Highlights: Chaos Monkeys, by Antonio Garcia Martinez


Raw Generation of Kindle Notes, up for Summary + Review

Location 216-217

United Airlines? Whatever the fuck Charlie McGarraugh said it was, as he was the “market maker” in airline credit for Goldman Sachs. The broker of public perception, he was both the market’s conduit and its lion tamer, buffeted by market forces out of his control, but also

Location 223-226

Even amid the perpetual convulsions of fear and greed that possessed everyone on the floor, reason would occasionally out. Like a rock-bottom alcoholic contemplating his vomit-stained sheets through the haze of another postbender hangover, you occasionally asked yourself: How did I get here? How could I do this to myself? Where

Location 293-296

People often asked me in the early days of my tech career how I had gone from Wall Street to ads technology. Such a person almost certainly knew nothing about either industry, or the answer would have been obvious. I did the same thing the whole time: putting a price on a human’s perception, be it of a General Motors bond or a pair of shoes coveted on Zappos. It’s the same difference either way; only the scale of the money pile changes.

Location 294-295

career how I had gone from Wall Street to ads technology. Such a person almost certainly knew nothing about either industry, or the answer would have been obvious. I did the same thing the whole time: putting a price on

Location 357-359

To paraphrase the very quotable Silicon Valley venture capitalist Marc Andreessen, in the future there will be two types of jobs: people who tell computers what to do, and people who are told by computers what to

Location 359-361

The next place where this shift would be seen at whopping scale in terms of both money and technology (though I didn’t realize at the time) was in Internet advertising. And after that, it would hit transportation (Uber), hostelry (Airbnb), food delivery (Instacart), and so

Location 376-379

doing so, with reasoning I’d see again at Facebook: an incumbent in a market dominated by a few, with total information asymmetry, and the ability to make prices on the market rather than just take them, has little incentive to increase transparency. The bid-ask spread—that is, Goldman’s difference in price when buying or selling the same thing—was huge on credit derivatives.

Location 429-433

“Don’t you ever think about venturing out and doing your own thing, and not this big-company stuff, Scott?” I asked, gesturing at the cloud of worker bees around us, pounding their sensible salads before heading back to the trading hive. “My parents ran a small family business, and I watched them go through all the stresses of that . . . the ups and the downs. The uncertainty of their lives was terrible on them. I could never imagine going through that myself. Goldman isn’t perfect, but it’ll be here a long time. I wouldn’t want to live with the insecurity.”

Location 431-431

“My parents ran a small family business,

Location 438-441

companies, including and especially at Facebook, I would always prefer—a hundred times prefer—being subject to the rigors of the market, the fickleness of luck, and the whims of users than to navigate the popularity-contest politics of a large company, surrounded by the mediocre duffers who’ve succeeded in life through nothing more than guile and appearances. Scott Weinstein’s unfortunate example was the best advice he (or anyone else) has ever given me, and one that I ignored to my extreme peril.

Location 452-458

Some police officers are bullies who enable their sadism via a badge. Most officers, however, are really just reactionaries, believers in some simplistic Manichaean duality of the universe, in which everyone is intrinsically good or evil. Sworn defenders of order, these officers protect property and its patrician owners from violence and thievery; they separate moral wheat from chaff, and discern the no-good criminal from the upstanding citizen. Such judgment is their calling, their vocation, with the full backing of whatever jerkwater town is paying for the uniforms and squad cars. To earn their leniency, you just have to be perceived as someone from the good side who has momentarily gone astray and ventured into the dark. Just a little nudge to put you back on the straight and narrow.

Location 485-488

Evaluation time: I was over the limit, but not by much. If I acted like a whiny little bitch and insisted on a blood test at the station (my right under California law), he’d have to arrest me, stick me in the car, and drive me to the station. Then they’d have to set up the blood test. By the time all the crap was done, I’d be below the limit, and he’d have missed two hours at whatever DUI trap I had driven into, ruining his night.

Location 496-498

One week into my new Silicon Valley life, and the lesson was this: if you want to be a startup entrepreneur, get used to negotiating from positions of weakness. I’d soon have trickier situations to negotiate than convincing a cop to let me take a cab. And so will you if you play the startup game.

Location 531-533

Media buying was no longer about putting a square on the automobile or real estate section, but about finding specific users anywhere and anyhow. All this data being generated, stored, and used, by both publisher and advertiser, made room for people who once priced credit derivatives to do the same for parcels of human attention instead.

