Home
Entrar
Registrar
Conteúdo de Aprendizagem
Loading...
Praticar Escuta
Praticar Escuta
/
Video
/
The Infographics Show
/
The $50,000,000 Employee. Why Electronic Arts Is DEAD
The $50,000,000 Employee. Why Electronic Arts Is DEAD
Selecionar modo de aprendizagem:
Ver legendas
Escolher palavra
Reescrever palavra
Highlight:
3000 Oxford Words
4000 IELTS Words
5000 Oxford Words
3000 Common Words
1000 TOEIC Words
5000 TOEFL Words
Legendas (241)
0:00
The gaming industry is on fire. Thousands of jobs are being cut. $200
0:04
million projects are being scrapped. Behind closed doors, CEOs are panicking. But in a small office
0:10
in Bellevue, 360 people are making more money that the entire workforce of Electronic Arts combined.
0:17
They aren’t just winning. They are changing the way we
0:20
think modern companies are supposed to work. And if they're right, then the entire gaming
0:24
industry isn’t just struggling. It’s operating on a lie.
0:28
CHAPTER 1: The Ghost in the Machine In the first half of 2024,
0:33
11,000 lost their jobs disappeared across the top 10 public gaming companies. Microsoft shut
0:38
down Arkane Austin. Tango Gameworks was gone almost overnight. Embracer Group
0:43
closed entire divisions across the world. Their restructuring dropped over 4,000 staff and shut
0:49
44 studios. Sony wasn’t spared either. Internal restructuring hit its PlayStation ecosystem,
0:54
with studios quietly shuttered or absorbed. Across the entire industry, the pattern was identical. It
1:00
seemed like no one seemed immune. Except for one private
1:04
company in Washington State. Valve's headcount stayed locked between
1:08
330 and 360 people. There were no layoffs or hiring sprees. And there was no quarterly drama.
1:15
The numbers back up Valve’s status. Steam alone generated over $16 billion in
1:20
revenue in 2025. The total business was projected to clear $17 billion. Divide that by their tiny
1:27
workforce and it’s a per-employee figure approaching $50 million. Microsoft, with its
1:32
operating system, cloud empire, and Xbox combined, generates roughly $1.24 million per employee.
1:39
Valve runs 40 times higher. Court documents from a 2021
1:44
antitrust suit revealed something even uglier for the rest of the industry.
1:47
The average Valve worker took home around $1.3 million annually. That figure dates back years.
1:54
The number today is almost certainly larger. But this isn’t just a story about compensation.
1:58
And it isn’t about luck either. Because when you look closely,
2:01
the difference isn’t performance… it’s structure. Public gaming companies, tied to investor
2:06
expectations, inflated their headcounts during the pandemic boom. They hired aggressively,
2:12
then cut just as fast when the market turned. When things went south, they didn’t adjust,
2:16
they bled people out in waves. Valve didn’t follow that cycle.
2:21
Their staffing level behaves like a hard ceiling, not a growth target.
2:25
It’s a structure that stays within its own limits. What does that say about the rest of the industry?
2:29
The $200 billion gaming sector has been preaching scale for decades. Bigger studios,
2:35
teams and budget. But this playbook may be the disease, not the cure. The early evidence comes
2:40
from the only company refusing to play along. CHAPTER 2: The Zero-Dollar Marketing Miracle
2:46
In modern AAA gaming, budgets don’t just get big anymore… they get extreme.
2:52
Sony’s Concord became one of the clearest examples of that. Reports place the development deal at
2:57
over $200 million. Some estimates push the total closer to $400 million. And that doesn’t even
3:02
include marketing, studio acquisition costs, or the launch push. 8 years of development went into
3:08
it. The studio behind it, Firewalk, was bought outright by Sony in April 2023. They bet big
3:14
on Concord as their next big live-service hit. The day Concord launched on Steam, 697 players
3:21
showed up. The title sold around 25,000 copies. Sony pulled it off the shelves 14 days later and
3:27
issued refunds to everyone who paid. Sony shut Firewalk down entirely 2 months after
3:32
that. Years of work. Hundreds of millions in capital. All wiped out in a matter of months.
