Valve Should Buy Discord

An Unintentionally Timely Piece

(Image generated with Gemini Nano Banana Pro)

Preface: this was supposed to be a shorter piece, a quick addendum of a thought that didn’t neatly fit into my last post and so would get broken out into its own fast-follow thing.

It grew.

Update before publishing: As I was putting the finishing touches on this monstrosity of an article, Bloomberg broke news that Discord has confidentially filed for an IPO. This doesn’t change my thesis — if anything, it sharpens it. Once Discord goes public, acquisition becomes significantly harder and more expensive. Below is the case for a deal whose window is actively closing, if not outright past.

---

In my last post regarding the significant parallels between Valve and Apple, I noted that one significant aspect where the parallels between the two break down is positioning within the social layer.

On the one hand, you have Apple, with iMessage and Facetime serving as soft lock-in within the Apple ecosystem, offering cross-device chat history, better integrations, and all the other features that make it “just work, seamlessly”... so long as the people you’re talking to also have Apple devices, of course. And heaven forfend you try to develop a way of undermining that lock-in.

On the other hand is Steam… which does technically have forums, Steam Chat, and groups built into its platform, but calling it “underutilized” would be polite. “Anemic” might be a better word.

No, the social side of gaming is dominated by Discord, and this is a (perhaps the only) critical market weakness for Steam long term. And that’s just one reason why Valve should buy Discord.

Financial Feasibility First

Before we dive into the arguments, we have to ask – is it even possible? Can Valve even afford to do this? Or, is Valve, as a poor small indie company, incapable of bringing the financial weight needed to buy a venture-backed social giant like Discord?

Valve’s Numbers

• Valve pulled in about $4 billion in revenue from Steam:

  • ~$17 billion in total store revenue…

  • Multiplied by a pessimistic factor of 23% (Valve takes either 30, 25, or 20% of sales depending on each game’s revenue)...

  • Gives roughly $3.9 billion

• The overall company profit estimate (and it’s only an estimate) is ~$2.5 billion

Discord’s Price

• Discord’s post-money valuation following their most recent funding round was $15 billion, though that was about 4.5 years ago

• Secondary markets estimate Discord’s current FMV at ~$7 billion.

Comps

Slightly less than 3x annual profit is hardly bank breaking in the world of tech M&A.

Facebook’s purchase of Whatsapp in 2014 was for $19 billion at a time when their profit was ~$1.5 billion. That’s a multiple of almost 13. Zuckerberg paid almost 13 years of Facebook’s at-the-time annual profits for a chat company with 55 employees that was burning $138 million per year (even discounting the stock-based compensation, it was burning ~40 million). Talk about value!

More recently, Salesforce’s acquisition of Slack was priced at ~$28 billion in a year in which Salesforce’s profit was ~$4 billion, meaning they paid (very roughly) 7 years of their own profit to buy Slack.

Even taking Discord’s funding round valuation of $15 billion, it’d be slightly less pricey (6x Valve’s annual profit) than Salesforce’s acquisition of Slack, and less than half of Facebook’s acquisition of WhatsApp.

Synergies

So, it is (or at least should be) financially possible. What’s the win-win on the table?

Discord’s Problems:

Well, just like the chat companies we talked about above, Discord is burning cash. In an attempt to stop burning cash, they’ve begun turning the metaphorical screws on users, heralding the start of the cycle of enshittification, including such beloved fan favorites as:

• Adding ads and brand promotions, even for paid subscribers;

• Introducing overpriced “profile customizations” that interfere with the “standard” profile features;

• Locking some basic features like profile banner images (a feature that’s free on any other social platform that supports it) behind an (at time of writing) 10 USD/mo subscription; and

• Getting rid of the middle-ground “Nitro Classic” subscription tier a la Netflix

All of this comes despite Discord (or at least its CTO) wanting to avoid enshittification.

Valve/Steam’s Problems

Well, they don’t really have any… right?

Foreshadowing is a literary device in wh-

The Streaming Wars (but not the one you’re thinking of)

Let’s take a short digression and talk about streaming. Specifically, the livestreaming market, especially gaming related.

We can learn something from this space, but in order to do so, we need to first get on the same page as regards…

The Mechanics and Economics of Streaming.

