They’re usually prompted by questions Future Media’s readers, viewers, and listeners.

Today, one wrote in: “Your narrative style isn’t good for my dyslexic brain, hence why I don’t read much of it - but from what I could read, that piece on the iPhone is really good and insightful.”

So, this one’s for you Liam. Welcome to Future Media Explains…

Your next iPhone is going to be more expensive.

Not a little more expensive.

A lot more expensive.

Research firm TechInsights estimates Apple will add US$270 to the price of the next iPhone Pro.

That’s in September.

And Apple’s CEO Tim Cook has already told The Wall Street Journal it’s coming.

He blames soaring chip prices. And AI driving up the cost of memory.

He called it a “hundred-year flood.”

But that’s not the whole story.

Because the real reason your iPhone is going up in price is a 25-year gamble.

A gamble that Tim Cook made - and lost.

And now you’re being asked to pay for it.

This is Future Media Explains. I’m Ricky Sutton.

Share

Let’s go back to the beginning.

Twenty-five years ago, Tim Cook joined Apple.

His job wasn’t to design products. His job was to build the supply chain.

And he made a decision that would define Apple for a generation.

He moved everything to China.

Manufacturing. Assembly. The whole operation.

And it worked brilliantly.

China was cheap. China was fast. China was enormous.

And the iPhone became the most successful consumer product in human history.

At its peak, Apple was 7% of the entire S&P 500 stock market.

That means when the iPhone sold well, the retirement savings of ordinary Americans went up.

That’s how big this thing got.

Share Future Media

But there was a problem buried inside all of that success.

Almost all the advanced chips inside Apple devices - every Mac, every iPhone, every iPad - they come from Taiwan.

Taiwan is a small island off the coast of China.

Twenty-three million people.

And the most critical semiconductor factory on the planet.

Here’s the geopolitical reality.

China has never accepted Taiwan’s independence.

Beijing has never ruled out going to war to take it back.

And a 180-kilometre stretch of water is all that separates them.

Tim Cook knew all of this.

He built Apple’s entire future on that fault line anyway.

And he didn’t have a Plan B.

Share

Now here’s where it gets really interesting.

To keep selling in China, the world’s biggest market, Apple did a deal.

It invested over US$1 trillion in China.

It trained 30 million Chinese workers in advanced mobile technology.

In return, it got access to Chinese consumers.

But China used what it learned.

It took that technology. That knowledge. That expertise.

And it built its own smartphone industry.

The Council on Foreign Relations estimates China now banks $360 billion a year from its Apple relationship.

China makes almost as much from Apple as Apple does.

And the homegrown rivals Apple helped create - companies like Huawei - are now beating it in China.

Apple’s China sales fell 11% earlier this year.

Tim Cook was trapped.

Two superpowers - America and China - pulling at his company from opposite sides.

And he was stuck in the middle with no way out.

Share

When it became clear the situation was unsustainable, Apple did something extraordinary.

It started chartering jets.

Planes packed with 600 tonnes of Apple components.

Flying them from China and Taiwan to India and Vietnam.

Just to get them out of China fast enough.

At the peak, that operation was costing Apple US$10 million a day.

Cook told analysts the shift combined with US tariffs on China had cost $900 million in just 91 days.

He also pledged to invest $500 billion in American factories over four years.

And within months, half of all US iPhones were shipping from India.

He expects the majority to come from India by mid-year.

It’s an incredible logistical achievement.

But it came 25 years too late.

And it came at enormous cost.

Share

So let’s add it all up.

Memory chip prices are up four times what they were.

The cost of moving production out of China is in the billions.

Competition is rising - in China, in the US, everywhere.

Apple’s AI strategy? It arrived late. And it’s still catching up.

Google is also paying Apple US$26.3 billion a year to be the default search engine on the iPhone.

Think about that. Apple’s biggest rival in AI is also its biggest revenue source.

And Apple’s AI product - Apple Intelligence - has been quietly delayed and underdelivered.

Meanwhile, the iPhone itself hasn’t had a genuinely new idea in years.

Android phones are overtaking it on features.

The next big thing - a foldable iPhone - isn’t coming until September.

So when Tim Cook says chip prices are why your iPhone is getting more expensive?

He’s not lying.

But he’s not telling you the whole truth either.

The chip prices are the spark.

The dumpster fire underneath them is 25 years of decisions.

Share

Here’s one more thing worth knowing.

Tim Cook is leaving.

He’s stepping down later this year with a pay packet worth US$74 million in his last year alone - and a net worth of US$2.9 billion.

He’ll hand the keys to John Ternus.

Ternus is a hardware guy. An Apple lifer.

He’s inheriting a company that is simultaneously the most valuable on the planet and under more pressure than at any point in its history.

Revenue is flattening.

Hardware costs are rising.

AI strategy is behind.

And the geopolitical faultline Cook built on is still very much alive.

Share

So what does this mean for you?

It means that US$270 price increase isn’t really about chips.

It’s the invoice for a 25-year strategy that ran out of road.

You’re not paying for a better iPhone.

You’re paying for decisions that were made before most of you had one.

The iPhone has always been expensive because it was the best.

Now it’s expensive because Apple has a very big bill to pay.

And you’re the one holding it.

I’m Ricky Sutton. This has been Future Media Explains.

Subscribe for more - and if you have a question you’d like me to answer, send it in.

See you next time.

It arrived quietly and revealed it was allowing users to personalise their favourite news sources in Google.

Once done, publishers the readers had selected as their favourites would be prioritised in Google search results.

Wow! That’s a big deal.

Today’s mail is the tip of a giant story which first broke the surface in Johannesburg just before Christmas and reveals the direction of Google’s next ad monopoly.

It’s a big story, so this one’s on the house.

And my thanks to today’s sponsor Cerby. 🙏

⚽ World Cup 2026 is about to kick-off and thousands of publishers will be relying on social media. But are your brand accounts secure? Protect them with Cerby.

Protect your social accounts with Cerby

Let’s go…

If you’re a regular at Future Media you know that I live by a simple premise: The first rule of tracking Big Tech is to analyse anything it says in detail.

Google’s mail today says it’s changing its $294 billion search ad cash cow to give us all “more control” over search.

We can even tell it which news providers we want at the top of our search results.

But if you look closely, you’ll see Google prioritised search, maps, shopping, hotels, flights and even translations ahead of news.

And that betrays the real reason this is happening.

Alphabet’s banked $402.8 billion last year and 73 cents of every dollar came from the advertising that used to fund publishing and the open web.

And search ads are Google’s biggest earner.

So if Google’s reordering how Search works, it is performing open-heart surgery on the most profitable advertising business ever built.

So there has to be a very big reason.

