Riot Games wrote an entire internal document about circumventing Google Play.

Titled “Self Publishing Games — Circumventing Google Play” and written by a pair of senior Riot Games engineers (including the company’s Chief Architect if my eyes are clear), the document lays out a plan to build a launcher not unlike the Fortnite Launcher by Epic:

In order to support bypassing Google Play on Android phones we would need to build capabilities in the following areas:


Game Website

Game Client

Game Client Patcher

Mobile Payments

The “complexity” of most items is listed as “low” “very low” or “none,” with the exception of mobile payments, which is listed as “Medium (new capability)”.

“All of the capabilities built in the process of preparing self-publishing for LR are reusable and can be then applied (after proper skinning) to other mobile games released by Riot,” the page ends.

Riot’s CFO dismissed it as an “initial educational document from technical to business people” about “what it would take” to make such a thing happen.

Epic is trying to make Riot’s CFO look like a stooge.

Google got him to categorically shut down the idea that Riot was in any way persuaded not to bypass the Google Play Store with its mobile games — but Epic’s lawyer is pointing out that a separate Riot Games employee, Brian Cho, had conversations with Google that didn’t include the CFO.

Epic is also pointing out that Riot’s typical strategy of attempting to distribute a game as broadly as possible and prioritize its own payment system (where it pays “generally between 2 and 5 percent, depending on the partner” on PC) doesn’t gel with the idea it would only ship the Android version of its game on the Google Play Store.

In fact, a 2018 Riot mobile strategy document has a line reading “Steps can be taken to drive players into relatively low fee channels, preserving margin.” Yet Riot doesn’t allow players to sideload its APK on Android, where it could do just that.

I expect Epic will come back to some of these other bullet points in just a moment. Update: Yep, sure enough, Epic just brought them up moments after I pressed publish.

Riot Games says it’s still bringing its games to console — namedropping Xbox and PlayStation.

When will Riot’s games come to consoles, as the company has promised for years?

It’s not yet clear — but in sworn testimony for the Epic v. Google trial recorded just weeks ago on October 27th, 2023 that we heard live in court, Riot Games CFO says “we’re working toward making them available on console platforms such as Xbox and Sony PlayStation.”

“Did anyone ever tell you that the Games Velocity Agreements were offered to Riot in exchange for Riot’s agreement not to develop an app store?”

No, says Riot’s CFO. Google’s attorney Michelle Park Chiu asks the question many other ways, just to be crystal-clear — and Riot’s CFO says there was no implied agreement, no understanding, nothing of any sort that would have kept them from developing an app store.

And if there were, he would have been aware of it as CFO, he testified.

We’re off for lunch now — here’s more Tim Sweeney / Verge trivia while you wait.

I casually ambushed Epic co-founder and CEO Tim Sweeney in the hallway the other day, because yes, he’s still here all day every day in court watching the whole thing unfold.

He told me I got the exclamation points right in this story, which made me LOL. Later, I saw him laughing at one of our Verge bingo cards, asking a colleague if anyone’s said “Metaverse” yet. I don’t think they have?

Oh and — we’ve tentatively got a date for an interview with Tim, though it won’t happen till after the jury verdict. He says he doesn’t want to interrupt (or was it interfere?) with the court process.

Here’s the $90 million deal Google offered Riot Games:

“Google to offer $90M+ total enterprise value to Riot Games in 2020,” begins a document about the Google Games Velocity Program deal it offered to the game developer.

That $90M+ included:

$12M+ in pre-registration support on the store

$20M+ in merchandising launch support

$10M+ in post-launch promotion

$5M for a gift card retail program

$15M in co-marketing dollars (pre-reg and launch)

$10M in a Google Ads reimbursement program, $1 for every $3 spent up to $30 million dollars

$3M for a YouTube community development program

$20M+ in Google Cloud credits with “(2% of game rev on play as credits)”

“Dedicated technical, launch, growth, marketing, and BD Google staff for Riot Games”

“Dedicated quarterly code review, growth, optimization, and UI/UX workshops”

In exchange, Riot had to commit to sim-ship and content parity (with an exception for telcos) — but there was no requirement to stay exclusive to Google Play, we’re seeing. Google’s lawyer is walking Riot’s CFO through contract after contract, addendum after addendum, and none of them mention exclusivity or a Riot store.

So if Google stopped Riot from launching a store, it would have been through a private understanding — and Riot’s CFO says there wasn’t one.

He does say he believes it was Riot’s “perception” that Google believed Riot would sideload its games, bypassing Google’s store, if Google didn’t come through with the promised credits, though.

“Google is sensitive to us going direct-to-players,” Riot Games wrote internally.

“Google is sensitive to us going direct-to-players, and we can use that to ask for increased promotional support while staying on their platform,” reads one line.

