Inside Big Tech's epic 20-year war to win search
The fascinating collision of ego, power, politics, and money, that decided how four billion people access the open web...
This article was first published on my Future Media Substack. For first run news and cutting-edge analysis on the collision of Big Media and Big Tech in the age of AI, subscribe here .
Hello to new subs from News UK (hey Dave ??), the United Nations (guete daag Maya ??), Google (now the second largest subscriber after News Corp ??), Big Tech lobbyists the Digital Industry Group, the UK’s Evening Standard, Hello! magazine, mag group Bauer, ITV News, KPMG, sports streamer DAZN, InMobi, the University of Technology Sydney, The Tow Center for Digital Journalism at Columbia University, the international affairs division of the Republic of South Korea, The Royal Bank of Canada, ThinkTV, and Australia’s ABC among others.
Welcome all, let’s get to it…
There’s nothing like a string of internal mails to reveal the big, true story of what’s going on at the absolute top of Big Tech, and this post has some crackers.
The focus is a story that’s Silicon Valley folklore; how Micrsoft burned $100 billion and its best minds and never even scratched Google’s dominance in search.
The mere thought of it makes Microsoft bristle, and Google grin. It’s why Microsoft chief Satya Nadella was so incautious as to joke he would make Google dance over AI.
Microsoft believed the secret to taking Google’s crown was to break its exclusive default deal with Apple.
That was going to be expensive. Mind-blowingly, shake the Earth, kinda expensive.
But with Google’s search earnings ballooning to $162 billion-a-year, and the iPhone delivering half the search volume, anything was good business. And I mean anything.
It’s a strange story. Hard to believe were in not in court papers and internal mails. It somehow manages to rope in Eurythmics lead singer Annie Lennox at one point.
But this is how the three-way-war between Microsoft, Apple, and Google, played out…
Two kids at Stanford wrote a doctoral thesis on how to find stuff on the internet and turned it into a $2 trillion monopoly. It’s one of the great business stories of all time.
Larry and Sergey’s school experiment grew into Google, listed, gave up trying not to be evil and became the default entry point to the web.
Half the world’s population use Google today, far in excess of the 15 per cent ruled by Alexander the Great at his peak. Google is a significant landmark in human history.
But now we know how it did it. It pillaged competition, and crushed rivals by brandishing billions of dollah dollah bills to buy its dominance.
Rivals Apple, Samsung, and more were paid off from its $96 billion cash stash , to secure default search positions.
Google hoovered up data that was unlocked from iPhones, Android, and Chrome, and strip mined the world’s privacy to build its $238 billion search ad monopoly .
Now, that’s been deemed illegal, and Google faces break up. And well… hell yeah, let’s go to the future!
That has obviously been the focus since last week’s verdict, but after reading the 91,588-word ruling, and combining it with my own research, there’s another story.
It’s how Google kept Apple and Microsoft on search’s sidelines, and it’s told through the emails of those most closely involved - Sergey Brin, Steve Jobs, and Sundar Pichai.
It’s fascinating, and reeks of egos, power, politics, and money, but let’s begin at the beginning, with the mechanics.
This ISA big deal
Google knew it could hoover up the entire search advertising market if it could ensure that it got the lion’s share of search queries. Common sense, right? Business 101.
The epicentre of its strategy was a type of contract that Google called an ISA (or Internet Service Agreement).
It set out to sign ISAs with everyone and anyone that had a mass audience. It would go fish where the fish are…
These ISAs detailed what Google would pay rivals to hand over the valuable default search positions in their phones and browsers.
Google’s gold was enough to tempt almost everyone to stick the ubiquitous G on the world’s most popular devices.
The largest though was Apple, which was soon sending Google 10 billion weekly search queries through its Safari search browser.
That exploded when it extended to 1.5 billion iPhones. Google’s volumes ballooned to more than 50 billion queries a week - or 2.6 trillion annually.