Location 535-540

that power was shifting inexorably from the publisher, the owner of the eyeballs, to the advertiser, the person buying them. If my advertiser data about what you bought and browsed in the past was more important than publisher data like the fact that you were on Yahoo Autos right then, or that you were (supposedly) a thirty-five-year-old male in Ohio, then the power was mine as the advertiser to determine price and desirability of media, not the publisher’s. As it turned out (and as Facebook would painfully realize in 2011, forming the dramatic climax of this book), this “first-party” advertiser data—the data that companies like Amazon know about you—is more valuable than most any publisher

Location 657-663

It’s me, the author, taking the state inside my mind and, via the gift of language, grafting it onto yours. But man invented language in order to better deceive, not inform. That state I’m transmitting is often a false one, but you judge it not by the depth of its emotion in my mind, but by the beauty and pungency of the thought in yours. Thus the best deceivers are called articulate, as they make listeners and readers fall in love with the thoughts projected into their heads. It’s the essential step in getting men to write you large checks, women to take off their clothes, and the crowd to read and repeat what you’ve thought. All with mere words: memes of meaning strung together according to grammar and good taste. Astonishing when you think about

Location 673-677

Incidentally, the fastest way you can indicate your level of startup naïveté to a VC (or to anybody in tech), is either by claiming you’re in “stealth”—that is, with an idea so secretly valuable you can’t disclose it—or by forcing someone to sign a nondisclosure agreement before you even discuss it. You may as well tattoo LOSER on your forehead instead, to save everyone the trouble. To quote one Valley sage, if your idea is any good, it won’t get stolen, you’ll have to jam it down people’s throats instead.

Location 684-688

Here’s some startup pedagogy for you: When confronted with any startup idea, ask yourself one simple question: How many miracles have to happen for this to succeed? If the answer is zero, you’re not looking at a startup, you’re just dealing with a regular business like a laundry or a trucking business. All you need is capital and minimal execution, and assuming a two-way market, you’ll make some profit. To be a startup, miracles need to happen. But a precise number of miracles.

Location 696-697

The uniqueness and complete fickleness of such a miracle are what make investing in consumer-facing apps such a lottery. It really is a user-growth roulette wheel with razor-thin odds.

Location 728-732

This formulation of “the X of Y,” where X and Y are two easily understood things, but their intersection is novel or intriguing, was also classic Y Combinator thinking. PG had actually suggested such a trope in his essay on pitching to investors. It’s now terribly cliché, but at one point, like any meme, it enjoyed a period of almost classic establishment acceptance. Even now you hear it occasionally: “the Uber of bicycles”; “the Netflix of men’s underwear”; and so on. It’s a one-line way to instantly make your weird

Location 745-746

“Life is what happens when you’re making other plans.” If you

Location 794-797

tall in bare feet, and towering over me in heels. Most women in the Bay Area are soft and weak, cosseted and naive despite their claims of worldliness, and generally full of shit. They have their self-regarding entitlement feminism, and ceaselessly vaunt their independence, but the reality is, come the epidemic plague or foreign invasion, they’d become precisely the sort of useless baggage you’d trade for a box of shotgun shells or a jerry can of diesel.

Location 882-885

One major unpleasantness involved in writing a memoir is the historian’s task of rereading your personal archive of texts, messages, and emails. In contemplating an earlier version of yourself, you’ll realize that young and glorious you was in fact a total and complete fuckwit. An older you, going back and whispering in your young ear, would issue not praise and encouragement, but insults and dire warnings.

Location 1029-1032

Investors are people with more money than time. Employees are people with more time than money. Entrepreneurs are simply the seductive go-betweens.

Location 1057-1061

For three months, the selected startup founders meet weekly for a dinner with some eminent startup personage. The dinners are not relaxed social affairs: they’re competitive demos in which founders try to one-up each other with increasingly developed products, upward-sloping user graphs, or funding news, within a context of technocamaraderie and shared suffering. The weekly cadence imposes order on the always-full-throttle startup chaos, and is a welcome respite from the grinding toil and stomach-churning stress.

Location 1073-1075

The real YC selling points were the following: access to the YC partners, access to the network of YC founders, a bankable patina of prestige, and Demo Day. The value of the partners will be made clear as this story progresses.

Location 1079-1085

The real network effect, however, was this: With the pool of YC companies expanding every year, you could probably re-create 80 percent of the consumer and infrastructure technology used in our Internet-enabled age exclusively with Y Combinator companies. Whatever need you had, whether system-monitoring tools, mobile development, or even marketing tools (“Have you heard of our product, AdGrok?”), you could find a YC company to fill it. Given that you were one of the family, you could expect excellent service and a steep discount. You could also be sure that whatever you were building would receive preferential adoption by others in the YC network, providing you with an instant set of savvy and patient users, as well as impressive logos to feature in a pitch deck. This tendency to “dogfood” YC products came straight from the top. The day YC funds an airline, I suspect PG will fly that one, and no other.

Location 1093-1097

It doesn’t even need to figure out what a search query is worth and price it accordingly. It simply holds an auction for every search query entered, at the moment the query occurs. The net result is that billions of times a day, Google runs an auction of keywords and accompanying bids. By looking at the bid, and estimating the likelihood of a click, Google takes the product of the two (which is how much it will make per query) and picks the highest. Then it displays the associated ad that the advertiser has created and uploaded to Google for that keyword.