Here’s another, though (Riot used underline instead of bold, but I don’t have that option at this moment):

The best way to ensure broad distribution is to be on as many platforms as possible, regardless of their fees

recommendation: use google play

Among the other platforms Riot considered, according to the document: Singtel, and Epic Games.

On a page titled “Maximizing margins” that considered the pros and cons of going with Epic and Singtel lives this phrase:

“Cons: frustrating Google.”

The document shows that Riot was aware it could technically be on more stores in addition to Google Play, but I saw yet another bullet point that suggested it could be detrimental to the Google relationship, too.

Apparently this came up in a meeting where all of Google Play’s leadership was present:

Google leaned in very hard (entire Play leadership attended) as side loading / bypassing their store is still a reality and they cannot afford to lose us as a partner.

Riot’s CFO now says he doesn’t think sideloading was explicitly mentioned in that meeting, though.

Google also wrote internally that it didn’t want Riot to do an app store.

Riot Games CFO Mark Sottosanti is now speaking in Epic v. Google.

We’re done with Activision Blizzard’s CFO and Google is presenting another taped deposition, this one from quite recently: October 27th, 2023.

He says Riot began discussions to launch games on Google Play before it had ported any games to mobile. It wound up signing an $18M deal with Google, which, Epic has tried to suggest, blocked Riot from launching its own mobile app distribution platform.

He says his understanding is Riot pays Apple the standard 30 percent service fee.

Activision Blizzard played two big companies against each other.

While its Project Boston apparently ended with it getting “billions of dollars flowing” between ABK and Google rather than attempting to ship its own mobile app store, the app store wasn’t its only Plan B.

Lawyers just showed that ABK also considered a different deal with a redacted company. The business presentation included this note:

“Note: these deals are mutually exclusive and ABK will only enter into the one we determine as the best option.” Zerza confirms ABK choose Google over the other company, and (Google’s? Epic’s?) lawyer is driving home that ABK did so because the terms of the deal were simply better for ABK.

One of those terms that I managed to nab before it flashed off-screen:

$20/15/15M annual co-marketing contingent on $900/1,000/1,200M gross bookings of ABK titles on Google Play (2020 AOP: 985M+)

We briefly saw the entire term sheet, but it was shown too small for us to read.

As of 2022, Activision Blizzard was still “looking” at launching its own mobile store.

While Activision did indeed sign its desired partnership with Google in January 2020 — we just saw ABK CFO Zerza’s signature alongside that of Google’s Hiroshi Lockheimer — Zerza claims the company is “still looking at” launching its own app store as of today (read: 2022, when his deposition was taped).

However, there’s apparently a document that implies the January 2020 deal did derail the Project Boston plan to do so, at least: “It led us to make the decision to pivot the plans for Boston,” read a lawyer, off a document I didn’t manage to see myself.

Here is another bullet about its proposed app store I missed getting into earlier posts:

End state goal of single mobile distribution solution experience with the ability to eventually support all ABK titles (and potentially 3rd party) on Android devices first (Apple iOS to follow)

ABK had an “estimated steady state goal of 10-12%” for its store, rather than the 30 percent fee it pays Google, according to the Boston presentation. Zerza said he couldn’t confirm the number.

In 2020, Activision Blizzard plan was to ramp its mobile app store.

Forgive the formatting, I copied things down fast before they disappeared. Here’s what was planned for 2020’s “ramp” of the mobile app which never wound up launching to begin with.

ABK Integration / Ramp

joint abk approach to add capabilities throughout the year

multi account support

push notifications

store operations team and tools

market expansion

cut over to internally developed systems

us and other major markets

King/ built platform: ramp team with core functionality capabilities in place to build & maintain

technology – app development, account systems, web development, commerce systems, game integration, distributions ystems

business operations – business performance, editorial, finance/reporting, legal 

In 2021, the plan was to reach “ABK solution at scale,” with 45-70 employees working on the project.

ABK’s CFO says that sounds very low. “This is not consistent with what I think it would take to develop a store like this,” he says, saying other internal efforts have hundreds of people.

Activision Blizzard’s own mobile app store would have started with King.

In the secret Project Boston document, we’re now seeing Activision Blizzard’s proposed timeline for building the thing — starting with a minimum viable product (MVP) in 2019.

The first version would have exclusively had King games like the mega-popular Candy Crush, but could stretch to Activision and Blizzard games later.


King Led proof of concept

1 product for King games

3rd party ‘off the shelf’ systems

King games for pilot

US based only


Use for carrier & distribution negotiations

Put pressure on Google (ongoing negotiations)

Validate technology & business assumptions

Integrate key capabilities (PSP)

Proof of concept for product/game integration solution

Here’s the part where ABK suggests it could stretch to Blizzard games:

ABK optionality (future cross ABK functionality)

~5 resources needed (for ideal integration would include 1-3 from

The MVP would have had 15 people working on it within the company, according to the slide we’re seeing.