The terms of the ISAs meant the search traffic was sent to Google almost invisibly, with consumers none the wiser where their questions or data was going.
Let me ask you this: Back in 2002, did you know that when you typed a question into Safari, that Apple was handing you off to Google for money?
Probably not, but what about yesterday?
Even as Apple is marketing itself with attack ads against Google highlighting its privacy creds like this one - it’s still banking $26.3 billion from Google’s ISA.
The hypocrisy is breathtaking for most of us , but at Microsoft it was too much to bear.
CEO Nadella was pissed. He had burned $100 billion of shareholders’ money on trying to buy Bing a toehold, but it was nowhere, and a laughingstock.
It was time for some shenanigans, and a Big Tech stand-off for the ages.
Deal of the century
Google first hit on the idea of an ISA with Apple in 2002. Eminem had just won album of the year and Jenny was from the Block.
Google’s money led Apple’s famously product-focused CEO Steve Jobs to make changes so subtle they were near invisible for users.
They were so subtle in fact, there weren’t even any headlines about what was to become the deal of the century.
Out of sight, and without billions of customers being told, the economics of the web were rewritten in ways that would send Big Tech valuations into the trillions.
All that changed was that when Apple users went to Safari and typed a query into the address bar, that question was sent to Google to provide a response.
It seemed so innocent, and so small, that no money even changed hands - at least not at the beginning.
The fact was, Apple needed a search engine, and Google’s was the best.
Meanwhile, Google needed scale to suck up more data to provide better results.
The first deal wasn’t even exclusive, but that didn’t last long either.
A hurricane of traffic
Google soared, and by 2005, all Silicon Valley was agog.
The scuttlebutt among the Kombucha classes was that the Apple deal was powering Google along, and its crosstown rival Yahoo began hatching a plan to steal it.
It’s unthinkable now, but at that time, Google and Yahoo were battling to win the search market. Both had billions at stake and wanted that delicious Apple slice.
Yahoo’s plan was simple. Google wasn’t paying, so it would wow Apple with a big money offer to gazump the new kid on the block and replace it.
Yahoo was still revered as internet royalty back then, but it was beginning to look tired, and a little long in the tooth.
Google was a thrusting newbie, full of bravado, and flush with cash. When it heard that Yahoo was sniffing around, it launched a brutal counter-offensive.
The dawn of exclusivity
Google hopped on calls with Jobs and threw bank at the deal. Google penned a new ISA, paid Apple $10 million, and sweetened it with a 50 per cent rev share.
The ISA locked in a three-year handshake that made Google the exclusive default search engine on Safari. Looking back, it was an amazing leap of faith by Google.
Apple was having ups and downs, but Google was growing like a weed. Ad revenue was $6 billion , and doubling year on year.
Google’s ballsy move underlined the strategic importance of the tie-up, and time would later show that it turbocharged the trajectory of both companies.
The i in iPhone
Two years later in 2007, when Apple launched the iPhone, Google was the first to the party, locking in more defaults, this time on the iPhone, iPods, and Safari for Windows.
The deal made one rising star Google product manager a bit queasy. Sundar Pichai - later to become CEO - was so concerned about optics he emailed Larry and Sergey :
“I know we are insisting on default, but at the same time I think we should encourage them to have Yahoo as a choice in the pull down or some other easy option. I don’t think it is a good user experience nor the optics is great for us to be the only provider in the browser.”
But Google was now a digital superpower and ready to flex its muscle.
Apple tried multiple times to persuade Google to allow customers a choice of search engine, but it was a hard no.
Google was clear it would rip up the multi-billion-dollar revenue shares if any other search popped up and had its lawyers on speed dial.
Even with billions at stake, Apple persisted in testing Google’s resolve.
It offered an olive branch, saying it would accept a lower 40 per cent rev share, but Google slammed the door shut, and rejected it out of hand.
Apple tried again in 2012. Google still wasn’t having a bar of it. “If they wanted to receive revenue share” it must maintain Google as the exclusive Safari default.