Today’s Future Media Explains video shares what it is. And the trail leads back to a competition inquiry in Johannesburg in late 2023.

Subscribe now

The hidden hand

The Media and Digital Platforms Market Inquiry launched by South Africa’s Competition Commission was comprehensive.

It stretched over two years and included five rounds of information requests and public hearings where Google was forced to sit in a room and answer tough questions.

The architects of that inquiry joined me on the Future Media Podcast last December.

What made their inquiry different from every other attempt worldwide is that it wasn’t about money. It was about protecting the country’s Constitution.

South Africa has 12 official languages. They’re the architecture of the country’s identity and culture.

But when the inquiry looked under the hood, Google Search supported only three of them. Its constitutionally protected languages simply didn’t exist in search.

And that was a problem.

The regulators used that legal leverage to demand Google’s raw data including its impressions and click-through using the legal information powers it unlocked.

What they found was stark.

Almost half of all news queries from South Africans were redirected by Google to foreign outlets like CNN, Reuters and Bloomberg.

Google argued that was what its users wanted, but the data showed that was a convenient deception.

The data revealed that Google’s South African users were 60 per cent less likely to click foreign links than links to their own publishers.

Google was shown to be systematically serving people the news they didn’t want and burying the news they did.

The Commission did something genuinely clever.

Using Google’s own spreadsheets, it showed it was failing its own customers and covering it up with words.

Google was pressed into a remedy. It would roll out a controversial new solution it had been quietly testing in its labs. It was called Preferred Sources.

The mail in your inbox today is Google now releasing it to the rest of the world.

Subscribe now

Rapid reaction

Google’s reputation has been to be very slow and considered in launching major changes - and with good reason given the money involved.

But after criticism for being too ponderous with AI, it has been throwing caution to the wind.

AI Overviews was the fastest expansion in Google’s almost 30-year history and the preferences launch is not far behind.

Look at the timeline.

• Mid-2024: Google begins trialing it in Search Labs in US and India, and only in English.

• November 2025: South Africa’s final report lands in Google’s inbox demanding Preferred Sources

• December 2025: Google commits to releasing it in South Africa early.

• April 2026: It was live in every supported language and market.

• Two weeks ago: Google extended it into AI Overviews and AI Mode.

This morning’s email is the consumer-facing wrapper around all of that.

Google looked at South Africa’s numbers and learned something. People like personalised results.

More than 345,000 unique sources have already been selected by its users globally.

Share

Personalisation tax

But why is Google suddenly so keen to evangelise the feature?

There are two reasons.

First, it’s a terrific regulatory insurance.

It is in the most intense enforcement period it’s ever faced with antitrust convictions in the US, ad tech remedies pending and inquiries from Brussels to Canberra.

A personalisation dial showing its listening to consumers is the cheapest concession available, and photographs really well in a courtroom.

But there’s something else.

A happy user is an engaged user, and an engaged user is a better advertising target.

When users hand Google their source preferences, travel intentions and shopping tastes, they’re not just curating their experience. They’re completing Google’s profile.

News and niche content drive daily habits. Habits reveal profiles. Profiles drive targeted ad auctions.

And that’s what feeds Google’s monetised funnels of Search, Shopping, Hotels and Flights.

It’s dressing up the next era of targeting, as control. It’s quite clever really.

Share

The bigger picture

For 20 years, personalisation has meant a ranking algorithm that makes a best guess at the users’ preferences based on click data.

It’s been crude, passive and one-size-fits-most - but in the sunsetting era of the open web, Google made the best of it to create a $4.3 trillion mega-monopoly.

But AI is changing the game.

It doesn’t rank an experience, it manufactures one.

Every digital surface can now be assembled fresh, per person, in the moment.

AI Overviews has turned search results pages into publication with a circulation of one. You. 🫵

The South African inquiry found that only 12 per cent of users who saw AI Overviews clicked through to publishers.

It also found that South African news content was only 0.26 per cent of the data Google’s AIs were trained on.

My reporting has so far focused on how publishers are being bypassed by AI Overviews. South Africa’s data proved entire nations were being silenced.

Twelve languages and 60 million people just didn’t factor, until South Africa forced the issue.

What publishers should do

Any day now, the first Future Media Intelligence report will be released to 26,000 publishers by FIPP, part of the world’s largest publisher industry group WAN IFRA.

It contains insights from more than half a billion data points as well as hundreds of interviews over three years to chart a path forward in the AI era.

Preferred Sources uses AI to change the ranking signal of search from the backlink and algorithmic bias of PageRank to one that runs on reader loyalty.

That opens interesting doors for publishers who believe they can push back the tide of AI slop.

But really this is a story about the revival of premium journalism and use it to reduce their 20-year over-reliance on search and social to send clicks.

1. Show readers how to add you as a Preferred Source. Google has published the how-to. People who care enough to choose you in settings will be the most loyal and valuable - and a distribution signal too.

2. Change what you measure. Stop tracking page views and start tracking direct-relationship traffic. Readers who type your URL, click a bookmark, or open the newsletter are your real audience. Anonymous traffic never paid the bills.

3. Double down on the quality of your product. Referrals from Google Search and Meta are down 40-80 per cent since AI Overviews. It’s time to let the readers do the distribution for you.

As Paula Fray, who helped build the South African deal, told me in December: These remedies are a foundation, not a finish line. Let’s build on it.

My post was explaining why we haven’t had a decision on breaking up Google’s illegal ad tech monopoly.

Basically, the judge Leonie Brinkema has been super busy.

The Google break-up - and with it the fate of the open web - is just one of the many massive cases on her to-do list.

Judge Leonie Brinkema has been busy.

First, she ordered the release of a Venezuelan couple with temporary protected status whom the Trump administration had detained.

And days ago, she blocked Trump’s US$1.776 billion Anti-Weaponisation Fund - a slush fund critics said was rigged to reward political allies, potentially including January 6 rioters.

My post was to share the good news that now she’s free, and unravelling with Google’s illegal antitrust monopoly - and returning the ads that she called the “lifeblood" of the open web - is her top priority.

But back to the question the reader asked, because I had published this chart in the post.

It compared Google own revenue growth (in green) against the falling ad revenue flowing through to the open web, through what Google calls Network ads (in red).

The reader called Myles asked me: “Hey mate, can you explain this diagram to me? Google search I get, but Network?”

It felt like time for another episode of Future Media Explains, so here goes.

Chunky revenue

Google has a monopoly in search, and an illegal monopoly in ad tech.

We know this from the antitrust trials brought by the US Department of Justice over the past few years.

Google is required by law to report its revenue in quarterly updates to America’s Securities and Exchange Commission, the SEC.