3:38
Around the same time, Valve added a project called Deadlock to Steam. It was invite-only.
3:43
There was no reveal trailer, no press tour or IGN exclusive. There was no marketing budget,
3:48
period. Reviewers who tried to write about it were asked not to. Word spread through
3:53
Discord servers. On September 2nd, 2024, Deadlock peaked at just over 171,000 concurrent players.
4:00
Concord cost Sony roughly $287,000 per peak concurrent
4:05
user. Deadlock cost Valve basically nothing. The pattern extends beyond these two titles.
4:11
Helldivers 2 broke out early in 2024, almost entirely through word of mouth. It sold over
4:17
12 million copies in just three months. Most of the marketing was organic memes about a fictional
4:22
planet called Malevelon Creek. Manor Lords, built largely by a single Polish developer,
4:27
moved over 2 million copies in its first 3 weeks. The game had almost no advertising
4:32
spend, just pure community momentum. So, does massive marketing spend actually
4:36
push away the players these games want? When Sony spent the GDP of a small island
4:41
nation broadcasting Concord, what their target audience actually heard was different. They heard,
4:47
"this is corporate, and it's trying very hard." Real interest doesn't grow out of billboards,
4:51
it grows through excitement and word of mouth. Valve built their entire approach around
4:56
exactly that idea. Access gets earned,
4:58
not bought. It clearly works. Smaller studios have already started copying the format. The
5:03
advertising-fueled launch playbook is dying. Something better has finally been proven to work.
5:09
CHAPTER 3: The Output Anomaly What’s most unusual about Valve isn’t what they
5:14
ship… it’s how cohesively everything operates. Take Counter-Strike 2. This wasn’t just a
5:19
sequel. It meant a full migration to the Source 2 engine, a rebuilt lighting system,
5:24
and a complete overhaul of netcode for the most-played competitive shooter on Earth.
5:29
Then there is the hardware. The Steam Deck OLED wasn’t just a refresh;
5:33
it was a new revision of custom handheld hardware designed around Steam’s ecosystem
5:38
itself. Even their older work still behaves the same way over time. Half-Life: Alyx remains the
5:43
benchmark for modern VR years after release. Each piece feels separate at first glance,
5:49
but together, they don’t behave like isolated projects.
5:52
They behave like one system that never really stops.
5:56
And it all runs on roughly 360 people. All while operating the largest digital store on PC,
6:02
and running several of the biggest live-service games on the planet.
6:05
Now compare that to the industry standard. At Microsoft, EA, or Activision, it's not
6:10
unusual for a single AAA blockbuster to require larger teams than Valve's entire
6:14
workforce. And even with those resources, those giant games still suffer from massive delays.
6:20
How? The answer is simple: Valve
6:22
keeps stacking advantages. Every project builds on something already in place. Counter-Strike 2
6:27
didn’t need to build an audience, it inherited one formed over two decades. Deadlock didn’t
6:32
need to solve payments or distribution, Steam already handled it. The Steam Deck didn’t need
6:36
to create a software ecosystem from scratch, Proton was already doing the heavy lifting.
6:41
Nothing starts from zero. And because of that,
6:44
Valve doesn’t scale its team the way other companies do. It stays small by design.
6:49
There’s no boardroom bureaucracy, no focus group revisions or shareholder reviews.
6:54
Smaller teams pulling in the same direction tend to beat massive ones more often than people think.
6:59
And the opposite is worth paying attention to. When a company announces it’s hiring thousands
7:04
more developers, it can sound like momentum. But it doesn’t always lead to faster shipping.
7:08
Sometimes it’s the opposite. In some cases, it can mean slower
7:12
results further down the line. CHAPTER 4: The Shadow Hierarchy
7:16
There’s a polished version of Valve that gets repeated in tech articles. The flat hierarchy. The
7:22
rolling desks. The employee handbook that promises you can move to whatever project excites you most.
