Seven years ago, Epic launched a storefront with a simple pitch: developers keep 88% instead of 70%, and that better deal would eventually reach players. It was a genuinely reasonable argument, backed by real money, aimed at a near-monopoly nobody had successfully challenged.
In 2026, Steam holds roughly 74% of PC storefront revenue. Epic sits around 3%.
That looks like a rout, and most coverage treats it as one. I want to argue something more uncomfortable: the Steam vs Epic Games Store question isn't settled by those numbers, and judging it purely on market share means judging it from the shareholder's chair rather than yours.
The numbers, and what they hide
| Steam | Epic Games Store | |
|---|---|---|
| Share of PC storefront revenue | ~74% | ~3% |
| Monthly active users | ~132 million | ~78 million |
| Peak concurrent players | 41.81 million (January 2026 record) | Not comparably reported |
| Year-on-year growth | Steady | ~57% |
| Developer revenue share | 70% (tiered up at volume) | 88% |
Look at the second and third rows together, because that's where it gets interesting.
Epic has 78 million monthly active users against Steam's 132 million. That's not a rounding error — it's roughly 59% of Steam's audience. And yet Epic converts that into about 3% of the revenue.
The reason is the free game programme. Epic has spent years giving away a title every week, and it worked exactly as designed: enormous numbers of people installed the launcher and claimed free games. What it didn't do is change where those people spend money. They claim the freebie and buy on Steam.
Epic bought attention and assumed it would convert into habit. It bought attention and got attention. Habit turned out to be the expensive part.
The 57% growth figure is real and worth respecting — but growth from 3% is a much easier trick than growth from 74%. Doubling a small number is arithmetic, not momentum.
Why Steam won, and it isn't the games
The standard explanation is exclusives — Epic paid for them, players resented it, Steam kept the goodwill. That's part of it and it's not the main part.
Steam won on accumulated function. Two decades of features nobody thinks about until they're missing:
- Cloud saves that reliably work. Sounds trivial until you lose a fifty-hour save.
- Workshop. One-click mod installation, which for some games is the entire reason the game still has players.
- Reviews with hours-played attached. The single most useful purchasing signal in gaming, and one Epic still doesn't match.
- Forums and guides. Where you go when a game breaks at 1am.
- Family sharing, remote play, controller configuration, streaming to a handheld.
- A refund policy with clear rules — under two hours, under two weeks, no argument.
None of that is glamorous. All of it is why people stay. Epic spent a fortune on exclusives while the actual moat was a decade of unsexy plumbing, and by the time it started shipping comparable features the habit was set.
The refund policy deserves particular credit. It's the clearest rule in PC gaming: if a game doesn't work or isn't what you thought, you get your money back, no discretion involved. That's a genuine consumer protection Steam adopted years before it was forced to.
What Epic actually got right
Now the part that gets skipped, because dunking on Epic is easier.
The revenue split moved the whole industry
Epic's 88/12 split was the first credible pressure on a 70/30 standard that had gone unchallenged for over a decade. Steam responded with tiered rates that improve at volume. Other platforms adjusted. Developers across the industry now negotiate against a better benchmark than they had in 2018.
That's a structural win for the people making games, and it happened because someone with enough money was willing to lose some of it. Epic doesn't get credit for this nearly often enough, and it's worth more to the ecosystem than its market share suggests.
The free games were genuinely free
No strings, no subscription, no trial window. Claim it and own it. Over the years that's put a substantial library in front of people who couldn't otherwise have afforded one, including a lot of players in regions where a sixty-dollar game is out of the question.
I'm not going to sneer at that. It's the most straightforwardly generous thing a storefront has done for players in years, even if the commercial logic behind it didn't pay off.
Unreal Engine funds it
Worth understanding, because it explains why Epic can persist at 3%. The storefront isn't the business — the engine is. Epic can run a loss-making store indefinitely in a way a pure retailer never could. Which means Epic doesn't go away, and the pressure on Steam doesn't disappear either.
The exclusives fight, in hindsight
Time for an honest reassessment, because I was among the people annoyed about this at the time.
Epic paid for timed exclusives. Players hated it, review-bombed the affected games, and the strategy became the store's defining reputation. Seven years on, my view has softened without reversing.
The criticism was correct. Exclusivity on an open platform is a tax on players. On consoles, exclusivity is at least tied to hardware you bought. On PC, where every store runs on the same machine, it exists purely to force a launcher install. That's friction manufactured for leverage.