And it reports its earnings - which passed a record $402 billion last year - in chunks.

This is a screen from Google’s latest quarterly update showing its revenue (in billions) from Search, YouTube and the Google Network.

And it’s that Network line that I was posting about, and what led to the question.

Network is where Google reports all the other money it makes, and within this number, is how much advertising Google’s ad network shares with the rest of the open web.

Let me explain.

Google has an all encompassing planetary-scale monopoly on digital advertising through its ad tech. It controls how ads are sold, delivered and measured.

It has the world’s most dominant ad server called Google Ad Manager, known as GAM.

Almost all the world’s publishers and almost all ad-funded websites on the web use it to find and deliver ads on their websites.

Google then has an ad network called AdX. Think of it as a dam full of ads looking for a place to be served.

Almost every advertiser and advertising agency in the world relies on it to get their ads from their whiteboards to the people they are looking to sell products to.

And it’s massive.

Around half of all the world’s ad demand - that’s advertisers looking for audiences - is transacted through here.

And we know from evidence surfaced at the trial that its scale is almost impossible to the human mind to comprehend.

How it works

Let’s say Hyundai wants to sell its SUV to 25 to 35-year-old women who have recently had children and need trunk space who live on the East Coast of the United States.

Hyundai goes to Google’s ad tech, and drops its requirements.

Google’s AdX runs a super-fast auction in a fraction of a second to find websites that have the target customers, and GAM delivers the ad.

Google gets paid at least 35 cents on every dollar for doing that, and the website gets what’s left.

This is programmatic advertising at its most basic, and when it’s working right, then everyone wins.

🖥️ Google gets paid for the tech.

📝 The publisher gets paid for creating the content that attracted the reader.

🎯 Hyundai’s ad reaches the right potential buyer.

🚗 And the consumer gets the car they want.

Only, my chart - the one Myles was asking about - shows how this has been corrupted by the illegal ad tech monopoly that Judge Brinkema is now considering breaking up.

Now let’s explain what it shows.

We have already established that Google controls all the ads looking for customers through AdX, and it controls where they go through the taps and pipes of GAM.

But in 2026, Google wants all the ads for itself, because that’s how it sustains the double digit annual growth Wall Street demands.

Internal emails presented at the trials exposed Google executives “shaking the cushions” to increase prices to meet Wall Street goals.

But also because it needs cash to fund the colossal investments its making to stay ahead in the AI race.

That has led it to bend the rules, and this chart is the proof ☝️.

To keep as many of the ads for itself, Google launched AI Overviews to replace search.

You might remember that it was Future Media that revealed this back in 2023 when it was a secret Google patent.

Since then, Google has used AI Overviews to keep readers on its search pages by answering the questions.

Think what that does.

A person goes to search and asks what are the latest headlines in the Wall Street Journal today.

AI Overviews responds with the answers, and the reader does not need to go to the Wall Street Journal.

That means Google keeps the reader, and serves them ads, and does not need to send the reader to the WSJ, so it does not need to share its ad revenue with them.

This isn’t theoretical any more. It’s real. Here it is in action.

Self preferencing

When Google sends readers to a publisher, it’s known in the industry as referral traffic.

Since AI Overviews, referral traffic to publishers has fallen by as much as 80 per cent in the worst cases.

But Google’s not done, because it has also turned the taps and narrowed the pipes to send less of its AdX dam of ads with everyone else on the web too.

And the proof and the impact of both together is what the Network Ads chart I shared shows.

It shows Google keeping readers to itself and monetising them for growth, while simultaneously sharing less of the advertising it controls with the open web.

And the SEC filings show how it’s moving the ads that used to pay for every ad-funded website to platforms it owns, mainly search and YouTube.

The law has a term for this. It’s called self preferencing, and in Google’s court cases this was found to be anticompetitive and illegal.

For the layman though it’s a pincer movement. Google’s sending less readers and less ad revenue to everyone else and suffocating the open web, while using its monopoly to hoard it for itself.

And that is the problem that Judge Brinkema will soon offer her solution for.

Will she force Google to sell GAM and AdX. That’s what the US Department of Justice wants.

Or will she try to change Google’s behaviour so it plays nice and shares like it used to. The DoJ has bleakly called this “putting a BandAid on a severed limb”.

That’s what we are all waiting to hear. It will decide the fate of the $750 billion global digital advertising market

So it’s a great question, and thanks to Myles for sending it. Please keep them coming. There’s plenty to talk about. Stick any questions you have in the comments. 👇

This has been Future Media Explains.

PS: I asked Google’s AI Gemini to fact check this article. It said: “The post is highly accurate regarding the specific news events it references (the Venezuelan couple and the Anti-Weaponisation Fund) and correctly explains the basic mechanics of programmatic ad tech.”

Spotify… man, it was sounding sweet a year ago. It executed a US$1.8 billion turnaround with a glorious 14-point plan and everything was humming.

Today though, the music paused, so I’m breaking down why it’s paying artists less because of an ad flop that hit rewind on US$14 billion of its value.

And don’t miss a big earning day tomorrow when I’ll deep dive into Google and Meta’s latest results.

First, thank you to paying subs from independent DSP The Trade Desk, New Zealand’s biggest media company NZME and recommendation engine Taboola.

And welcome to new free subs from influential tech tracker Arete Research in London, the US lawyers running a US$218 billion action against Google Keller Postman, AI marketers Monks, public broadcaster 3Cat in Catalonia, Spain, and Opera Australia, among many others.

Let’s fire up the ghetto blaster, hit play, and learn what Spotify’s troubles tell us about a new phase in the future of media…

The Big Story: From a boom to a bum note

Read more

Happy birthday Instagram. Fourteen years ago today Mark Zuckerberg shook hands with Kevin Systrom and agreed to buy the fledgling app for $1 billion.

As the ink dried on the paperwork, Instagram’s internal analytics showed it had soared to 30 million users, with 13 employees, no revenue and no ads.

Today Insta boasts three billion monthly active users, Reels has a $50 billion run rate, and its parent Meta banks $200 billion in ads - roughly half of it from Insta.

Zuck’s $1 billion bet is now conservatively worth US$500 billion. It may be the most lopsided acquisition in corporate history.

But the deal has been dodgy since the start. Most people have forgotten that. This post is a reminder.

And explains why the Federal Trade Commission believes Zuck withheld critical documents and wants the deal unwound.

Many of the key documents remain under seal, but today I’m publishing internal mails that show how the deal was done - and why it may not survive another birthday.

The Big Story: The hunter and the hunted

Read more

A federal judge ruled that Mark Zuckerberg’s controversial US$20 billion acquisitions of Instagram and WhatsApp were not anticompetitive.

