances.

Post-launch

• Keep tags and store assets current, since Discounts & Events now draws on behavior and engagement.
• Use sales and updates to keep generating playtime and purchases.

What to measure

Steamworks reports where a game is shown. The traffic data shows the features on each page where the game appeared, the impressions in each placement, and the clicks to the store page. Tracking this by placement shows which discovery surfaces are working and which are not, a sharper read than wishlist totals alone. Wishlist velocity, demo plays, Next Fest follower growth, and review trends complete the picture.

The bottom line

Steam’s 2026 direction is toward personalization. Visibility increasingly follows what players buy and play, rather than a fixed threshold or scheduled placement. Studios that build real audience demand early, keep metadata accurate, and time festivals and launch activity to concentrate that demand give Steam stronger signals to work with.

At Game Marketing Genie, we help studios and publishers turn those signals into a focused discovery plan covering store page readiness, festival timing, creator outreach, and launch-window targets.

Planning your Steam release? Contact us today to build your discovery plan.