And with both sides making billions, and Microsoft’s baby Bing barely out of the cradle, they swallowed the brinkmanship and signed a new deal in 2014.
领英推荐
Bing or bust
Microsoft growled, clenched its fists, and decided enough was enough. It would go big to break the partnership, and with the ink barely dry, it made its pitch.
It would pay billions, but the main thrust was that “increased competition between Microsoft and Google is in Apple’s long-term economic interests”.
Ultimately though it would come down to numbers, and Microsoft made it clear it was willing to be mind-blowingly generous.
Internal emails showed Microsoft was “willing to provide the majority of profits” to Apple, before revealing it would throw down an unprecedented 90 per cent rev share.
At the time in 2015, that was worth just a smidgen under $20 billion over five years.
Apple, sitting pretty with two digital goliaths fighting for its favour, said no.
Microsoft then went all in, proposing to give 100 per cent of its Bing revenue. Apple still said no.
And when Microsoft went back a third time, it offered to give Bing lock, stock, and barrel, to Apple for free.
Microsoft “thought they had great search, and they said that with Apple’s volume, they could be even better”, the court papers say.
But even after being offered Microsoft’s $100 billion search contender, and every cent of advertising that it generated, Apple still said no.
I’d bet that stung.
Crunching Apples
Internal mails from Apple’s lead negotiator Eddy Cue revealed why.
He feared Microsoft was “horrible at monetising advertising”, believed Bing was inferior, and felt Apple’s “customers wouldn’t use it.”
Despite the doubts, Apple set their bean counters to run the numbers.
They concluded the Google deal was worth $40 billion over five years, and $70 billion over the five after that. It dwarfed Microsoft’s $20 billion offer.
Cue mailed the Apple leadership team: “Clearly, Microsoft can’t commit to these numbers or even anything close to them.”
The only way out, he said, was if Apple “viewed Google as somebody they don’t want to be in business with and are willing to jeopardise revenue to get out.
“Otherwise, it’s a no brainer to stay with Google as it is as close to a sure thing as can be.
“They have the best search engine, they know how to advertise, and they’re monetising really well.”
Microsoft promised to go back to the drawing board, but at Apple, Cue was hardening a view that Bing wasn’t just a no for now, it was a hell no forever.
Through the looking glass
Over at the Googleplex, execs were keeping a wary eye, and launched a top-secret project called Alice in Wonderland.
A small team of handpicked analysts were tasked with running scenarios on what it would take for Apple to defect to Microsoft.
The math suggested Microsoft would need to pay Apple a 122 per cent revenue share to match the sums Google was delivering.
Project Wonderland concluded: “It will not be possible for Alice to match our payments profitably.”
Across at Cupertino, Cue had already made his mind up, mailing his bosses: “I don’t believe there’s a price in the world that Microsoft could offer us.
“They offered to give us Bing for free. They could give us the whole company.”
That was quite a call. Apple’s 2005 market cap was $60 billion while Microsoft’s was $271 billion. Google, listed just a year earlier, was about to pass $10 billion.
But Google had learned something valuable. Microsoft was not a viable alternative at any price, and that gave Google primacy in all future negotiations with Apple.
And until this year’s antitrust trial, there was almost no chance the deal would be unpicked.
Apple search?
We now know from the court case that Apple banked $26.3 billion from Google in 2021 (the most recent financials revealed).
Apple is Google’s largest single expense, and far more than it spent on R&D to improve search, which has raised a few eyebrows.
In 2024, more than half of all US searches came from Apple devices, and the current Google Apple ISA is dated to run until 2026. Siri also defaults to Google now.
Google payments come at no cost and represent 17.5 per cent of Apple’s operating profit.
It’s expensive, but what Google has paid for is to keep Apple on the sidelines of search.
In court, the judge noted that Apple has the financial, technological, and human talent, to buy - or build - a rival search engine.
It even poached Google’s former head of search John Giannandrea in 2018 to buy in the expertise, but to date, it has not done so to protect the income from the ISA.