Judge James Boasberg agreed Meta may have held a monopoly but rising competition from TikTok had changed the board and the FTC had “struggled” to make its case.

It’s a big loss after a five-year investigation by the Federal Trade Commission, which has strong echoes of Google’s search antitrust decision.

There, a different judge ruled Google held an illegal monopoly over search, but that he would not take any meaningful action as Google now faced competition from AI.

In both cases, the judges have chosen not to rule on past misbehaviour, but to bet on the future and help America maintain its momentum in building AI.

Meta’s value will now likely soar. It also opens the way for a raft of Big Tech acquisitions which will likely cement its supremacy for the time being.

Meta and Google’s cases were similar but different.

Google built its search tech and used its scale to dominate the browser market with Chrome, and mobile with Android. It then used its billions to buy dominance in ad tech.

Meta was different. It grew Facebook but when that began to lose its lustre, it bought its future acquiring rising stars Instagram and WhatsApp. And this is why.

The FTC argued that Zuck bought them to remove them from the board and to ensure Meta had an uncontested monopoly in social and social advertising.

Insta brought cool cachet. great mobile tools, and audience segments that were catnip to advertisers.

WhatsApp brought rocketship growth and engagement that smothered Meta’s marketing messaging in honey.

Meta used both to build an unassailable lead in social. Its ad revenue boomed, and it was the fastest rising stock on the US S&P index soaring 82 per cent. 📈

Rivals were flattened by its heft.

Snap spent billions but remained anchored to a 2.1 per cent share. The FTC argued that proved Meta was the 20-ton gorilla monopolising the market.

The question though was what market? The FTC argued the market was for personal social networking, which included Snap and minnow MeWe.

Viewed through that lens, Meta was an unquestioned monopolist, and the case was a slam dunk for the 6’ 7” basketball playing judge Boasberg.

Only the ground shifted beneath the FTC since the case was brought. Meta’s lawyers argued social was in a generational pivot from friends and family to video.

That put Meta in the same market as YouTube, Reddit, and X - but it was TikTok that tipped the scale. It was big, and more threatening, growing almost twice as fast.

Uh oh. US supremacy at risk again, and this time from China. That’s bad news.

But during the trial, The White House and Beijing agreed to sell TikTok’s US operations to a consortium of American owners.

The deal is yet to be signed but the new owners include Trump supporter Oracle’s Larry Ellison and potential Fox Corp, owned by the Murdochs.

News Corp outlets have actively campaigned for regulation of tech giants, including TikTok, over issues of content compensation and market power.

Meta’s legal eagles argued that TikTok was taking market share forcing Meta to invest billions to defend itself by adapting.

It pivoted from friends and family posts to video, spending $4 billion building Reels to combat TikTok and rival YouTube Shorts.

Zuck - the risk-taking honey badger of tech, did it again and was rewarded with a cash windfall.

Meta’s 2024 financials showed a 22 per cent growth to US$164.5 billion, with ad revenue making up 97.2 per cent of that.

As users grew five per cent to 3.35 billion, Zuck blasted them with rising ad loads (+11 per cent) and ground advertisers to pay more (ad prices up 10 per cent).

And the impact was global.

• US/Canada: US$63.2 billion (38 per cent)

• Asia/Pac: US$45 billion (27 per cent)

• UK/Europe: US$38.4 billion (23 per cent) and

• ROW US$17.9 billion (11 per cent).

I’d also be remiss not to point out that former senior exec-turned-whistleblower Sarah Wynn Williams testified to the US Senate that Meta had a secret business in China worth US$18.3 billion. That’s still under investigation.

I reported on this in my investigation for paid subs: Zuck’s hiding a secret reliance that’ll 心烦 the White House

Boasberg said TikTok, Meta and others have “evolved to have nearly identical” features. “The evidence resoundingly shows that.”

“The FTC’s hypothesis about how people use these apps is disproven by the data.” TikTok and YouTube “compete fiercely over a meaningful share of Meta’s business.”

“The court finds that YouTube and TikTok prevent Meta from holding a monopoly. Even if YouTube is out, including TikTok alone defeats the FTC’s case,” he wrote.

Meta’s is understandably thrilled with the court decision overnight. It fought hard for the case to never get to court. Zuck begged The White House to let him off the hook.

Last month, he sat next to President Donald Trump at a high-profile tech dinner and boasted he would spend $600 billion in the US through 2028.

Spokesman Christopher Sgro told Bloomberg: “The decision today recognises Meta that faces fierce competition.

“Our products are beneficial for people and businesses and exemplify American innovation and economic growth. We look forward to continuing to partner with the administration and to invest in America.”

Yeh haw - go ‘Merica.

Only, there’s still a barb in the victory.

In its latest quarterly results, Meta’s CFO did something she’s never done before. She warned investors there were major legal risks ahead.

Susan Li said the company “continues to monitor the increasing legal and regulatory headwinds in the EU and the US that could significantly impact our financial results”.

She’s talking about this 👇 A howitzer of a case brought by 41 US States alleging they can provide Meta causes child harm, knew about it, and covered it up.

The 233-page indictment says:

• Meta engaged in a “scheme to exploit young users for profit”

• Misled the world about safety features and harmful content.

• Harvested children’s data in violation of federal law, and

• Knowingly deployed changes that addicted children.

Finally, the FTC could appeal, but it feels unlikely. It lost the case on failing to prove the market. That’s not likely to be overturned.

So, Meta is free to push on, with its plans for synthetic social, and eyeglasses, and the Metaverse.

And if you still haven’t got enough, you can get full coverage and analysis of all Meta’s antitrust and other actions around the world here.

Meta on trial: Full coverage

But it might just hold the key to solving one of the biggest problems we have online right now: How creators get paid by AI.

Share

Think about this for a second.

We’ve got these incredible AIs that are basically vacuuming up the entire sum of human knowledge from the web.

But that’s kicked up a massive, massive question.

When an AI learns from your Substack, or your exclusive article, your song, your art, your video, who pays you for that?

And what if I told you the answer to that super modern, futuristic problem wasn’t some new tech, but an idea someone had more than 60 years ago?

Yeah, way before the internet we know was even a thing.

And that gets us to the internet’s big fundamental problem.

It’s the very thing one visionary tried so hard to prevent all those years ago.

A world where our collective knowledge gets scraped up with absolutely no payment or even credit going back to the people who created it in the first place.

Sound familiar?

Ideas just being ripped away from their creators. WTF?

Then copied and pasted over and over again, losing all their original meaning.

And the creators, they get nothing. No credit and definitely no compensation. we have a system that just doesn’t value the act of creating things online.