Location 1125-1129

Lung-cancer words, despite their superlative cost, are still a pretty niche market. What are the costliest Google keywords among relatively high-volume keywords? The ranking changes, but the top ten is always composed of some combination of “insurance,” “loans,” “mortgage,” “classes,” “credit,” “lawyer,” and so on. These are Google’s moneymakers, which pay for the Android phones, the Chrome browser, the self-driving cars, the flying Wi-Fi balloons, and whatever weird, geeky,

Location 1132-1135

built on snippets of words and phrases flitting through people’s minds. You as a mere consumer don’t see it, but how Google ranks keywords and runs the auction determines the fate of billion-dollar companies and industries. As the gatekeeper on buying interest, Google is the bouncer at the door of almost any Internet-enabled business today, and if you as the business owner don’t pay your bouncer well, he’ll shut you down, as Google has done more than once.

Location 1178-1187

Most small startups decide to go after the small-to-medium-sized businesses (SMB) market because they think it’s an easy mark. The enterprise sale is too hard, the sales cycle too long, and big companies too untrusting (perhaps reasonably so) of early-stage companies. So they sell to that venerable couple, that mythical bedrock of American values so loved by politicians: Mom and Pop. While it’s true that Mom and Pop are likely to try anything once, and that the quality of their typical software and service would make Italian phone companies look cutting edge by comparison, it is also true that Mom and Pop are phenomenally flaky, and are likely to cancel their subscription to even a useful service, making user turnover a problem. Also, given that you’re making fifty to one hundred dollars per month for every sale, those nickels and dimes mean any sort of high-touch sales process is unscalable even for underpaid startup founders. So you’ve got to scale the sale somehow, either by partnering with someone who already owns the SMB relationship (think Salesforce, or the advertising departments of newspapers), or by building on an existing platform with small-business clients, assuming that’s possible with your technology.

Location 1196-1201

The cofounder relationship goes way beyond the typical professional collegiality one finds in blah corporate life. One can stretch these military analogies too far—nobody is taking incoming artillery fire here, who are we kidding?—but the startup experience does have a certain comrade-in-arms, foxhole quality to it. Nobody believes in what you’re doing except this other poor fool sitting next to you, who’s just as fucked as you are if you don’t succeed. Nothing is keeping the entity going except your shared delusion. And there you sit, working, raging, doing both the best, and also the most poorly thought out, work of your life.

Location 1210-1216

At AdGrok, every major decision became this group decision-making circle jerk, which was fine for minor product tweaks and day-to-day twiddles, but was absolutely fatal for larger questions of strategic vision and corporate culture. If we had had to make some later-stage product pivot—and we likely would have had to were we not sold—there’s no way we’d have gotten agreement among the three of us on it. Such decisions aren’t data-driven collaborative conclusions, driven by spreadsheets and pie charts. No, they’re bold, intuitive, bet-the-company moves decided by one individual, similar to a ship’s captain in a storm, or a Wall Street trader in the midst of a market move. You live and die by such decisions, and they may well be wrong, but it’s more fatal to not make decisions than to make them. Long story

Location 1220-1222

Here’s how it needs to be: either you’ve achieved a certain Vulcan-quality mind-meld with your founders, your brains welded together in the crucible of some formative life experience like the military or hard-won work experience, or there’s one guy running the show (with at least 51 percent of the equity). End of

Location 1274-1278

People go into startups thinking that the technical problems are the challenges. In practice, the technical stuff is easy, unless you’re incompetent or really at the hairy edge of human knowledge—for example, putting a man on Mars. No, every real problem in startups is a people problem, and as such they’re the hardest to solve, as they often don’t have a real solution, much less a ready software fix. Startups are experiments in group psychology. As CEO, you’re both the therapist leader, and the patient most in need of therapy.

Location 1278-1279

people don’t really change, they just become better actors.

Location 1321-1323

The king holds the ring close and reads: THIS TOO SHALL PASS. So remember that, when lamenting your troubles, contemplating the perceived triumphs of peers and competitors, or rejoicing in that rare entrepreneurial triumph. It will all soon pass, and much faster than you think.

Location 1332-1336

Recalling my Goldman days, I imagined everything that was supremely wrong with New York startupwise: the lack of VC, the hustler rather than builder culture, Wall Street drawing away the best talent, the sniggering looks I got when I announced I was leaving Wall Street for a startup. Anyone who thought New York was fertile breeding ground for startups had never lived or worked there. PG was our genius guru, but like many brilliant minds, he occasionally got things flamingly and egregiously wrong, and this was one such instance.

Location 1349-1350

New York tech and life at Goldman Sachs, those would be our first two forays into the corporate-driven, mercenary written word—“content marketing,” to use the hideous marketer’s term

Location 1359-1360

hustling YC founders, and that species of simultaneously frustrated and sanctimonious poseur called a “wantrapreneur.”

Location 1362-1364

Robert Scoble is a mysterious if potent figure on the tech scene. He is one of these old, pasty white guys who seem from a previous generation, if not from the outright technological Jurassic. But via the conferences he attends, the people he knows, and the gadgets he messes around with and reviews, as well as his constant tweeting, he is maniacally embroiled in the Valley ecosystem.

