Evergreen or Evergrave: 10 Genres That Keep Paying, and 10 That Will Eat You Alive

The Ten That Will Eat You Alive — I've Got the Scars

Okay, before I show you the survivors, let's walk past the bodies. These ten look so tempting from the outside, and every single one of 'em has a graveyard full of indie teams who thought they were the exception. I thought I was too.

I'm saying this from the trenches, friend. I built an open-world game in Unreal Engine 5 — alone — and let me tell you, the map design and scene scale broke me in ways I still feel in my bones. Every empty field I placed was a promise I couldn't keep. And before that, I wrestled with mechanics-driven sandboxes: I'm not kidding, a single inventory system is enough to swallow months of your life, let alone weaving a dozen systems into one sprawling sandbox. I own every scar on this list, personally.

FPS / TPS. The genre that eats studios whole.

Look, here's the number that made my stomach drop: 931 shooters released on Steam in 2023; the new successes made up 0.3% of that cohort, and 80% of the genre's revenue comes from old games (Naavik). And the graveyard writes itself, you've seen it: Spectre Divide was venture-backed, a hundred people, promoted by Shroud — shut down within months (PCGamesN). Remedy's FBC: Firebreak fell from 1,992 concurrent players on day one to 309 a week later. The Finals lost 80% of its players in two months. As the analysts put it: all recent breakthroughs in the FPS market have been incubated by mega-publishers. Gun feel, netcode, anti-cheat, servers, matchmaking — it's a machine built to grind small teams like ours to dust. And here's the sneaky trap I want you to watch for: the genre is so popular that every new dev feels the pull, the same way we all feel the pull of making a platformer. Popularity is exactly the problem — you'd be walking into the most competitive room in gaming with the smallest toolbox.

Open World (3D). The scope trap with the biggest sign.

My Unreal Engine 5 experience in one sentence, and I'm still a little haunted by it: the map was the product, and I was one person. The content math is just brutal — a hundred hours of bespoke content is mathematically impossible for a team of four, let alone solo like I tried. The tech stack is worse: streaming, LODs, landscape tools, each one a project in itself. UE5's World Partition system only pays off past 4km², and even the people who use it professionally say a two- or three-person team shouldn't bother (Hyperdense). The industry consensus stings, but I've gotta quote it: a lone developer aiming to create an open-world RPG is delusional (Wayline). The ceiling sits so high the floor never gets finished — I learned that the hard way, one empty field at a time. The open world doesn't just fail slowly; it fails expensively, because you only see the failure after years of work when your map is still 10% full and you're out of money.

Large-Scale 3D Sandbox. Same dream, worse math.

Oh man, this one's personal for me. You're not competing with an idea here, you're competing with Minecraft — hundreds of millions of copies and a decade of free updates — plus Rust and Valheim, teams of ten to twenty with years of runway. And the systems problem is the silent killer that got me: I built sandbox systems once, and I'm telling you, one inventory system was a project on its own. A sandbox is a dozen projects wearing a trench coat — inventory, crafting, building, physics, AI, all woven together, and the moment any one of 'em is janky, the whole sandbox feels broken. The proof that the dream is real is Terraria — and Terraria is 2D pixel art made by a handful of people. Same dream, deliverable size. That's why I've put the 2D version on my good list below, you'll see.

Hidden Object. A nostalgia market with no new customers.

I'll keep this one short 'cause the numbers say it all: 0.39% success rate — 5 of 1,290 games released in 2024 reached 1,000 reviews (Zukowski). The folks who loved these games already own the classics from a decade ago, and the genre's design language hasn't changed enough to pull in new players like you. It's a museum of a genre, and I've gotta be honest with you, museums don't pay rent.

VR-Only. The platform that won't arrive.

I wish I had better news here, I really do: 0.43% — 2 of 467 games reached 1,000 reviews. Chris Zukowski, who tracks this data every year: "VR games still underperform and they don't seem to be improving." The hardware base is tiny, the dev cost is high, and the platform keeps promising mass adoption next year while the graveyard keeps filling up. The numbers have been flat for years, and flat isn't a trend you wanna build your career on, trust me. Build for the screens people actually own.

Souls-likes. You're fighting FromSoftware's polish with a fraction of their team.

Here's the painful truth: 0.83% — 3 of 363. Your audience has zero tolerance for jank, because the genre's entire appeal is combat precision, and precision is exactly what small teams like ours struggle to fund and polish. Every hitbox, every animation frame, every boss tells players you're not Elden Ring — and they'll be right, and they'll refund. The genre looks like a shortcut 'cause the mechanics feel familiar, but familiar mechanics with worse execution is the definition of a lost bet. Don't take that bet with your rent money.

RTS. Three mountains before your first player.

Ugh, I feel for anyone tempted by this one: 0.82% — 4 of 488. AI, balance, matchmaking — each one's a full project on its own, and the genre's attention is locked onto franchises with decades of content and established esports scenes. You'd be building the whole mountain before you find out if anyone even wants to climb it, and the climb is a multi-year commitment before your first player ever joins a match. That's years of your life before you get a single signal.

Social Deduction / Party Games. The anti-evergreen.

The canonical corpse here is Among Us, and it haunts me: roughly $86M of revenue in 2021 collapsed to about $1.5M by 2025 — a 98% decline (Business of Apps). These games depend on a critical mass of players that only exists during a viral spike, and spikes are powered by streamers who move on the moment the next game shows up. I'm begging you to remember this: a spike is not a long tail; it's weather, and weather ends. The game is only alive while the crowd is there, and crowds are the one thing you can't build with code, no matter how good yours is.

MMO / Live-Service. You're not making a game, you're running a service.

Think about what you're really signing up for: servers, content treadmills, anti-cheat, patch cadence, community management — you're suddenly operating a 7x24 business on an indie budget. The game only exists while you're paying for it, which means your savings account is the server. That's exactly why I'm such a believer in serverless co-op design — the Lethal Company model — it's a lifeline for small teams like us: the game keeps existing when you stop paying. If your design can't survive without you paying a hosting bill every month, your design is the wrong shape. Kill it now before it kills your wallet.

Licensed Sports / IP Games. The door that's locked.

I know the fantasy, I've had it too, but here's the wall: you can't get the FIFA or NBA license, and the audience wants precisely the license — a football game without real teams is a football game nobody asked for. The "Batman" tag is statistically one of the best-performing tags in Steam history, and every single game under it is backed by a giant IP with a giant budget (indienova). The door doesn't open for indies like us, so stop rattling it — build your own door instead.

And one meta-warning before I close out this graveyard, 'cause I see friends fall for this constantly: fad fast-follows. Vampire-Survivor-likes went from 17 hits in 2023 to exactly one in the first half of 2024 (Game World Observer). The gold rush works for the first person per idea; everyone after 'em is buying shovels in a mine that's already played out. My rule for you: a fad is worth at most a two-month side project — box in the time, don't over-invest — never a strategy.

Every corpse in this graveyard died the same death, I swear: the genre was the dream, and the dream ate the budget.