Livestreaming video has most of the challenges of YouTube with the added complication of temporal synchronicity for everything:

• Giant bandwidth requirements (for sending video to viewers)

• Giant compute requirements (for processing/compressing video from streamers)

• That compute has to be fast so that stream data is going out as fast as it comes in, and it’s potentially coordinated with any special features your platform adds on top

• Storage isn’t as much of an issue because VODs can be shorter-lived than YouTube’s “up forever unless you get copystruck” model… though you still need some of it

As such, the same heavy capex and opex requirements associated with these mechanical requirements mean that the same heavy financial burden applies before profitability.

Since we’re talking about profit, how do you go about making money in this business? Well, so far, there are 2 (or maybe 2.5) avenues of monetization:

1. Ads (like other “free” internet services), ideally targeted based on data you get from knowing what streamers viewers follow and what games they’re interested in; and

2. Subscriptions to individual streamers/creators that the platform takes a cut of

   1. Platforms also have started introducing “donations”, which are one-time transfers from viewers to streamers (minus the cut going to the platform, of course) – the question of whether to treat that as the same as subscriptions, its own thing, or a hybrid is what makes it 2.5

What do all of these methods need to work? Unsurprisingly, users:

• Ads need people to look at them for value

• Subscribers require people who are invested enough in the streamers’ personalities to actually open their wallets

• Donations need the same as subscribers, though maybe with a lower barrier since it’s (theoretically) one-time rather than recurring

So we know how the market works; who are (and perhaps more relevantly, were) the major competitors?

Our Fighters

The undisputed 800 pound gorilla of this sector, owned by Amazon, is Twitch. Relative to everyone else we’re going to talk about, Twitch is the incumbent with first-mover advantage and a market stranglehold. Everyone else is trying to steal the eyeballs from Twitch. The ones that are still alive and trying to do so are:

• YouTube, serving as the ironic Bing to Twitch’s Google, i.e. the deep-pocketed, long-running competitor to the industry leader;

• Kick, the… “niche” competitor (that is the most polite phrasing I have); and

• TikTok Live, which has broken out of streams of the “adjacent” content TikTok is known for (i.e. beauty/fashion and music/dancing) by serving mobile game streaming (Garena Free Fire, Mobile Legends Bang Bang, PUBG Mobile), a segment which is not as present/popular on YouTube/Twitch

And then, 6 feet beneath some oddly shaped piles of dirt, we have:

• Mixer (bought and then killed by Microsoft in 2020 when it had insufficient growth); and

• Facebook Gaming, shut down in 2022 after previously serving as the offramp for Mixer

Let’s take a deeper look at them, shall we?

Mixer

• Signed exclusivity deals with big streamers (Ninja, Shroud)

• Their bet: bring streamers with large, invested existing viewerbases onto the platform, and they’ll “filter out” to other streams

• Verdict: utter failure

  • Mixer as a platform was unable to bring and keep users or distinct communities from these big signings, leading to stagnant growth

  • One analyst compared the whole affair to Google’s attempt to challenge Facebook with Google+, which strikes me as a touch harsh, but…

Facebook Gaming

• Leveraged existing userbase (i.e. Facebook users) by pushing streams based on games being played that might interest users

  • Also employed a similar (though less flashy) strategy signing names like DisguisedToast and ZeRo, who might not as recognizable outside of gaming but still had sizable fanbases

• Bet: enough people will come in via algorithmic discovery in main app, boosted by audience from imported streamers

• Verdict: kept up with the treadmill for a while before tripping and falling off

  • Solid 3rd place in mid 2022 (Twitch @ 76.7%, YouTube @ 15.4%, FG @ 7.9%; ref report)

  • However, when the high tide of COVID stream watchers receded, Facebook was (to quote Buffett) revealed to be swimming naked

  • What was the root cause? Facebook Gaming’s audience wasn’t invested in watching game streams, just there because Facebook brought them there via algorithmic fiat, and so didn’t stick around long

    • This view is corroborated by DisguisedToast, who quickly returned to Twitch after his exclusivity contract expired in Nov 2021, specifically because the community’s vibes were off.

Learnings

So, why did I go through those case studies? Well, Mixer and Facebook Gaming teach us some important things about social graphs in tech.

1. You cannot transplant communities from elsewhere and expect the users to just proliferate outwards (Mixer); and

2. You cannot force an existing audience to “change form” – your product has to match the existing audience (as TikTok did), not the other way around (as Facebook tried)

In short, social is hard to do when not grown organically. Just buying it outright (as Facebook did with Instagram and Whatsapp), while a blunt way of doing things, is far more likely to play well if you can’t directly compete.

And oh, look over here, we have a social media platform that completely organically came to dominate gaming and might serve as a wedge for challenging Valve’s dominance in the space! There’s that foreshadowing I mentioned!