7:28
It’s a great story. But it’s a half-truth,
7:31
at least to people who've worked there. Glassdoor reviews and ex-employee accounts
7:35
describe something the brochure doesn't mention. Cabals. Not in the conspiratorial
7:40
sense. But there’s long-tenured groups of senior engineers, designers, and economists.
7:45
They quietly decide which projects get funded and which get shelved. Which new
7:49
hires get integrated, and which get iced out. One former employee compared the social dynamic to
7:54
"a high-school cafeteria with budget authority." Another described being technically free to work
7:59
on anything, while being unable to get anyone to collaborate on the project they chose.
8:04
Even without formal titles, there is still power inside the company. Pay can swing wildly based on
8:09
peer perception. Coworkers evaluate each other and bonuses or influence are determined by it.
8:15
If someone can’t read the unwritten rules of the room, they usually don’t last long.
8:19
The freedom is real,but so is the pressure surrounding it.
8:22
For the people who fit, Valve offers an unusual amount of autonomy and enormous
8:26
pay. For the people who don’t… the exit tends to happen quickly.
8:30
A public company would never survive this culture. Compliance and HR teams would step
8:35
in. There would be a raft of class actions. The intensity that makes Valve's team punch above
8:40
its weight doesn’t work inside public company rules.The system works, but not in the clean,
8:45
idealistic way the mythology suggests. There’s a harder edge underneath it.
8:50
Pretending otherwise ignores the people who actually had to live inside it.
8:54
But the model proves something the rest of the industry has gotten lazy about. Hiring
8:58
small teams of really good people who fit the culture beats hiring big just
9:03
to hire big. Nobody is saying go copy the cabals. The real lesson is easier.
9:08
Lowering your hiring bar to grow fast quietly damages the company. A company that stays
9:13
small on purpose hires carefully. It protects what makes it work.
9:17
That company may be doing more for its people than one that tripled its team to look good on
9:22
LinkedIn… especially when half got laid off 2 years later. The setup favors the smaller,
9:27
harder operation. Workers actually get paid more and output increases.
9:32
Whether it works as a business is no longer up for debate.
9:35
CHAPTER 5: The Passive Empire But the real scale of Valve
9:39
doesn’t come from selling games. Steam takes a cut of every game sold
9:43
on the platform, historically around 30%, scaling down for high-volume publishers.
9:48
That alone is a massive business. But it’s only the surface layer.
9:52
There are also marketplaces for in-game items, especially in Counter-Strike 2
9:56
and Dota 2. Players trade weapon skins, charms, and cosmetics in volumes that start to resemble
10:02
financial markets more than game economies. And Valve takes 15% of every transaction.
10:08
The Counter-Strike case-and-skin economy alone made over $1 billion in 2023.
10:13
The income arrives whether or not Valve releases new content. During peak periods,
10:18
the marketplace alone is estimated to generate serious revenue. At times, more than the
10:23
entire annual earnings of mid-sized publishers… And none of it comes from releasing a new game.
10:29
The strategic payoff is huge. Most games companies are stuck. Every quarter they
10:33
have to ship something new just to pay the bills. Valve doesn't. Its systems create
10:37
something like a permanent cash floor. That lets them experiment, kill or pause projects,
10:42
and wait a decade if they want. They have no pressure to ship sequels. They can afford to
10:47
wait until something is interesting. Compare that to a publisher burning hundreds of millions
10:51
every quarter. Their boards demand progress across multiple live-service projects at once.
10:56
The difference isn’t just scale. It’s breathing room.
10:59
So why hasn't Valve gone full Microsoft and bought every studio in sight?
11:04
The restraint itself is part of the strategy. Acquisitions force you to merge cultures and
11:09
expand headcount. They open up new legal risks and management problems that Valve has spent 30
11:14
years working to avoid. Public companies pushed by dividend demands or shareholders almost always
11:20
deploy excess cash on something. Valve just lets the money pile up and sink it into something that
11:25
appeals. Real independence means owning your own distribution, not renting it from someone else.