The reaction was disproportionate. Review-bombing games — punishing developers who took a deal that funded their studio — was aimed at the wrong target. Small teams took Epic's money because it de-risked their project. Attacking them for that was a failure of aim, and some of those studios took real damage from it.
The strategy is largely wound down now, which tells you Epic reached the same conclusion. It bought installs and generated resentment, and resentment turned out to be more durable than the installs.
The third option nobody mentions
Every comparison of these two skips the store I'd actually argue is the most player-friendly: GOG.
GOG sells games without DRM. You download an installer, you keep it, and it works whether the store exists tomorrow or not. No launcher required, no authentication, no account check.
That matters more than it sounds, and here's the uncomfortable framing: on Steam and Epic, you don't own your games. You own a licence to access them through a service. If that service ends, changes terms, or bans your account, your library goes with it. This is stated plainly in the terms of both.
For most people that risk is theoretical and the convenience is worth it. I use Steam daily and I'm not pretending otherwise. But GOG is the only major storefront where "I bought this" means what it normally means, and for anything you genuinely want to keep, that's worth the occasional inconvenience.
For games you consider permanent — the ones you'll reinstall in ten years — check whether a DRM-free version exists and buy that one. It usually costs the same. The difference only becomes visible on the day it matters, which is exactly too late to act on.
The features that decide it day to day
Market share is an abstraction. Here's what you actually bump into.
Refunds
Steam's policy is the clearest in the industry: under two hours played, under two weeks owned, refunded without an argument. It's automated, it's predictable, and it functions as a de facto demo system for a medium that stopped shipping demos.
Epic operates a comparable window. The difference is that Steam's has been in place long enough to be culturally understood — people buy more adventurously because they know the exit exists. That's a real behavioural effect, and it took years to build.
Reviews with hours attached
Steam shows how long a reviewer played before writing. That single piece of metadata does more work than any score: a negative review from someone with four hundred hours means something entirely different from one written after twenty minutes.
Epic added user reviews late and they carry nowhere near the same signal. For a genre like strategy, where problems only surface deep into a campaign, playtime-weighted reviews are the difference between an informed purchase and a guess. It's the feature I'd miss most.
Family sharing
Steam lets household members play from a shared library. In practice this is one of the most valuable features either store offers and it rarely appears in comparisons — probably because it doesn't generate revenue, which is also why nobody rushed to copy it.
Wishlists and price tracking
Both notify you when a wishlisted game goes on sale. Steam's wishlist has a secondary function worth knowing: developers watch wishlist counts to gauge demand, and for small studios that number genuinely influences whether a project gets continued funding. Adding a game costs you nothing and does something.
Which should you actually use?
The practical answer is all of them, and the friction of that is lower than people pretend.
| Situation | Store |
|---|---|
| Default purchase, anything with mods or a community | Steam |
| The weekly free game | Epic — it costs nothing |
| Games you want to own permanently | GOG |
| A significantly cheaper price during a sale | Whichever is cheaper |
| Playing on a Steam handheld | Steam, for the compatibility layer |
The "I only use one launcher" position is a preference dressed as a principle. Claiming a free game weekly from Epic has never once cost me anything, and over the years it's added up to a genuinely large library.
The one real caveat: if you play mostly on a Steam handheld, buying elsewhere adds friction with the compatibility layer. That's a legitimate reason to consolidate — I go into how that shapes the platform decision in PC vs console.
Mike's take
Steam won and deserved to, on the strength of two decades of infrastructure rather than any single decision. Cloud saves, Workshop, reviews with playtime attached, and a refund policy with actual rules — that's the moat, and Epic spent years attacking a wall that wasn't there.
Epic still mattered. The 88/12 split forced a revenue conversation the industry had avoided for a decade, and the free games put real libraries in front of people who couldn't afford one. Losing on market share isn't the same as accomplishing nothing.
Where I land: buy on Steam by default, claim Epic's free games because they cost nothing, and buy anything you want to keep permanently from GOG. Store loyalty benefits the store, not you.
What I'd actually want next
Not more competition on exclusives — that fight was bad for everyone. Competition on the things that affect you after the purchase.
A store that guaranteed offline installers. One that let you transfer a library. One that committed to keeping games playable after delisting. Those would be genuine differentiators, and none of them require paying a publisher to withhold a game from you.
Until someone tries that, Steam's position is safe — not because it's unassailable, but because nobody has attacked the part that actually matters.
For the practical side of PC gaming, the build guide covers where your money goes furthest, and the monetisation checklist applies the same scrutiny to what happens inside the games once you've bought them.