The Threat

Money troubles aside, Discord owns the gaming community social layer in a way that even Twitch and YouTube don’t. Video game developers, whether small or large, make Discord servers to reach their players (e.g. SuperGiant Games). Other forums for gamers (e.g. Reddit) have “associated” Discord servers (e.g. the League of Legends subreddit has an official subreddit Discord). Esports teams/organizations organize fan events in their own Discord servers (e.g. Cloud9). In a world where authenticity and direct connection are strengths, Discord is the way gamers get those things.

By contrast, Valve’s social features, despite a 2018 push for improvement, are no comparison (e.g. no persistent servers, groups are a joke, in-client streaming is difficult to find and finicky, etc). Valve does not have any real social lock that mirrors even the limited lock-in of Apple’s Facetime and iMessage

Valve knows the importance of this; another quote from GabeN’s University of Texas talk that I opened my last post with a quote from openly acknowledges it:

We’re happy to do it if nobody else will do it, mainly because everybody else will pile on, and people will have a lot of choices, but they’ll have those characteristics. They’ll say, “Well, I could buy a console, which assumes I’ll re-buy all my content, have a completely different video system, and, oh, I have a completely different group of friends, apparently. Or I can just extend everything I love about the PC and the internet into the living room.”

(via Polygon as last time; emphasis mine)

He understood it a decade ago.

What does this mean? Put simply, Discord can be turned into a weapon against Steam.

Epic Games Store (EGS)

EGS was originally explicitly built to not have social features, supposedly based on feedback from developers:

“For instance, forums and other social media-like tools—a cornerstone of Steam—won’t be part of the package. Galyonkin said that this is because ‘not a single developer I talked to wanted forums’ and ‘the toxicity it brings,’ preferring to interact with communities on their own terms on platforms like Reddit and Discord instead.”

(via Kotaku)

In the year 2025 2026, both game devs on EGS and Epic themselves may be regretting that particular design decision:

> “There’s nothing to do there but to buy. So that will always lose to a shop that is also emotional.”

(Lead Developer of Witchfire, via FRVR)

Indeed, Epic themselves have already signalled a partial pivot on this front; their Trello Board has “Social Overhaul” in the “Future Development” category. That said, at time of writing, it seems to have been there for almost 5 years, so maybe not all that high priority… though one way to bump this out of the icebox “future development” category would be to buy/merge with a finished solution :P

Microsoft Xbox

Xbox Gaming on PC already has some integration and market space (Xbox Game Pass, Screen Capture/Streaming, etc). Microsoft has also attempted to match Epic’s attempt to attract developers by taking a smaller share of the pie of sales on the Microsoft Store.

However, while there is an Xbox Social included in those “Xbox Gaming on PC” integrations, I had no idea it existed until doing research for this. A big name purchase like Discord, though, would bring that feature into real prominence.

Both of these potential acquirers would instantly threaten Steam’s current dominant position. That might seem odd to say, but look back at that quote from the Witchfire dev: “There’s nothing to do [on EGS] but to buy. So that will always lose to a shop that is also emotional.”

Steam might have your achievements, your library, your cloud saves, and now the hardware of Steam Deck and Machine 2.0, but Discord has your conversations. It has your custom emotes. It has your friend group’s dumb jokes from 3 years ago.

If it came to it, which would you pick?

Aside: Discord vs Valve

With everything I’ve said thus far, you might reasonably wonder, why do I not think Discord can/should take Valve on directly, mano a mano?

That’d be because they already tried that once and failed.

Why did they fail? Well, there were definitely a lot of subtle, nuanced reasons, like:

• The fact that Discord users themselves seemingly didn’t know the store existed until they were informed it no longer existed;

• The downright baffling idea of seemingly only selling games via verified servers rather than a single storefront, kneecapping discovery; and

• The suggestion (see the earlier article re: Discord’s CTO and enshittification) that the company didn’t know what it was even doing with the storefront

All this, combined with the fact that the EGS and Microsoft Store are failing to take market share from Steam (which still holds strong at 75%) despite lucrative deals for developers, free games for customers, and market leverage (whether through Fortnite or direct integration on Windows), suggests that maybe running a digital game storefront isn’t as simple as Steam has historically made it look. At minimum, Discord’s been burned once trying to do it and doesn’t seem eager to try again.

