Your game isn't dead. It's just quiet.
Of the hundred struggling indie launches read in depth by the Shypiard observatory, 84% share exactly this silence — not a bad game, not a marketing mistake, just velocity that stalled after release. The number, we've already made the case for. Here's the 90-day plan that reverses it.
Not your rating. Your velocity.
The Steam algorithm doesn't compare your game to its own release six months ago. It compares how fast wishlists and sales move over a rolling fourteen-day window. A game that steadily sells ten units a day beats, in the algorithmic ranking, a game that sold a thousand on day one and nothing since.
Concretely: if your sales/wishlist curve has been flat for more than two weeks, the algorithm has already filed your game as "done". That's not fatal — it's a reversible state. But it takes a new event to reset the measurement, not just waiting it out.
Every two-week silence costs you rank.
Northgard ran five years of velocity on twelve DLC. You don't have five years, but the mechanism is the same at 90-day scale: every release hands the algorithm a fresh signal to measure.
Reaction patch
A visible fix within 15 days of release, with patch notes that explicitly cite player feedback (Steam reviews, Discord, Reddit). The content of the patch matters less than the proof that you're listening.
Substantial devlog
An article or video that shows what's coming next — not a bug-fix list, a direction. Published on Steam News (triggers a notification to wishlisters) and reposted to the community.
Content wave or event
An addition visible enough to justify a new capsule or a short new trailer: game mode, level, season. Coordinated with a Steam promotion if the calendar allows (seasonal sales also restore velocity).
Creator reactivation
See section 03 — this is the moment to re-contact creators who didn't respond at launch, with a new angle (the W5-W8 content) rather than a repeat of the same pitch.
A quiet release isn't a failed one.
The common mistake is treating launch-week media silence as final. It isn't — but the pitch has to change. "My game just came out" interests no creator at D+30. "We added X based on feedback" interests far more, because it's news, not a promotion.
The concrete method: rebuild a list of ten to fifteen creators actively covering your genre (not the biggest — the most active on comparable titles released in the last six months, findable via Steam "curators" pages and the video comments of comparable games). Contact them with the devlog or content wave from section 02 as the hook, never with "will you play my game" alone.
Give each one a code, individually, with a line on why you think this specific creator fits thisgame. The reply rate of a generic message sent to a hundred creators is close to zero. The reply rate of fifteen specific messages is far higher — it's the same mechanism that drove the breakthrough of Among Us and Balatro, applied after the fact rather than before.
Four numbers, not a dashboard.
Wishlist velocity
7-day rolling delta, not the total. A total that stalls despite a high headline number is a warning sign, not a comfort.
30-day positive review ratio
Different from the displayed overall rating — an old rating can mask a recent decline if the last-30-days rate drops.
7-day rolling CCU
Recent concurrent player count shows real retention, not just acquisition.
Reddit/Discord mentions outside your community
A signal that a conversation is starting without you initiating it — the only one that predicts a real snowball effect.
This plan assumes a game that, once seen, retains players (decent positive rate, no critical breakage). If the positive review ratio is structurally low, the problem is no longer distribution — it's the fit between the promise and the game delivered, a different diagnosis. The free diagnostic tells the two apart before proposing a plan.
Break the silence
before it becomes permanent.
Test your page first in 90 seconds — or jump straight to the full diagnostic if you're already ready to go deeper.