Overall, the app would be:

An Android mobile app

enables purchasing, download, and patching games outside of the Play Store (side loading)

pre-installed and carrier scertified on certain Carrier devices

downloadable via website

ABK Store App would not be available on the Play Store

Activision Blizzard’s “Project Boston” overview.

“We see two potential paths… both are being pursued in parallel,” Activision Blizzard wrote in the Project Boston plan.

I’m copying down the details about those paths best I can:

1) Enterprise negotiations with Google first (and then Apple following)

2) Development of a direct to consumer mobile distribution platform that bypass existing storefronts […] with two approaches to the store

2A) ABK storefront only

2B) ABK storefront and external

“At the time, we were talking to Google about an enterprise-wide deal which we obviously signed with Google in early 2020,” says the company’s CFO.

“At the same time we were also looking at our own mobile store,” he says. ABK wants you to think it genuinely pursued these ideas in parallel, but he hasn’t yet told us when ABK stopped pursuing the store.

Correction: I typo’d: the Google deal was early 2020, not early 2022.

Introducing Activision Blizzard’s “Project Boston” — it wasn’t just an app store.

We’re seeing some internal ABK documents from December 2019 — and it looks like the app store was ABK was hedging its bets!

“Project Boston” shows two parallel “paths” for ABK through the mobile app landscape, but only one of which would be chosen in the end.

ABK wanted either a mobile app store, or alternatively a $100M+ deal with Google. We know which one it got.

“Path 1: Enterprise negotiation” read: “capture stronger economics for ABK across mobile, YouTube, advertising, media spend, and cloud” and “$100M+ per year value creation for ABK and growth in overall book of business between ABK and Google”

“Path 2: Build own mobile store” includes this incredibly important bit which ABK bolded for emphasis: “Should we secure real savings with Google, we would deprioritize path 2.”

“Our ultimate goal here would be to have a joint mobile app store,” said Activision Blizzard CFO.

“Our ultimate goal here would be to have a joint mobile app store — we strongly believe that scale at the front end will benefit all of us and our players, plus increase the odds of success,” Zerza told Tim Sweeney in May 2019.

“We have already announced CoD and Diablo mobile, more to come,” he teased.

In a 2022 video deposition, he told the court there were many reasons to explore the idea:

one of them was to create a more direct relationship with our players, for example, and when we have more direct relationships, then we generally are better able to see what our players are actually engaging with faster. We do have those date form our partners to, but there’s always a time lag.

It would have had an interoperable account system, and the idea was to build the storefront first, then add marketing and promotions later on, according to the emails.

But in the 2022 deposition we’re hearing now, Zerza claims they were “very early exploratory discussions.”

“Obviously we never pursued it because it wasn’t financially attractive for us,” he says.

Lawyers questioning him are asking questions designed to see how serious ABK really was, and whether it lied to partners about its plans. Whether Google and Epic believed Activision Blizzard or not is relevant to this case.

“Steam of mobile”: Activision Blizzard, Epic, and Supercell scrapped a plan to launch their own app store.

In 2019, Epic Games CEO Tim Sweeney and Armin Zerza discussed a plan to “create a consortium between Epic, Supercell, and ABK.”

The goal: “one single mobile games store focused on distribution of mobile games (“STEAM of Mobile” concept), including a single payment system.”

We’re seeing their emails in court right now:

Activision Blizzard’s CFO confirms ABK was paying 30 percent — and asked Epic’s CEO if he wanted to change that.

ABK CFO: “I asked Mr. Sweeney whether Epic, in this case, would be open or interested in, you know, joining a conversation on developing our own mobile distribution payment systems.”

Now, we’re hearing they wanted to build a mobile game store together, too.

Epic and Google’s economists are done and we’re moving on.

Google got a brief chance to ask their economist Tucker a few last questions, but Epic successfully objected to most of them. She did get to say that she doesn’t believe the gap between the time Apple introduced new pricing and Google followed Apple’s lead was relevant, suggesting these things take time in the real world.

We’ve now moved on to a September 22nd, 2022 video deposition with Activision Blizzard CFO Armin Zerza.

“I see partnerships, not market power.”

That’s Google’s economist Catherine Tucker, making an interesting suggestion that fits her proposed market definition: she suggests that OEMs are not customers for Android, and thus Google isn’t a seller in a market, because she sees them as partnerships instead.

Epic moved on to attacking some of Tucker’s earlier points about competing with Apple, suggesting that whether or not Apple has a better product that indeed competes with Google’s products doesn’t tell us anything about whether Apple “constrains the Play Store as a competitor.”

“Isn’t everything on the internet designed to facilitate the delivery of digital content?”

That’s Judge Donato, who interrupted yet again to ask if Tucker’s market definition is… broad. “What’s not in your vision?” he asks.