So to find a way out of this mess, we have to go back. Way back.

We need to talk about this guy, a philosopher and tech visionary named Ted Nelson and his totally radical, almost mythical idea, Project Xanadu.

Now, to really understand how far ahead of his time this guy was, just look at this. Nelson dreamed up his global docuverse in 1960.

That is a full nine years before two computers sent the first ever message on the ARPANET and an incredible 29 years before the World Wide Web was even invented.

And Nelson has never, ever been quiet about how he feels.

To him, the web we use every single day isn’t some great achievement. It’s the cheap knockoff that replaced his far, far better idea.

So what was his grand vision?

Well, Xanadu was supposed to be this universal library where every single document was connected with permanent, unbreakable, two-way links.

You could actually see how ideas were connected to each other.

And here’s the killer feature.

It had a system for micropayments built right in from the very beginning to automatically pay creators any time their work was quoted or used.

So that was the dream, but that’s not the web we got, so let’s talk about that.

In Nelson’s eyes, what Tim Berners-Lee built was just a stripped down, oversimplified shadow of what Xanadu was supposed to be.

You know that thing we click on a hundred times a day? The hyperlink? Nelson absolutely hated it. He called them jump links. Just a blind, one-way leap into another document.

To him, it was like imitating paper, not inventing a whole new way for ideas to connect. It was a failure of imagination.

Xanadu promised two-way links that could never break, so you’d always have context and credit. The web? It gave us one-way links that break all the time.

Hello, 404 not found.

Xanadu also had payments built in. The web gave us free. Paid for by ads, and all the problems that have flowed from that.

Basically, Nelson’s whole point was that we got was completely backward.

He wanted computers to be organised like the human mind, messy, interconnected, associative.

Instead, we got a web that forces us to think like a computer with these rigid files and folders.

He literally sees modern computing as a prison for our minds.

Share

Righto. So if Xanadu’s so brilliant, why aren’t we using it right now?

Well, the simple answer is, it was an idea that showed up decades before the technology needed to actually build it existed.

The web won because it was simple, maybe too simple, but it worked on the computers of the day.

Xanadu, on the other hand, was an absolute beast.

It needed processing power, storage, and networking that was basically science fiction back then.

And most importantly, there was just no way to make those tiny frictionless micropayments that the whole system relied on.

Which brings us to this little number, 402.

What on earth could that mean? Get ready for this, because it’s a pretty wild piece of internet trivia.

It’s the HTTP status code 402 Payment Required.

And what’s so fascinating is that Tim Berners-Lee himself actually put this code into the web’s original design.

It was a direct shout out to Nelson’s vision, but it was never, ever used.

Why?

A web browser can’t exactly have a credit card, and there was just no way for it to send and receive money.

But that was then. And now things have changed. A lot.

Which brings us all the way back to our first question. Has technology finally caught up to Ted Nelson’s 60-year-old dream? What’s different now?

Well, the good news, is it’s just about everything.

• In 2009, we got internet native money.

• In 2014, we got stable dollars that live on blockchains.

• And in 2025, we’re looking at blockchains that can handle tons of transactions, almost instantly, for next to nothing.

All the pieces are finally here.

And here’s the bottom line.

The one thing, the main technical hurdle that stopped Xanadu dead in its tracks - the inability to do tiny, fast, cheap payments - that problem is solved.

The system of commerce that Nelson was dreaming about is now for the first time actually possible.

His original vision of a fairer, more connected digital world, it’s within our grasp.

Which leaves us with one last huge question.

Is it maybe time to look back at that mad loony dream from the Sixties?

Should we now be try to rebuild our digital world on these old, almost forgotten principles?

Well, it’s happening.

A few months ago, I reported how $70 billion cyber security giant Cloudflare is was dusting off 402 to make AI’s pay. I had them on the pod.

Then I featured the CEO launching in with some choice commentary and declaring himself the new AI Sheriff who make make the thieving varmints pay.

Microsoft has also begun quietly signing deals with major publishers using the same pay-for-use protocols. It even has a name now: Pay per crawl.

Whether this becomes the way creators are paid for their contribution to the web is yet to be seen.

But if a 60-year-old idea is still animating the biggest brains at the largest most innovative companies, there must be something in it.

Don’t you think?

Epic Games, the maker of hit game Fortnite, squaring off against Google, the giant behind Android. It’s been rough and tumble - and very entertaining.

And just when I thought we were settling in for years of appeals… Boom! It’s over.

But not in the way I, or most commentators, expected. It’s turned into something completely different.

Today, I’m revealing what’s just happened.

• How it will lead to better products and lower prices for three billion consumers.

• Help publishers who make apps and sell subscriptions in the Play Store, and

• Put Apple on notice that it’s now an outlier with its high anticompetitive pricing.

Welcome to new paid subs from the digital leadership team at German publisher Burda which has 500+ brands in 17 countries, Canada’s Island Press, and The Ideas Business in Australia.

Share

And more subs from the BBC newsroom in London as well as BBC Arabic, the boardroom at Swedish publishing giant Bonnier, the leadership team at FT Strategies, the C-suite at India Today which reaches 295 million through its TV and online network, ad giant Omnicom, a director at British copyright lawyers Addleshaw Goddard, the IAB in Australia, the C-suite at The Telegraph in London, and our first sub from ecommerce platform Shopify. Hey there :)

Upgrade to paid

Headlines

There’s a lot going on, so let’s kick off.

🚨 AdExchanger reports Google and the US Justice Department have filed updated papers ahead of the ad tech antitrust ruling.

It’s the last chance for both sides to sway judge Leonie Brinkema. At stake is the future of the US$750 billion digital advertising market.

Google’s final argument is that f*cking with it will break the web, and it’s happy to make some (minor) changes, so this is all a waste of time and energy.

The DoJ - backed by a slew of publishers - argue it must happen because Google’s a three-times illegal monopoly and can’t be trusted. They want GAM and AdX sold off.

Google lost its app store antitrust trial (see today’s lead story), it also lost the search trial but walked away with a soft penalty, but this is the biggie. Expect a decision before the end of the year.

Refresher: Google on trial

In other news:

• The UK’s largest parenting site launches a child-safe smartphone - LinkedIn

• Apple closing in on a $1 billion-a-year deal to use Google for Siri - Bloomberg

• Pinterest crashes 18 per cent as it loses ad ground to its rivals - Reuters

• Snap signs a $400 million deal with bad boy Perplexity for chat AI - Bloomberg

• Amazon sues to stop Perplexity using AI agents to buy stuff - Bloomberg

• New York Post says Facebook is censoring its BLM investigation - New York Post

• Getty largely loses a landmark UK lawsuit over an AI image generator - Reuters

• Anthropic says its revenue will rise to $70 billion by 2028 - TechCrunch

• Rocketship Reddit launches its pro toolset for publishers - Press Gazette

• Spotify says it now has 500k video podcasts, and 400 million viewers - TechCrunch

• NY Mayor candidate adds Big Tech’s favourite nemesis to his team - Bloomberg

Got a story?