Location 1366-1367

use the term without succumbing to the vapors, he is a tech influencer, a person whose mere tweet could make or break a company.

Location 1403-1405

As a result of the Amazon Web Services near fiasco, plus several more outages and server meltdowns, MRM suggested we run a “chaos monkey” from time to time. This was a software tool created and open-sourced by Netflix, meant to test a product or website’s resiliency against random server failures (such as we’d just witnessed with the blog).

Location 1410-1414

More symbolically, technology entrepreneurs are society’s chaos monkeys, pulling the plug on everything from taxi medallions (Uber) to traditional hotels (Airbnb) to dating (Tinder). One industry after another is simply knocked out via venture-backed entrepreneurial daring and hastily shipped software. Silicon Valley is the zoo where the chaos monkeys are kept, and their numbers only grow in time. With the explosion of venture capital, there is no shortage of bananas to feed them. The question for society is whether it can survive these entrepreneurial chaos monkeys intact, and at what human

Location 1438-1441

had practiced the pitch to the point that I’d be repeating the script in my dreams for years, so go-time was itself uneventful. The AdGrok pitch was simple: Google AdWords represented an immense river of money, much of which was completely undammed or untouched by anybody, as it flowed in its majestic course from the world to Google. We were going to get our hands on a part of it. Even if our take was a small fraction, it would be enough to satisfy the dreams of startup avarice.

Location 1540-1541

Startups are business experiments performed with other people’s money. Here’s how you

Location 1545-1546

debt that is paid off. All early-stage companies, except those that raise an exceptionally

Location 1545-1547

All early-stage companies, except those that raise an exceptionally large round of funding from the get-go, raise on what’s called a convertible note. Despite the fancy name, it’s essentially debt with a nominal interest rate.

Location 1563-1565

As a result, this cap is perceived in essence, if not in contractual reality, to be a proxy for the valuation of the company at the time of the angel’s investment. Early-stage entrepreneurs will bandy about their cap number as if it were a real company valuation, when in truth it’s an input to a hypothetical calculation that may or may not play out in the future.

Location 1566-1568

different beast. With a capped note, there’s no universal agreed-upon value. You can bounce from investor to investor, like a bee from flower to flower gathering pollen, getting notes signed at various caps, and no one need be the wiser. In priced-equity rounds, however, everyone has to agree on a share price and a total amount sold,

Location 1771-1775

One quick way to cut through the shit: ask your pretender-to-influence, “Do you have decision-making power?” If he or she even remotely hesitates or hedges, you’re speaking to a lackey (whether he or she acts like one or not). His or her only utility is to get you to the person who does have that power; everything else is so much pig swill. So route around such people if a real investment check is your goal. Arguably (and this is the canonical YC advice) don’t even accept a meeting from someone who can’t answer yes honestly to the above question. You’re wasting your time.

Location 1785-1790

This stresses the importance of cultivating a network in Silicon Valley. Unlike on Wall Street, where a professional network is conceived more or less when you are, popping out of an affluent uterus in Rye, New York, and setting you on a track of Andover, Yale, Goldman for life (or not), in the Valley things are more fluid and impromptu. Any hustler who can make superficial friends of the California variety and publish a few blog thought pieces, while collecting the echo of confirmatory social media approval, is as much part of the elite as any member of a Harvard Final Club. Of course, you can lose your place just as easily, something an East Coast elite need never fear. But such is the greased pole of Silicon Valley fame and power; anyone can try to ascend, but nothing will arrest your fall.

Location 1825-1827

The lifelong Machiavelli fan in me remembered that memorable line from The Prince: war is never avoided; it’s only postponed to someone’s advantage. I had thought it would be postponed to our advantage, but we’d clearly miscalculated Murthy’s vindictiveness.

Location 1868-1870

Here is a key insight for any startup: You may think yourself a puny midget among giants when you stride out into a marketplace, and suddenly confront such a giant via litigation or direct competition. But the reality is that larger companies often have much more to fear from you than you from them.

Location 1930-1934

So I lied. I diminished the costs of the lawsuit to far below the Undertaker’s projections, meanwhile moving our projected launch date forward to next month to generate revenues immediately, an impossibility given all the changes the boys were making to the product. Then I jacked up the growth rate to an unconscionable amount. It was outright chicanery, cooking the books in the worst form. But it was either that or give up now, and surrender was unthinkable. I still can’t believe the investors believed my numbers, but they did.

Location 1931-1932

I diminished the costs of the lawsuit to far below the Undertaker’s projections, meanwhile moving our projected launch date forward to next month to generate revenues immediately, an impossibility given all the changes the boys were making to the product. Then I jacked up

Location 2037-2040

Fast-forward thirty-five years: Gates now tours Africa as a great philanthropist and single-handedly cures malaria. Kildall eventually succumbed to alcoholism, dying in mysterious circumstances—probably a drunken brawl—in a Monterrey biker bar at age fifty-two. To quote Balzac, “The secret of great fortunes without apparent cause is a forgotten crime, as the crime was properly done.” Never was a crime better concealed.

