Sony's Concord launched on August 23, 2024 and went offline on September 6 — fourteen days of commercial life for a project reported by press to have cost in the range of $200-400 million in development, figures Sony has not confirmed. Square Enix's Babylon's Fall ran about ten months before its March 2023 shutdown, and Firewalk Studios, Concord's developer, closed that October. These are not outliers. They are the predictable output of an economic model in which only one or two games per genre can survive, and dozens of publishers keep funding attempts anyway.
Understanding the mechanics of live-service failure explains both the industry's churn and the player fatigue directed at it — and it tells you which live games are actually safe to invest hundreds of hours into.
Why can the market only support a couple of winners?
A live-service game is a habit, not a purchase. Its revenue comes from battle passes, cosmetics and expansions purchased by players who show up weekly, and human attention does not scale: a player with 20 hours a week gives them to one or two games, not ten. Every extraction shooter competes directly with established genre incumbents; every hero shooter competes with Overwatch-era patterns and whatever is currently ascendant. The result is a winner-take-most distribution that console generations never had — in 2025, persistent charts were dominated by a short list of evergreens while new service titles launched into a market with no empty hours to claim.
Development cost compounds the problem. Service games carry MMO-like content pipelines — seasonal updates, balance cycles, live events — meaning their burn rate continues after launch at a scale premium games never see. A single-player game that disappoints can be written off; a service game that disappoints must be either funded indefinitely or shut down in front of the players who did show up, which is its own reputational event. Publishers have absorbed the lesson unevenly: Sony publicly committed to live-service expansion in 2023 with a stated target of roughly half its releases, then spent 2024 and 2025 trimming those plans after Concord — a full public cycle from doctrine to retreat.
What does shutdown actually mean for players?
When a service game ends, what dies varies by architecture — and this is the single most important fact for buyers:
- Server-dependent games die completely. A title whose core loops require developer-run servers, matchmakers or account services becomes unplayable at shutdown, as Babylon's Fall and The Crew (servers closed March 2024, sparking a wider preservation debate) demonstrated.
- Service hybrids can survive partially. Games with offline modes — many shooters and RPGs with single-player campaigns — retain a playable remainder after cosmetic shops and ladders close. Suicide Squad: Kill the Justice League's 2025 support ending left its campaign playable even as its live roadmap stopped.
- Purchases do not survive. Battle pass progress, shop currency and event-exclusive cosmetics are entry tickets to a venue, not property. When the venue closes, so do they — refunds have been rare and discretionary.
The Stop Killing Games petition campaign of 2024-2026, which gathered over a million signatures urging publishers to leave games in an end-of-life playable state, formalized exactly this consumer grievance, per campaign organizers' public counts.
Related stories: Every Genre Is a Mod: How Counter-Strike, DOTA and Battle Royale Escaped Their Hosts · Buying Unfinished Games on Purpose: How Early Access Rewired Trust.
Why did players get fatigued, exactly?
Fatigue is often described as a mood, but it has identifiable components. First, structural sameness: the battle pass / daily quest / seasonal event loop was standardized so thoroughly that genre differences became cosmetic. Second, opportunity cost made visible: every hour on a new service game is an hour not spent on an established one, and players increasingly refuse that trade without a standout hook. Third, betrayal asymmetry: a cancelled single-player franchise disappoints, while a shutdown wipes a game players are mid-relationship with. Helldivers 2's 2024 breakout and the durable success of Fortnite, Roblox-scale platforms and Warframe show the model itself still works — for titles that either arrive with an identity or have spent a decade earning trust. What died is the middle: competent service games asking for a main-game commitment without offering a reason.
Which live games are actually safe to commit to?
- Massive, profitable incumbents. Fortnite, League of Legends, Warframe-scale games with decade track records and proven revenue. Shutdown risk is lowest where the cash flow is largest.
- Games with an offline floor. If the campaign or PvE mode runs without company servers, your worst case is a normal game with a dead shop attached.
- Nothing at launch. The Concord pattern says wait: the discount arrives within months, the population question resolves within weeks, and your initial $40 saves you the disappointment premium.
For the industry side, the correction is already visible in hiring and portfolio plans: studios built for seasonal content pipelines have been restructured toward smaller teams or absorbed into support roles on the winners, and 2025's release calendars show fewer mid-budget service bets than 2022's. The remaining question is architectural, not commercial — whether end-of-life obligations, like an offline patch or private-server tooling, become a standard part of the launch checklist. Publishers who answer that before regulators or petitions force the answer will hold a trust advantage that no battle pass can price.
Verdict: live-service is not dying — it is barbelled. The top handful of platforms get stronger while the mid-budget 'also a service game' tier collapses, and 2024-2025's shutdowns were that collapse completing itself. Buy service games like subscriptions, not like products: commit only to incumbents or offline-floored titles, and treat every launch-week purchase as a bet the publisher has not earned.