Current Discord Stakeholders

None of this is going to come as news to industry insiders/veterans. Discord’s importance is apparent to everyone with an interest in the space. For example, MS previously had talks about buying out Discord (though they ultimately went nowhere).

However, Discord does have existing industry players as major stakeholders. For one, in response to Microsoft’s buyout talks, Sony began integration of Discord into PlayStation Network and bought a minority stake. It’s hard to see this as anything other than a defensive move to “have a seat at the table” in case MS tries to buy Discord again and cut them out of the picture.

Similarly, Tencent has been invested in Discord since 2015. Importantly, Tencent is also a significant minority (~35%) stakeholder in Epic Games.

Valve’s Palatability as Acquirer

Sony would probably prefer Valve to Microsoft as owner/acquirer. Valve has been making inroads into competing against the PS5 with the Steam Machine 2.0, which (while not explicit) is a “console substitute”. However, Valve takes a relatively open approach with their things (ref: Proton/SteamOS, which they’ve made free and usable by other “handheld gaming PC” manufacturers like Lenovo/ASUS). At present, it’s unlikely Valve’s acquisition would lead to PS5 being locked out of Discord.

Harder to say how Tencent would view Valve. My take is that, fundamentally, Tencent probably doesn’t care too much so long as they make money out of the arrangement; while they do have their fingers in a lot of gaming pies (full owner of Riot Games, maker of League of Legends; significant minority stake in Epic, as noted earlier; positions in Ubisoft and Roblox; previously had a stake in Activision-Blizzard-King before MS acquired it), they don’t seem interested in completely dominating the space (as evidenced by the seeming lack of objection to Microsoft buying ATVI). If given an option, though, they might prefer a buyout by/merger with Epic Games, since they’re also invested there, and could perhaps facilitate that themselves if they were so inclined as investors in both entities.

The Consumer Interest

I think it’s entirely possible that Valve buying Discord would result in ticker tape parades by gamers. While there will definitely be some hesitancy, both in the monopolization and due to Valve’s… questionable commitment to social (as well as the historic downsides consolidation has had for gamers), it’s hard not to see the upsides:

• Valve doesn’t (currently) do ads, and Steam has resisted the siren call of enshittification for the past 20+ years; this might represent an about face on a lot of the recent changes for the worse made by Discord in the name of raising revenue

• Valve is known for running a solid business and service, and could “subsidize” Discord’s cash burn while a better monetization model is found (iteration on Discord’s Nitro subscriptions? Integration between servers and games for developers that can be monetized (e.g. achievement announcements/titles/whatnot)? Just efficiencies by porting Discord off of GCP and onto Valve’s bare metal?).

• Valve is also the most likely to be relatively hands off of any potential acquirer

  • Discord going public would, like Reddit’s IPO, be a one way street to “revenue squeezing time” (Pre-publication note: And it seems that road is now being paved).

Speed Bumps: Potential Arguments Why Valve Shouldn’t Do It

So far, I’ve made it out like that this is a no-brainer. Obviously, the fact that it hasn’t happened yet is an indication that things aren’t quite so straightforward.

Valve Doesn’t Really Do Acquisitions

No, really.

They’ve acqui-hired some individual game developers and bought some IP rights to game properties (e.g. DotA, Portal, Team Fortress). They acquired an audio company by the name of Impulsonic in 2017. They’ve made some game developer studio acquisitions in the past (Campo Santo, Hopoo Games most recently).

That’s… it. That’s the whole list. Valve has made as many Half-Life games as they have major corporate acquisitions (which is saying something, considering Valve has made exactly three Half-Life games… and that’s if you count Alyx).

This is obviously not by accident, but because I like quotes:

“‘I don’t lie awake late at night saying “What should we do? What’s our competitive response?” for something like this, because it’s sort of orthogonal to any of the things that we think are interesting in terms of how Valve runs its business,’ Newell said.”

(via PCGamer)

Suffering From Success

Valve’s most eye-catching stat is their revenue and profit per employee – the 2025 estimate linked earlier was about $50mm/employee in revenue. Their dominance in the space is obviously part of that, but the other side of it is their limited headcount – they only have about 350 employees

Discord, by comparison, has roughly 3x the headcount of Steam. In the article I linked regarding Discord burning money, they mentioned laying off 17% of their employees, which at the time was 170 employees. Thankfully, that makes the math easy – Discord had roughly 1000 employees at that time.

Valve also famously has a “flat” corporate structure, which makes absorbing a large number of employees (especially at a ratio greater than 1) much harder to do without losing the culture that has made Valve so successful.