“Things which are cumbersome, burdensome, not easy ways of delivering digital content,” she replies.

“Can you name a couple for us?“

She suggests that physically going to GameStop to buy a Nintendo Switch cartridge, which technically contains digital content, might be one.

“Is there anything online that is not in the relevant product market as you define it? Anything at all?”

A long pause. “So….”

“I think because I’m including the ability to self distribute…”

She lands on Disney+. 

“There’s many ways I can get that content to you… that would include subscribing on my PC through a website […] a smart TV would be another way […] I’d include all those ways because we’re allowing for the possibility of self distribution,” she says.

The judge leaves it at that for now. We’ve just moved on from market definition to other parts of Tucker’s report, but not before Epic tried its best to make Tucker look like she’s in fantasy land, with questions like “In your view there are not even transactions between the smartphone OEMs and Google, correct?”

Why I think Google is keen on Tucker’s market definition:

I went back to check last night, and sure enough: back when Epic lost its first fight against Apple in that separate and earlier case, Judge Yvonne Gonzalez Rogers handed out that loss/victory based on her own chosen market definition of “digital mobile gaming transactions.”

In a vacuum, that sounds pretty similar to Tucker’s definition of “facilitation of digital content transactions” — except for Google, it’s even better! Because it’s not limited to mobile, Tucker can point out that Android also competes with other gaming platforms like PlayStation, Xbox, and Nintendo, all of which have rigid 30 percent fees.

Judge Donato: “are browsers included?”

The judge interrupts Epic’s cross-examination to drill down on Tucker’s market definition yet again, asking if the “the facilitation of digital content interactions” would include web browsers.

She initially says no, it’s about “an instance where a platform can add value.” But he probes further, asking if they would be such a platform in the online world. She waffles slightly, saying they need to have the ability to self-distribute.

The judge asks one more time: “Would you include browsers in the relevant product market?”

She says yes.

It’s Epic’s turn to question Google’s expert economist Catherine Tucker.

Google finished up its first round with her yesterday, and now Epic gets to cross-examine. Epic lead attorney Gary Bornstein is beginning with likely the key to this entire case — the relevant product market, which determines whether or not Google has monopoly power to begin with.

Tucker chose a market definition that gave the judge a double-take: “the facilitation of digital content transactions.”

Now, Epic will try to poke holes in it.

So far, Bornstein seems to be implying that Tucker’s definition is not a product — he suggests neither Google nor Apple ever refer to their products this way — but he’s early on in his line of questioning so I guess we’ll see!

Epic v. Google day 13 begins — the judge demands demands.

On Tuesday, right before end of day, Epic revealed that it’s never actually formally discussed settling with Google, and Judge Donato ordered a settlement conference before we go to jury verdict.

This morning, the judge got clearer: Epic must file demands by 9AM Saturday. “The demand is going to be all-inclusive, detailed, and specific,” he says.

Google gets 24 hours to respond: “Google will serve an all-inclusive, detailed response… by 9AM on Monday morning,” he adds. The settlement conference will be just four people: Epic Games CEO Tim Sweeney, Google’s “right person to sign an enforceable deal,” and lead attorneys for both parties.

Epic v. Google Day 12 is done — but not before some revelations.

Judge Donato is ordering Epic and Google to have settlement discussions — which Epic lead attorney Gary Bornstein has just revealed have not yet happened at all. Repeat: Epic and Google have never formally attempted to settle this case before now.

“If you win, what are you planning to ask for?” asks the judge — because Epic has not articulated that clearly before now either. (The Verge has also asked Epic and been partly rebuffed.)

Epic says it’s asking for three things: freedom for Epic and other developers to introduce their own stores without restriction, total freedom to use its own billing system, and an anti-circumvention provision “just to be sure Google can’t reintroduce the same problems through some alternative creative solution.”

Judge Donato says the last won’t happen: “We don’t do don’t break the law injunctions… if you have a problem, you can come back.”

He says the first two asks feel like they’re doable in a settlement: “Spotify pays 4 percent or 0 percent and has its own billing… you need to be clear with your client who’s sitting behind you that [settlement negotations are] going to happen.”

(Epic CEO Tim Sweeney is sitting behind him. I don’t see a visible change of expression on Sweeney’s face.)

Google will call Rich Miner, a co-founder of Android, as soon as tomorrow as it begins its case for counterclaims.

The plan is to finish up evidence on Friday the 1st, and begin closing arguments on Monday the 11th.

google’s-nest-cameras-can-now-tell-you-when-your-garage-door-is-left-open Previous post Google’s Nest Cameras Can Now Tell You When Your Garage Door Is Left Open
tesla-cybertruck-starts-at-$60,990-and-can-get-up-to-340-miles-of-range Next post Tesla Cybertruck Starts At $60,990 And Can Get Up To 340 Miles Of Range