Reuters also reports that global investors are sweating on whether AI is a bubble. More than half think it is, yet global markets are booming. WDYT?

Kill your commute

🫣 Watch: A Redditor has used AI to bring a prehistoric zoo to life. It’s roarsome 🦖

📖 Read: Just WTF are we now in the chaos of rules and regulations in the AI era?

📻 Listen: How a cold-blooded murder in Africa could unravel Meta’s algorithm.

Listen to my pods on Substack

🤯 Did you know

Google Gemini can create professional presentations from a prompt and some uploaded docs. Yeah, I was doubtful too, so I uploaded Reddit’s financials and its shareholder update, said go, and…

Download

Reddit Q3 2025 Shareholder Letter

1.93MB ∙ PDF file

Download

Download

Wow, that’s just saved about a year of work! WDYT?

The Big Story: The Epic truce

The opening lines to Faithless’ club anthem God is a DJ contains these words: “It’s in minor keys, solutions and remedies. Enemies becoming friends, when bitterness ends.”

I’m not sure whether Google’s Sundar Pichai and/or Epic Games’ combative CEO Tim Sweeney are fans of progressive trance house, but they seem to be feeling the love.

The two have been fighting through the courts for years, running up massive legal bills.

Google wanted to protect the massive margins in its app store business. Epic wanted to tear them down and sell its own stuff without paying Google’s tax.

Google then lost the case. It appealed, and then lost that too, and in a few weeks, a judge is due to tell them both what the penalties will be.

At least that was the plan until this week, when the two warriors downed weapons and had a very acid house love in.

But no-one has surrendered here. Not by a long shot. What’s actually happened is kind of wild.

They’ve called a truce so they can co-author the new rules of the road for the entire Android ecosystem.

Today, I’m breaking down what exactly is in this surprise peace treaty using the court papers both parties just filed.

And explaining what it means for publishers selling apps and subscriptions, and for price-conscious consumers.

Upgrade to paid

Long term and global

Both parties are framing this as a win for everyone.

After Epic won its big case in December 2023, the court basically handed down a set of rules, and the bickering started.

This new settlement ends that. It urges the court to scrap that confusing court order and create a new plan.

The first fix is making it way easier for other app stores - like Epic’s Game Store to get into three billion Android phones without having to go through Google’s Play Store.

The court’s plan was the same, but only lasted three years and was just for the US.

Google and Epic’s new deal lasts until 2032, and because it changes the Android operating system itself, the effects are basically global.

That’s huge.

And instead of forcing Google to list competitors in its own store - which was always awkward - it creates a totally new system for installing them right from the web.

Epic argued at trial that Google used scary-looking warning screens to freak people out if they tried to install apps from outside the Play Store.

Now that all goes away, and is replaced with a single neutral-looking permission screen. Users find it, approve it, and install it. Done.

But what’s really important to note is that this doesn’t wipe the slate clean. A lot of key antitrust protections from the original court order remain.

Google is still banned from paying off phone makers to block competition, or forcing developers to keep their apps exclusively on the Play Store.

Google - it seems - has come to the party.

And that has major implications for Apple which continues to fight actions across the world to cut its app store fees and open up too. Oh, to have been Tin Man Tim this week…

Upgrade to paid

Lower fees

So that’s app stores, but it’s only half the deal. The other half is payments.

Google has steadfastly defended its rule that developers must use its billing system and pay a 30 per cent fee on everything.

Under this new settlement, developers get a ton more freedom.

• They can offer their own payment system right alongside Google’s and

• They can actually encourage users to sign by charging less than Google.

They can even show consumers both prices side by side to show who has the better offer.

Think about that. Google spent tens of millions arguing developers couldn’t even hint there was a better deal on their own website.

Now it’s OK, and that’s a level of direct price competition never before seen inside one of these huge app ecosystems.

That leads to the $48 billion question for developers: If they use their own payment system, do they still owe Google a cut?

The answer is yes, but it’s a lot lower.

Google’s argument has always been that it deserves to be paid for the platform and services the Play Store provides.

This new settlement puts a hard ceiling on how much Google can charge.

The first cap is 20 per cent for some transactions (but it doesn’t yet say which) but that’s the absolute max Google can take and a third less than now.

And that sounds bad, but wait, because there’s more coming.

For other transactions (not yet stated) the cap is much much lower, down to just nine per cent.

By sitting down and negotiating these, Google and Epic have avoided what would have been another long messy court fight over what a reasonable fee is.

Who falls into the nine and 20 per cent categories is still to be decided, but this is a massive concession by Google, and a giant win for Epic Games’ CEO Tim Sweeney.

Done deal?

Well, not quite.

Epic and Google are on the same page, but now they have to get the judge to sign off on it, so there are still a couple of hurdles left.

First up, there’s a court hearing on December 11 where the judge will review this whole new plan - but there’s also a second really critical condition.

Google is also settling a different case on the same issue with a bunch of US states.

The new deal only holds up if the judges agree to the changes Epic and Google are suggesting.

And Judge James Donato is no friend of Google. He lambasted the search giant at the trial for deleting millions of instant messages that he wanted as evidence.

That is still headed to another trial - which will be big and which hardly anyone is covering, so keep an eye on this one.

That all means it’s still a little complicated, but will he get in the way of a mediated settlement that achieves the goal of reducing fees and increasing competition?

That feels kinda unlikely.

Peace in our time?

Umm, what’s going on here? When you step back, what this settlement is really about is creating a final, lasting peace.

• It ends a five-year war.

• It gets them out of fighting over how to implement rules, and

• It creates a clear, stable playbook for Android that lasts almost a decade.

This all leaves a really fascinating question.

For years, the entire narrative has been about regulators and competitors dragging Big Tech into court, kicking and screaming.

But here, we have two of the biggest adversaries sitting down together and literally writing their own regulations.

Could this be the new playbook? Is this antitrust pressure doing its job? Is this how Big Tech is tamed, through negotiated peace instead of endless war?

I’m often accused of optimism. Sorry haters, that’s just me, but it’s a huge question. Honestly, only time will tell.