Social Culture Issues

Discord and Steam, despite having significant interest overlap, have very different cultures. While it wouldn’t be impossible to integrate the two, and it might not even be required (i.e. Valve could just let Discord communities mostly exist independently of Steam), it would be a challenge if Valve wanted to integrate the two more comprehensively (to e.g. implement the ideas about achievement-specific monetization in Discord servers).

Operational Challenges

Discord, as a social platform first, has and requires a much greater moderation arm (e.g. ID verification in response to things like laws against minors accessing social media, explicit/illegal content scanning, etc). While Steam might have some of that already (responding to DMCA claims against store entries, taking down forum posts that advocate violence), Valve has historically been very “hands off” when it comes to Steam’s social side.

Regulatory issues

An acquisition like this will almost certainly put Valve “on the radar” for regulators to look at and consider taking action. After all, the 75% market share holder in digital game storefronts acquiring the leading independent social platform for gaming seems like a flashing warning sign for antitrust, no?

Why Valve Should Do It Anyway

Addressing Concerns

While all of the issues outlined above are serious, none of them are insurmountable:

1. Not doing acquisitions isn’t so much a “problem” to be solved as it is just a shift in mindset, even if only a temporary one.

2. Integrating a large number of employees without significant adverse effects to culture can be done. And, while it sounds harsh, if Discord does get bought by Valve and loses the incentive to enshittify, a lot of the staff related to those efforts (e.g. ML engineers for ad targeting, sales staff for promos, etc) can be let go.
   That is the flipside of enshittification – it sucks for consumers, but it makes jobs.

3. The integration of the two platforms’ communities can (and probably should) be done gradually, and there can (and, again, probably should) be some separation despite integration.
   The acquisition of Instagram by Facebook is a good example here – while from a technical standpoint, there’s basically no difference between an Instagram user and a Facebook user, the experiences, cultures, and dominant content styles are different across the platforms in very distinct ways.

4. When it comes to operational challenges and regulatory issues, while an acquisition of Discord might speed up the clock on them, I don’t think not moving on Discord will allow Valve to sidestep them entirely. Valve has, thus far, skated by quietly, but all it takes is one sensational story about gambling for knives gone wrong that attracts the attention of lawmakers. It’s happened once before (albeit not with Valve). That’s not to say that the concerns here aren’t real, and they should absolutely be factored into the decision making, but I don’t think they should be treated as blockers, just as problems needing solution. In CS terms, these are expected problems of scaling a system up, and buying Discord just means you hit the scaling problems faster.

A Positive Outlook

Why, despite those challenges, should Valve do it anyway? There’s a two pronged argument:

Business Strategy

A defensive acquisition seems absolutely warranted here. We’ve already talked about the worlds where Epic Games or Microsoft jumps in, but those are hardly the only possibilities; there’s a number of tech giants out there who might see the money printer Valve has as an opportunity and look to cut into it themselves.

Defensive acquisitions are hardly unprecedented in tech – look again to Facebook’s purchases of Instagram/Whatsapp, or Google’s purchase of Motorola simply to prevent the patents from being used against Android, or Amazon’s purchase of Twitch so that someone else couldn’t have it.

Remember why Apple’s iMessage lock-in works? Apple had full control of the hardware interface. When you text another iPhone user, iMessage is just there – frictionless, default, seamless.

Valve can’t replicate that. Steam Deck is one device among many, not the universal PC gaming platform. They can’t just make Steam Chat the default between gamers in the same way Apple made iMessage the default between iPhone users.

Valve hasn’t managed to organically capture social lock-in. If Valve ends up needing it – and Newell clearly recognizes its importance – they can’t build it organically the way Apple did. They’ll have to buy it in the form of Discord.

“Consumer Welfare”

While I hate Bork’s consumer welfare standard, if we view things from that lens, we return to GabeN’s interview referenced earlier on the matter of acquisitions, this time talking about “good” acquisitions:

“Because [in those cases] whoever is actually doing the acquisition has thought through how it’s an additive thing. Not ‘how did I make my empire bigger,’ but: ‘What’s the opportunity to do a better job of building great game experiences for people?’”

Buying and integrating Discord can turn Valve’s offerings into even more of a complete package.

Your games. Your achievements. Your in-game items. Your save files. Your friend list. Your gaming group’s conversational history and in-jokes. Your custom emotes. All living in one place. All travelling with you, from your PC rig, to your living room Steam Machine, to your Steam Deck on the go.

It’ll just work, seamlessly.

\0