11:31
So what does that mean for studios stuck in stores they don't own?
11:35
The answer is not great. But the framing makes things much clearer for anyone trying to figure
11:40
out where the real advantages actually are. CHAPTER 6: The 10-Year Gamble
11:45
Most people remember the Steam Deck launching in 2022 and being a hit.
11:49
What gets lost is everything that came before. The full version of SteamOS eventually launched
11:54
alongside the original commercial Steam Machines in late 2015. The standalone controllers only
11:59
sold 500,000 units by mid-2016, and the actual consoles sold far worse than that.
12:05
The whole project looked like a flop. A normal company would have killed
12:08
it. Google launched Stadia in 2019 and buried it just 3.5 years later.
12:13
Amazon's Crucible barely lasted 6 months. Valve didn’t kill SteamOS, they kept it alive.
12:19
They built Proton, software that lets Windows games run on Linux without the developer doing
12:25
anything. They made parts of it open to everyone. They kept improving Linux in the
12:29
background by contributing directly to it. Most importantly, they didn’t rush it.
12:33
For almost a decade, it looked like an expensive side project.
12:37
Then, on February 25th, 2022, the Steam Deck shipped. It didn’t just sell,
12:42
it created a whole new category. Suddenly the decade of Linux work was the foundation of a
12:47
hardware ecosystem. One that was chipping away at Microsoft's grip on PC gaming.
12:51
The assumption that Windows is the only place serious games can survive started to crumble.
12:56
Every one of those purchases pulled PC gaming further away from Microsoft's old revenue model.
13:01
It worked so well that competitors began scrambling, trying to catch up to a future
13:06
Valve started building years earlier. ASUS, Lenovo, MSI, and others have since developed
13:11
their own Windows-based handhelds. You can’t do this as a public company.
13:15
Shareholders wouldn’t entertain the idea of spending 10 years building
13:19
an operating system. They would want to see numbers in the profit column.
13:23
Valve doesn’t have that problem. Gabe Newell reportedly owns over 50% of the
13:28
company. That alone changes everything. There’s no board pushing for a different direction. There’s
13:33
no pressure to show short-term wins at the expense of long-term work. Decisions don’t have to look
13:38
good in the next financial report. They just have to still make sense years down the line.
13:43
Patient money, when it's actually used patiently, beats short-term money by a mile. It's not even
13:49
a fair fight. Founder-controlled companies, dual-class shares, and staying private may be
13:53
the only ways to make the long game work. People are starting to take notice because more of those
13:58
structures are emerging across tech now. The playbook is now clear enough to copy.
14:03
CHAPTER 7: The Last Gatekeeper Valve markets Steam as an open platform. Anyone
14:09
with a developer account and $100 can publish a game. The store famously refuses to curate,
14:15
instead the market should decide what succeeds. That stance has won them goodwill from indie
14:20
developers for a generation. But it’s not the whole story.
14:24
Steam's recommendation algorithm quietly buries thousands of low-effort releases every year. The
14:29
recommendation engine decides which games get shown to which users. Asset-flipped titles get
14:34
buried. The AI-generated slop games that have flooded every major storefront also get buried.
14:39
Valve now requires disclosure for AI-generated content and any game that breaks the basic quality
14:45
rules gets banned outright. So the idea that Steam is “non-curated” doesn’t really hold up.
14:51
It is curated. Just not in the obvious way.
14:55
That puts Valve in a tricky spot. They get a huge boost from being seen as the indie
14:59
developer's friend. The platform that lets the market speak. But they also have to protect
15:04
that platform. Otherwise the store drowns in 50,000 annual releases of drag-and-drop
15:09
junk. The two goals aren't fully compatible. So Valve does what public companies can't pull off.
15:15
They make a quiet, practical compromise. They take the small hit of looking a bit
15:19
hypocritical and then they move on. Are they protecting the ecosystem,
15:24
or are they the gatekeeper who decides which developers get to eat?