Location 2052-2055

Steve Jobs was all smoldering ambition, ruthless will to power, and narcissistic amour propre; by all accounts of people who actually worked with him, he was a mediocre engineer who had good taste and knew how to recognize in others talents he didn’t posses, and got them to work like hell for him, while fending off competitors in the meantime. In that sense, he was the absolute exemplar of your successful startup CEO, even if not in the way people commonly think.

Location 2063-2067

The picture I’ve painted of AdGrok—and there’s more sordidness where this came from, not to worry—is not some weird exception; it is the absolute rule. The tech startup scene, for all its pretensions of transparency, principled innovation, and a counterculture renouncement of pressed shirts and staid social convention, is actually a surprisingly reactionary crowd. Its members preen and puff and protect their public image, like a Victorian lady powdering her nose, and refuse to acknowledge anything contrary to their well-marketed exteriors. Sure, it’s no worse than traditional industry or politics, but certainly no better either.

Location 2068-2071

In the case of AdGrok, this meant no subterfuge, no underhanded blow, would be disallowed. In the startup game, there are no real rules, only laws, and weakly enforced ones at that. In the end, success would forgive any sins, as it did for Gates and Jobs, and continues to do for countless startup entrepreneurs. Do we begrudge David the use of his sling, after all, against the towering giant Goliath?

Location 2087-2089

These partnerships, like some potential royal marriage, hinged on a delicate combination of greed, long-term strategy, and a certain mutual seduction. Ugly realities like lawsuits, no matter how petty, dissolved the magic, and would send the more powerful partner fleeing.

Location 2110-2114

Google and a passel of Silicon Valley acquisitions made very rich men of many Valley players. In the self-perpetuating genius of the Valley, that wealth wanted to create more tech wealth, creating an almost embarrassing glut of early-stage angel-investor money. Not only did many of these startup pickers invest their own money, they raised small funds in the $20 million to $40 million range to scale their stakes in budding companies. Angels who used to write $20,000 checks on a deal could now easily write one for $200,000 or more (e.g., our boy Sacca).

Location 2114-2118

the popularity of accelerators like Y Combinator, plus a general acceptance of entrepreneurship as a career, meant lots of very skilled engineers and product people were skipping the corporate trajectory and building exciting products. The emergence of turnkey, on-demand computation like Amazon Web Services, plus off-the-shelf Web-development frameworks like Ruby on Rails, meant that new ideas were easier than ever to test. Many entrepreneurs chose to build shovels rather than dig for gold, creating more complex software building blocks to underpin the innovation, such as back-end services like Parse, accelerating the startup explosion in an almost exponential way.

Location 2126-2128

Heavyweight funds like Mayfield and August knew this, and started doing seed investing, not to own some little piece of a company (they could write small checks all day and still not invest their entire funds) but merely

Location 2128-2136

Which brings us (finally) to our point: in the day-to-day, the lifeblood of a VC wasn’t money, it was deal flow. Getting a first look at a potential Uber or Airbnb is what distinguished a first-class VC from an also-ran. Given Y Combinator’s immense success in drawing the best entrepreneurs, it had a quasi-stranglehold on the best early-stage deal flow in the Valley. And since early-stage deal flow today translated into later-stage deal flow tomorrow, via the follow-on investing phenomena described, Y Combinator was the gatekeeper to the best present and future deals in the Valley. Like control of the water supply in some arid agricultural region, whoever had the most upstream control of the water sluice controlled everything else—which is what Y Combinator’s Demo Day represented. Thus, powerful and haughty VCs who wanted to attend Y Combinator’s showcase pitch event had to kneel and kowtow to a sandal-wearing bear of a man with a distaste for bullshit and a flair for the written word. That man was Paul Graham, without question the canniest tech investor in human history. And it was to Paul Graham we first turned with our existential problem in those desperate

Location 2128-2136

Which brings us (finally) to our point: in the day-to-day, the lifeblood of a VC wasn’t money, it was deal flow. Getting a first look at a potential Uber or Airbnb is what distinguished a first-class VC from an also-ran. Given Y Combinator’s immense success in drawing the best entrepreneurs, it had a quasi-stranglehold on the best early-stage deal flow in the Valley. And since early-stage deal flow today translated into later-stage deal flow tomorrow, via the follow-on investing phenomena described, Y Combinator was the gatekeeper to the best present and future deals in the Valley. Like control of the water supply in some arid agricultural region, whoever had the most upstream control of the water sluice controlled everything else—which is what Y Combinator’s Demo Day represented. Thus, powerful and haughty VCs who wanted to attend Y Combinator’s showcase pitch event had to kneel and kowtow to a sandal-wearing bear of a man with a distaste for bullshit and a flair for the written word. That man was Paul Graham, without question the canniest tech investor in human history. And it was to Paul Graham we first turned with our existential problem in those desperate

Location 2140-2142

We needn’t have worried. Even if we were the runt bastard child of the family, Papa PG brought all his protectiveness to bear. In a world where superficial relationships and glib bullshit are the rule, there’s nothing more fearsome than to see true loyalty at work. PG mustered YC’s full network for our cause.