Wins and losses
Google and Apple have both run the numbers to see if they are winners or losers under the deal.
First Google.
“Google has long recognized that, if Apple were to develop and deploy its own search engine as the default in Safari, it would come at great cost,” the court heard
“Google projected that without the ISA, it would lose around 65 per cent of its revenue, even assuming that it could retain some users without the Safari default.”
In real numbers, that 65 per cent is $154.7 billion, meaning the deal is a good one for Google.
And Apple?
Well, Apple has invested billions in its own search experiments over the years.
Giannandrea’s team has done the hard yards by crawling and indexing the web, and creating a knowledge graph, but it has not yet hit go.
Fourth time lucky?
In 2018, Microsoft was back at Apple again. This time Apple new boy Giannandrea was leading the review.
In comms revealed by my mates at the Internal Tech Emails Substack, he said:
“1/ their traffic is small. This is presumably because they are nowhere on mobile.
“2/ they are investing a lot more in search than we are. They implied >1000 people.
“3/ they are not launched in most of the countries I think we care about.
“4/ if they need to ‘give away’ the product to get close to the current economics we would need to be really sure that its strategic for both of us.
“It’s a sure sign that their ad tech + marketplace is way worse (more than twice as bad in fact).”
Apple’s VP of corporate development Adrian Perica, who reports directly to Apple CEO Tim Cook, asked: “Could MSFT and us catch up or is the gap much larger in the user space than appears on paper?”
Giannandrea responds: “I have been living on Bing for the last few days. Mostly it works fine. Then in the odd long tail query it just doesn’t.
“A recent example is “Annie Lennox first band”. Google gets “The Tourists” as a web answer. Bing highlights the same answer but shows a box highlighting the Eurythmics. This worries me a lot.
“I don’t think Bing can do better than Google for the search use case unless it spends more on it or has a better mousetrap.
“Not having mobile queries at scale is a huge liability for them since the most important search signal is engagement.
“But it is not impossible. As we noted yesterday the reason a better search engine has not appeared is that it’s not a VC fundable proposition even though it’s a lucrative business.
“Can I imagine that Apple can build a search engine to compete? Yes, but it’s probably not the best way to differentiate our products.”
Deep pockets
The court papers said: “In late 2020, Google estimated how much it would cost Apple to create and maintain a search engine that could compete with Google.
“Google estimated that the total capital expenditures required would be in the rough order of $20 billion.
“Google further estimated that, if Apple needed only half of Google’s infrastructure, it would have to spend $10 billion to get it off the ground, plus $4 billion annually
“On top, if Apple could sustain a business with only a third of Google’s engineering and product costs, it still would cost Apple $7 billion annually.
“The cost of maintaining a search engine once built runs into billions. In 2020, Google spent $8.4 billion to operate its search engine.”
Apple even ran its own numbers.
“Apple has estimated that it would cost $6 billion annually…” and “even assuming that Apple retained 80 per cent of queries with its own search, it would still lose $12 billion in the first five years”.
The court added: “Monetising it is also expensive. In 2020, Google spent $11.1 billion to operate its search ads business.”
That same year “Bing earned only $7.7 billion total in search ads revenue”.
And that’s before the challenges and costs of AI, so you can see why Apple is in a holding pattern.
And why Nadella darkly joked in his testimony that VCs see search as the “biggest no fly zone” in Silicon Valley.
Now, if your mates ever ask you why Apple hasn’t built its own search engine, you have an answer.
What’s left hanging though is how Apple reacts, if the antitrust judge rules that its default deal with Google has to be unwound. That’s quite likely.
And it’s also the likely reason Apple signed the recent deal with OpenAI to power Apple Intelligence .
That question is the one at the forefront of investors’ minds. How long Apple continues to be the world’s most valuable company depends on making the right call.
Tim ?
For more reports from the collision of Big Media and Big Tech, subscribe to Future Media >