Follow me on YouTube

🗓️ Coming up

• Tonight: Join me and Rod Sims, the architect of Australia’s ground-breaking news media bargaining code, as we explore the future of Big Tech in a special Scotch & Watch session before a live audience at Bookoccino in Avalon Beach, Sydney, tonight at 7pm. 🥃🥃🥃

• And join me on Monday for another special edition when and I are joined by Substack founder Hamish McKenzie to hear his plans for publishers

For now, this is the future checking out. See you later 👋

The deal terms were clear: Google would pay Reddit US$60 million-a-year - and that’s what the headlines reported - but there was far more to it .

I’ve been investigating it ever since, and today, it’s time for the real story.

I can now reveal how that deal has so far earned Reddit $39 billion, and it’s still accelerating.

The social site that was loss-making before, is now worth the same as Coca Cola, Samsung, and Honda.

Today, I’m sharing how Reddit did it, and the profound implications for publishers as they explore the right model to licence content to AI.

Strap in... This is a big one.

Before we get to it, welcome to new subs from The Washington Post, BBC, Time, The Independent, The Guardian, WSJ, Bandcamp Daily, FT Strategies at The Financial Times in London, German mag and digital group BurdaVerlag in Munich, Axel Springer in Berlin, Fortune magazine in London, Tamil publisher Vikatan in Chennai, India, boutique media M&A broker MediaFund in London, Canva in Sydney, Puzzler Media in the UK, and many more.

🗓️ I also have a special alert. On Friday night, I’m joined on Scotch and Watch live by Rod Sims, the former chair of Australia’s competition commission.

He’s the plain-speaking regulator who did what the world failed to do - made Big Tech pay for news. Join us at Bookoccino in Avalon Beach as he shares his insights.

No need to buy tickets - just turn up - but’ll be standing room only. 🥃🥃

Here we go…

Headlines

🚨 People Inc has become the first publisher to sign an Airgap-style deal with Microsoft. This, in combination with Cloudflare’s pay-per-crawl, fires the starting gun on AI licensing at scale - Digiday

Plus:

• Brussels really wants Google broken up as deadline closes - Politico

• Almost one in 10 US articles at least partly AI generated - Press Gazette

• Tech’s cricular deals raise fears of a bubble? - NY Times

• Google yanks Gemma AI over false Senator sex claims - Ars Technica

• Fox News reports largest ad revenue quarter in its history - Forbes

• GM plans to dump CarPlay, Android Auto, for its own tech - The Atlantic

Got a story?

Kill your commute

🫣 Watch: A judge has gone for Apple’s jugular collapsing its ability to charge 30 per cent fees in its app store, and one of its execs might be headed for jail.

📖 Read: Meta has warned for the first time that upcoming court cases may have a material impact on its business. Read into the court case that has Zuck worried.

📻 Listen: DuckDuckGo, the browser built with privacy in mind, explains how publishers can turn doing the right thing, into a profit.

Listen to my pods on Substack

The Big story: Reddit’s friend in high places

In February 2024, Reddit did a game-changing deal with Google. Since then - and out of the sight of most people - Reddit’s been on an unprecedented tear.

Its growth was revealed last week in its latest financials, when it reported revenue up 68 per cent to US$585 million - in one quarter.

And after years of losses, its net profit over that same period was a jaw-dropping US$163 million.

A company that for years struggled to make a cent is suddenly printing money. Its forward guidance is to pass US$2.1 billion over the full year.

The question is how? And is it a blueprint for the future of media and AI deals?

That’s what I’m breaking down today.

Reddit signed a formal AI training agreement with Google where it was paid US$60 million-a-year to give Google access to its community conversations.

These are considered by many as the most human dataset on the internet. It was that exchange that made the headlines.

It was - after all - more than News Corp was able to get out of OpenAI just a few weeks later.

But there was more to the Reddit deal. Much more.

On the face of it, it was a match made in heaven.

Google needed billions of real human conversations to teach its ailing AI models how people actually talk. Gemini was falling behind. OpenAI was racing ahead.

And Reddit needed money, but far more than that, it needed momentum, because it was headed to a listing on the US markets.

Each had what the other wanted badly - and the timing was perfect.

The deal delivered two immediate advantages for Reddit.

• First, credibility for the IPO. The years of losses had dulled investor appetite, but with Google showing it so much attention, the listing story suddenly got sexy.

The float, that looked like it would be US$5 billion, suddenly priced closer to $9 billion. Reddit coined it in.

• Second, distribution. Google had scale that it could share, so it twiddled its dials and juiced Reddit’s presence in search and, notably, in AI Overviews.

You probably noticed. Search almost anything over that period and suddenly a Reddit thread was right at the top.

That oxygen mattered. Daily active users leapt, and Reddit now has 116 million people showing up every day.

On last week’s earnings call, CEO Steve Huffman revealed Reddit is now the third most-visited site in the US after Google and YouTube, and ahead of Facebook and Amazon.

Quite a feat in less than two years, and whether you love it or hate the deal, Reddit is now in the big leagues.

Here’s the core strategy. Think of it as a four-step flywheel.

Step one: Reddit’s users made Google’s AI better, and Google gave that better AI to Reddit to improve its product, through faster answers, smarter search on-site, better tools for moderators...

Step two: That better product attracted more users and kept them around longer. Engagement went through the ceiling.

Step three: More engaged audiences delivered better ad performance - especially when Google shared its AI secret sauce to match creative with context. CPMs, sell through and yield all shot up. Reddit’s ARPU on US users is now US$9.04.

Step four: Those ad gains drove revenue to a record $585 million, which Reddit is reinvesting to improve the product again.

Round and round, each step is amplifying the next. It’s live and working and delivering jaw-dropping results.

Let’s focus on a number that flips the story with Reddit. US$163 million in quarterly profit - and the implications for the wider industry.

While most publishers are looking at AI as a way to cut costs, Reddit is using it to drive top-line power and growth.

Quarterly revenue leapt 68 per cent YOY. Advertising revenue is up 74 per cent - and that’s three times the ad growth at Google, Meta, Amazon and other ad gains.

The takeaway is that this isn’t accidental traffic - it’s an effective monetisation engine that’s accelerating - executed by a publisher that’s in lock-step with a hyperscaler.

Here’s why it’s so effective.

Reddit is not just another content site, like most publishers. It’s a town square. It’s where the conversations of the world are taking place.

It’s a montage of lived experiences, with more than 1,000 distinct communities of interest - each a little city of people who care passionately about something shared.

It’s a living library of human experience with more than 23 billion posts and comments; messy, human, wildly specific - and therefore incredibly valuable.

Where most publishers shut down comments years ago and outsourced them to social to dodge moderation risks and legal issues, Reddit did the opposite and made conversation the product.

That now gives the AI-era Google something it can’t buy from publishers: Authentic, constantly updated human dialogue at scale.

Looking in publishing’s rear-view mirror, that feels like a missed opportunity.