15:27
Both, probably. The platform's scale is what made
15:30
the slop flood possible in the first place. Over 20,000 games hit Steam in 2025 alone. Valve also
15:36
built systems that actively shape how that ranking works. It’s not random. It’s structured. The whole
15:41
setup is messy, but it keeps Steam’s discovery feeling intact. Games still feel like they can
15:47
break through on their own. Even as the store quietly filters, sorts, and surfaces what rises
15:51
to the top. While newer storefronts struggle under the same explosion of content, Steam doesn’t.
15:57
It adapts. This type of moderation might
15:59
be the only way platforms this size last without turning into heavy-handed publishers or becoming
16:05
chaotic dumping grounds. Valve has shown a third, messier path. Sure, it doesn’t quite match their
16:10
own marketing, but it works. CHAPTER 8: The Pivot
16:14
None of this really makes sense in isolation. Valve ships more than its peers with a fraction
16:20
of the staff, does it without marketing, runs a culture that wouldn’t survive
16:24
public ownership… but still produces results that work. It’s not a structure
16:28
that’s replicated in the big name studios. Any one of these looks impossible by itself.
16:33
But the combination is something else entirely. It works because Valve has refused, from day one,
16:38
to play the same game everyone else is playing. Just one founder, a handful of senior partners
16:44
and the freedom to ignore everyone's opinion on what a modern tech company should look like.
16:49
Most coverage of Valve treats this as something quirky. But there’s a clearer pattern underneath.
16:54
A lot of what feels normal in big companies only exists because of outside money. Quarterly
16:59
reporting. Acquisitions. Large staff numbers. Heavy marketing expenditure..
17:04
None of it is strictly required to make or distribute software.
17:08
A small group of people building what they actually want to build, with enough money to take
17:12
their time. That’s what makes it feel unusual. Why?
17:16
Because everyone else has been so used to the status quo, that they
17:19
forgot this one was even an option. CHAPTER 9: The Sovereign State
17:24
Valve isn’t really a company in the standard sense. It behaves more
17:27
like a small, private state. Headquartered in Bellevue,
17:30
it runs with its own internal rules that cover hiring, pay packages and long term strategies.
17:35
There’s nothing outside the company that forces it to behave to the industry norm.
17:40
And that’s the key difference. Valve doesn’t behave like a company trying
17:43
to win inside a market. It behaves like something that stepped out of the usual system entirely. And
17:49
once you see it that way, the decisions that look unusual from the outside start to make more sense.
17:55
Replicating Valve completely is almost impossible. Very few founders still own
17:59
a majority of a profitable software giant. Very few cultures have been tuned this carefully over
18:04
20 years. Their advantage is specific. Owning the biggest PC games store on
18:09
Earth is something only one company can do. But the rest? Companies can learn from it.
18:15
Smaller studios watched the public-market wreckage and decided they wanted no part of it.
18:19
Look at LocalThunk, the solo developer behind Balatro. He turned down buyout
18:24
offers and deliberately kept the operation small. Or Swen Vincke at Larian Studios,
18:29
the team behind Baldur’s Gate 3. He’s run the studio for decades without taking it public. They
18:34
recently shipped a generational hit - on their own terms - after walking away from publisher deals.
18:40
None of these companies will become Valve. They don't need to.
18:43
Each has picked up enough of the playbook to punch above their financial weight. Their
18:47
combined success quietly proves the bigger-is-better idea wrong.
18:52
Gaming isn’t finished. The industry layoffs aren’t even over yet, the next wave is probably already
18:57
in motion. AAA budgets are still blowing up past anything sustainable and marketing departments are
19:02
still pouring money into campaigns nobody asked for. The bleeding will go on. Some of the studios
19:08
closing right now were doing really good work. The people losing their jobs deserved better than what
19:13
their employers' money problems pushed on them. But somewhere in Bellevue, 360 people are
19:18
proving a point. Year after year.