Location 2144-2147

Y Combinator was the sort of unforgiving power player that remembered the names of investors who had crossed portfolio companies in the past, or who had disseminated unflattering portraits of YC, and blacklisted them from any YC dealings, or from the minds of YC founders. This was done with little thought of the size of their funds or their influence in the Valley, leaving more than one self-important VC sputtering at a dressing-down, or left out of a round because the company’s founders had gotten a warning from PG.

Location 2147-2148

The harsh reality is this: to have influence in the world, you need to be willing and able to reward your friends and punish your enemies.

Location 2179-2181

But in the passive-aggressive popularity contest that is Silicon Valley, someone actually going to bat for you—really going to bat, like telling important people to go fuck themselves—that’s rarer and more short-lived than a snowflake in a bonfire.

Location 2200-2204

I’m high-strung, fast-talking, and wired on a combination of caffeine, fear, and greed at all times. But “Sama,” as he is known on Hacker News and Twitter, really takes the cake. After an hour with him, I was looking for the closest beer bar. Standing maybe five-seven, lean and wiry, with perpetually hunched shoulders, he has clear blue eyes of an unusual intensity. A typical meeting would involve him conducting a conversation about A, with a side tangent on topic B, while considering C, and simultaneously texting on his phone and scanning his laptop screen. The topics veered from detailed wisdom about fund-raising

Location 2209-2210

my desperation to find an in with Microsoft, I had stalked people on LinkedIn like some recruiter trying to poach a hire, searching for an intersection point with Microsoft. Sama was one of those highly connected nodes that pepper the Valley; you know him, and you’re

Location 2239-2247

What’s it take to do startups? It certainly isn’t intelligence. I was in the lower third of my PhD class in physics at Berkeley, and I had to take my preliminary exams three times before I passed. Most of the founders I know are certainly crafty and quick-witted, but compared with some of the certified geniuses I met in academia, they aren’t going to win any Fields Medals or Nobels. It certainly isn’t technical skill. I’m a crappy programmer, and I can hack crude prototypes of finished products at best. Some founders are technical virtuosos, but I suspect most weren’t the top students in their respective computer science classes (assuming they even had a formal education). It’s not unique product or market vision. Anyone who used Google’s ads-buying tool for all of five minutes, and then registered that that piece of shit was a $70 billion per year moneymaker, could see the need for AdGrok. Some startup ideas were visionary, like Airbnb, but many, like Dropbox, were merely extremely well executed versions of existing technologies.

Location 2248-2249

distinguish successful startup founders at whatever level of the game, from the forgettably minuscule (e.g., AdGrok) to the epoch changing (e.g., SpaceX).

Location 2249-2250

First, the ability to monomaniacally and obsessively focus on one thing and one thing only, at the expense of everything else in life. I lived, breathed, and shat AdGrok. Thanks

Location 2256-2261

Second, the ability to take and endure endless amounts of shit. I was raised under the sadistic care of a sister ten years my senior who delighted in unleashing endless taunts and abuses. My father was a domineering man, in every way. I spent years in an all-boys Catholic school filled with brutish bullies and cold, aloof priests. We did nothing but beat each other up and jerk off. I scraped by on nothing for six years as a penniless grad student in an expensive city, and then survived three years on Wall Street’s most competitive trading floor during its biggest market catastrophe. Long story short: within the limited purview of white-collar travails, there’s nothing I can’t take for an extended period of time.

Location 2262-2264

Sure, there are times when an entrepreneur’s life will make you feel like you’ve got the world by the tail. You just talked the biggest check you’ve ever seen out of a respected investor, and boom! there the amount appears, as big as life, in your previously barren bank account.

Location 2265-2269

But those moments are far outnumbered by the countless moments of gnawing doubt, nauseating anxiety, and endless toil. If you’ve got a great relationship with your cofounders, then that camaraderie can maintain you. In fact, as in sports or war, the overpowering desire to not let down your fellows is often all that keeps you going. But if you have no cofounders, or if that band-of-brothers relationship isn’t there, then the only thing keeping the whole construct going is sheer, stubborn bloody-mindedness. You get up every morning, get kicked in the face, and come back for more the next day.

Location 2273-2275

Incidentally, it helps to have enemies. While love is a beautiful emotion, far more empires have been built, books written, wrongs righted, fights won, and ambitions realized out of vengeful desire to prove some critic wrong, or existential dread of some perceived enemy, than all the love in the world. Love is grand, but hate and fear last longer.

Location 2291-2293

Murthy passed on to the oblivion he deserved. The only chapter in Silicon Valley history he’ll ever have is this one, the one I’ve written for him. As for AdGrok and the boys, it was time to think about a real launch.

Location 2326-2331

My first thought was to not tell the boys; they’d get more stressed-out about it than I was, and I needed them alert and productive for our looming product launch. My second thought was, well, so what? I was already neck-deep in risk; our return profile (to use Wall Street quant-speak) was absolutely binary: either we succeeded dramatically, with a million-dollar outcome, or we ignominiously failed. What was one more child in the mix? I’d have the heir and the spare. A little redundancy, whether in technology or heredity, never hurt. Launching!