Over at the Googleplex, they knew what to do.

They turned the knobs on search and AI Overviews to point traffic in Reddit’s direction and the predictable happened.

Traffic climbed, engagement deepened, millions of new users discovered communities they didn’t know existed, and the surface area for relevant ads expanded.

Google juiced Reddit - shamelessly - and changed the game. Google got lakes of data back to close the gap on its AI rivals.

And then there’s the money engine.

Reddit - like TikTok - invested early in building its own ad platform and auctions, and chose not to gamble its future by relying on Google’s ad tech.

The majority of Reddit’s ads run in-app, and the backbone of its sales story is contextual.

It uses AI and other tools to match ads to the interest and the pages, instead of shadowing users around the web.

That will become more strategically important as third-party tracking erodes and context and intent become increasingly valuable.

Reddit’s filing said: “The foundation of our ad performance is based on context and interest instead of tracking users based on personally identifiable information.

“Our lack of reliance on third-party data makes our offering more resilient to the loss of signal that other platforms rely upon as well as forthcoming changes to the internet ecosystem.”

And with Google’s ad tech machine learning tools now in the loop too, Reddit is reporting materially better conversions. It reported a 20 per cent lift on customers’ KPIs.

That’s the kind of delta that CFOs - and advertisers - remember.

Let’s add it all up.

• Reddit has been transformed into a highly profitable, fast-growing, ad-funded media company which 444 million people use every week.

• It has developed a symbiotic data partnership with Google that improves both AI and user experience and feeds it back into growth, and

• It’s backed by an ad system that scales with content without creeping into surveillance as privacy becomes a trigger topic in political circles.

The result is that Reddit has moved from cult favourite to central pillar of the web.

It’s market cap this morning was $39 billion.

Let that sink in. $39 billion. That’s more than the combined value of News Corp, Paramount and the New York Times - and it’s achieved the turnaround in 18 months.

So, what can the rest of media and tech learn from Reddit’s play?

Four plain lessons.

One: Publishers’ most valuable asset is the authentic human content that only they can host. Publishers are still audience rich - and that audience is key. It’s a moat. But the way to unlock the value is to engage them in conversation, not just tell them the news.

Two: A smart AI partnership is not a press release - it’s a key that unlocks the value inside that moat. Choose partners who have a genuine need - and who will put their shoulder into making it work. Be strategic, not desperate. Figure out the why?

During my Scotch & Watch podcast’s world tour stop in London I pointed out that News Corp’s deal signed two months later with OpenAI has fallen flat.

News Corp has barely mentioned its OpenAI deal in its latest financial filings because it’s not making any material money. Remember: Reddit’s at $39 billion and rising.

Three: Design the product so user experience and monetisation reinforce each other. Better UX → more time on site → better ad relevance → more revenue → better UX - and then reinvest it in building what comes next.

Four: Treat creators and moderators as partners, not inventory. Empower them with tools and they will compound your growth. I love what Reach is doing with Substack. We need more of this.

The creator economy is outpacing traditional media, as my deep dive into The Reuters Institute study last week proved.

If Reddit is a model for publishing, then it also poses some tough and uncomfortable questions.

It asks publishers to dump much of the accepted wisdom of the past 50 years to rebuild for the next 50.

It means being open to partnering with hyperscalers. I’ve favoured Microsoft since my very first post on this newsletter in July 2023, but Google and OpenAI are there too.

It means ascertaining terms that respect the value of your data, your brand, and your community - and finding where it dovetails with the needs of your partner.

We’re still years from trusted AI, yet publishing holds a key to getting humanity there. Meanwhile, the AI companies are spending $1 billion-a-day and need to show a return.

There’s a path.

Publishers need to lean into context over surveillance, performance over page views, and conversation over one-way publishing. Stop telling, start listening.

It’s long past time to give readers a platform to talk about the stories that we create.

And in a world where AI can spit out infinite good enough content, what still defends its value?

The answer is messy, passionate, specific communities of real people. People showing up to talk to other people they trust, in a place they trust, with a brand they trust.

That’s impossible to copy with machines, and therefore uniquely defensible.

Reddit bet the company on it - and then chose a partner who could amplify it.

So, in conclusion, friends in high places helped Reddit.

But the groundwork was already there - living, breathing communities at scale, willing to do the one thing algorithms can’t - and that’s being human together.

Tie that to an AI platform that needs you as much as you need it, and the flywheel spins.

Reddit didn’t just get lucky. It got aligned. And alignment, in this era, can be everything.

And these are the numbers Reddit’s Q3 FY2025 filing: (All US$)

• Revenue up 68 per cent YoY to $585 million.

• Ad revenue up 74 per cent to $549 million.

• Daily active users up 19 per cent to 116 million.

• Weekly active users up 21 per cent to 443.8 million.

• Average revenue per user (ARPU) up 41 per cent to $5.04.

  • US ARPU $9.04.

  • International ARPU $1.84.

• Net income was $163 million.

  • Profit margin 28 per cent.

• Adjusted EBITDA $236 million (40 per cent margin).

• US revenue $480 million - up 67 per cent.

• International revenue $105 million - up 74 per cent.

Guidance for the next quarter of revenue between $655 million and $665 million - putting it on track for FY revenue of ~$2.13 billion.

Wow.

🥳 Join the party

Planning your end-of-year party? Book me for a punchy year-in-review or rapid-fire predictions for the year ahead? I’ve headlined stages from SXSW to Broadway.

Sydney-based, happy to travel, but dates in late November/December are limited. Want a night your team will still be talking about in January? Drop me a line.

Of course it wants it. Chrome directs 66 per cent of all web traffic, and whoever has it has the data for a US$200 billion ad business and a way to super-charge AI training.

But there are three very good reasons why Perplexity has a snowball’s chance in hell of getting its hand on the most valuable digital ad accelerator the world has ever seen.

First, join me in welcoming a crowd of new subs in what’s been the fastest grown of new sign-ups to the newsletter in a year. Great to have you.

Hi then to the leadership team at Magnite, the commercial team at The Guardian, The New York Post product team, The Financial Times in London, La Nacion in Argentina, the agency team at Amazon Ads, global consultants Edelman (publishers of the trust barometer), Australian radio giant ARN, advisors Kekst CNC in Riyadh, Nexxen in Japan, the UK’s University of Westminster, the US bureau of Aussie broadcaster Nine, among more than 100 others. Please share with your friends and join the party 🥳

Share Future Media

And don’t forget, we are now days from hearing how Judge Amit Mehta intends to break up Google’s illegal monopoly on search.

But first, Perplexity ain’t gonna meet muster on getting Chrome ‘cos…

Read more