Location 2326-2330

My first thought was to not tell the boys; they’d get more stressed-out about it than I was, and I needed them alert and productive for our looming product launch. My second thought was, well, so what? I was already neck-deep in risk; our return profile (to use Wall Street quant-speak) was absolutely binary: either we succeeded dramatically, with a million-dollar outcome, or we ignominiously failed. What was one more child in the mix? I’d have the heir and the spare. A little redundancy, whether in technology or heredity, never hurt.

Location 2342-2345

To paraphrase Jon Bond, an agency guy we had pitched: marketing was like sex; only the losers paid for it. Like losers, we paid for it in this case, hiring a PR agency to juice the metaphorical printing presses for us. The result was a series of endless interviews in such august publications as Internet Retailer and Direct Marketing News.

Location 2343-2343

pitched: marketing was like sex; only the losers paid for it. Like losers, we paid for it in this case, hiring a PR agency

Location 2372-2378

However, such fairy tales reveal themselves to be the fantasies they are following a launch. You’ve spent months and gambled your career on a product, and then after a bit of excitement, you realize what a misshapen and misbegotten piece of shit it really is. This is the crisis that kills 90 percent or more of startups that manage to survive the initial plagues of founder disagreements, failure to ship code, and failure to raise money. This nearly uncrossable chasm is behind all those “What happened?” posts on TechCrunch or other tech rags that crop up when some seemingly unstoppable startup suddenly announces it is ceasing operations and giving back what’s left of the investors’ money. Its founders and management could not negotiate the trough of despair, and like some lost British adventurer trying to cross a forbidding desert, the company basically gave up and died.

Location 2382-2383

What’s the solution? The trendy answer is that you need to find what’s called

Location 2382-2384

What’s the solution? The trendy answer is that you need to find what’s called “product-market fit,” which is a fancy way of saying that you need to build a thing people are willing to pay for.

Location 2384-2388

turns out that’s pretty hard, because you don’t know what people will pay for until you ask them to, and if what you’re building is truly novel, there’s no history for you to go on. Fortunately, the iteration cycle in a software business is fast (we don’t need to retool milling machines here), and from some approximately correct starting point we can converge on a final product, almost like successive guesses at the solution to an equation. The quicker we iterate, the more steps we can take in the direction of this mythical point of perfect

Location 2389-2392

If, however, we get close enough to that perfect product-market fit, in which users pay more money for the product than it costs to operate (and than it cost to acquire those users to begin with), we’ve escaped the financial free fall. Recall our cliché about startups being like building a plane after jumping off a cliff; getting to a positive “run rate,” as it’s called, is like finally getting the wings on, firing the engine, and watching the contraption actually maintain or gain altitude (no matter how badly built it may

Location 2395-2397

The combination of these two elements, enchantment and surrender, is then essential to this love we are discussing. Their combination is not a mere coexistence; they’re not two parts placed together, one next to the other, but rather one is born and nourished from the other. It’s love due to surrender via enchantment.

Location 2409-2412

Read between the lines of their idealized and cosmetically enhanced CVs: less than a year at a company means they didn’t vest their first year of equity, which means either they or the company sucked, possibly both. Leaving at precisely four years means they’re plodders, punching the equivalent of the white-collar Silicon Valley clock until the next shift.

Location 2412-2414

Which friends or connections do you have in common? Do they know your dipshit douche friends you maintain for career reasons, or the ones you actually respect? As true now as in Machiavelli’s time, men are judged by the company they

Location 2414-2420

Do they have that “lean and hungry look” of Cassius in Julius Caesar, and eat at dingy taquerias, or perhaps live off cheap staples like ramen or food substitutes? Those are the truly dangerous ones, the ones who live like organized crime trigger men, guerrilla fighters, or sailors at sea—eating shitty food and living in cheap, dumpy crash pads—and who couldn’t give a damn about quality of life.* Those are the people to fear, because they don’t need anything an antagonist can deprive them of. Or do they have a score of check-ins at Benu, Saison, and Quince? If they’re not financially independent, then they are harmless tools; they’ll do anything to keep the parade of fungi terrine monkfish and tangerine-peel abalone coming.

Location 2423-2427

Total commitment, like unconditional love, is the only type that matters. The bike-riding, date-night-going types will never give everything to a company or an idea, and are nothing more than complacent bourgeois, whatever trappings of the “disruptive innovator” they may sport, often in the form of ponderous blog posts or a bookshelf of bound B-school-level bloviation. The ones who could pass for a homeless person, though, those are the startup kamikazes who will give everything for the entrepreneurial cause, and are stopped only by death or

Location 2424-2427

only type that matters. The bike-riding, date-night-going types will never give everything to a company or an idea, and are nothing more than complacent bourgeois, whatever trappings of the “disruptive innovator” they may sport, often in the form of ponderous blog posts or a bookshelf of bound B-school-level bloviation. The ones who could pass for a homeless person, though, those are the startup kamikazes who will give everything for the entrepreneurial cause, and are stopped only by death or jail.

Location 2427-2431

Are they what passes for the American ruling elite? Do they hail from the urban archipelago of American privilege: Chevy Chase, Maryland; Winnetka, Illinois; Tiburon, California; Scarsdale, New York; and so forth? Or are they from Visalia, California; or Chimacum, Washington; and did they go to a high school named after an astronaut or a president? Did they go to a high school with either “Day” or “Prep” in the name, or with the titular formula “The——School,” where the blank is something majestic and/or Anglo-Saxon sounding? Then they’re playing the slow-and-steady, long game,

Location 2432-2435

reach the coddled precincts of the tech elite? Did they trudge along on the upper-class tour bus from the Ivy League to consulting and/or finance and burn out, or did they fight their way out of some backwater, earning their seat at the table based only on chutzpah and/or blitzkrieg-style success? The latter are fearsome, the former . . . not so much.

Location 2436-2437

Know what they want—whether money, power, social approval of various flavors, or merely a comfortable life—and you can predict 90 percent of what they eventually do.

Location 2468-2469

In a culture in which people show up in chicken suits to formal meetings, security is one of the few things the self-consciously irreverent tech company takes seriously.

Location 2515-2519

As I hope I’ve made clear in many disclaimers throughout this work, there was nothing badass about my career in technology. The scant success I had was due purely to happenstance, combined with being a ruthless little shit when it counted. No, I mention this to underline how the snowballing interest of a larger company in a smaller one resembles the intoxicating enchantment one feels for a new romantic interest, house, or car. The target of your interest lays out his or her wares and tries to beguile, but past a certain point, it’s you

Location 2530-2531

For example, the Twitter API allows someone to build a tool that gathers all your tweets and presents their engagement data (e.g., retweets, favorites) in an elegant and useful way. Effectively, it’s the way computers talk to each other when they are owned by different companies.

Location 2545-2548

use the word “offer” lightly; companies will mention the possibility, but often the entire process is just an excuse for dissecting a company on the founders’ time, the market intelligence a big company does to remain relevant, as well as keep from getting blindsided by an upstart. He also advised me privately to shield the boys from any acquisition chatter.

Location 2577-2577

10 Tripel, sucked down while gloomily reading Michel

Location 2587-2588

SPOILER ALERT! Do not read if you like suspense, Dostoyevsky, and/or incremental character development.

Location 2590-2591

I squelch them from my feed—I don’t need to see the weekly puppy or child pictures—but I tune in every couple of years when my mind trips over some stale memory. Who did that beauty you crushed on finally marry? Did the pretentious kid become the smarmy yuppie you thought he would?

Location 2597-2600

What a company builds (SVP, Product), how it builds it (SVP, Engineering), how that eventual product is operationally run (COO), and what other companies it buys (Corp Dev): those are the core functions of any large tech company, and the people from the Ads team we met during forty-eight busy hours in 2011 would, by 2015, be that core leadership of Twitter.

Location 2600-2604

Often at a tech company with an advertising business model bolted on to an otherwise user-facing product, the Ads team will be the weakest team in the company. Ads is perceived as a necessary evil, and marketing ain’t cool, so the hotshot twenty-two-year-old kids who are the recruiting cannon fodder don’t join up. The visionary CEO doesn’t care about money, only the user experience, and manages by looking at usage dashboards, not revenue ones. Bold product initiatives that will rekindle always-fickle app usage are green-lit, and their resources are allocated by CEO fiat.

Location 2602-2604

who are the recruiting cannon fodder don’t join up. The visionary CEO doesn’t care about money, only the user experience, and manages by looking at usage dashboards, not revenue ones. Bold product initiatives that

Location 2606-2609

But Twitter was different. The Ads team was actually the most dynamic, ready-to-ship, and bustling team in the organization. While the core Twitter product hasn’t changed in years, the Ads team had been crackling with new products shipped regularly, one step either behind or ahead of rivals like Facebook, but always churning and burning. They also convinced the company’s management to make ambitious (and stunningly large) acquisitions in the ads-technology space.

Location 2621-2623

founders of Twitter’s first acquisition, a geolocation technology startup named Mixer Labs. One founder was Elad Gil, a YC investor and noted blogger, whose eloquent and knowledgeable takes on the early-stage startup game I routinely devoured. The other was Othman Laraki, a former Google product manager who was then a member of the Twitter Growth team.

Location 2626-2627

After an effusive phone call, he arranged for his junior partner Harold Yu to give us a legal checkup.

Location 2633-2637

Here’s another lesson for the aspiring startupista: don’t skimp on lawyers. An error at the hour of signing a big contract, or negotiating an acquisition, could easily cost you millions, or be the deciding factor between summers in Ibiza with your model girlfriend or taking a consolation-prize job as product manager at Oracle instead (look, you get pretax commuter cost benefits there!). Get the best legal guns you can pay for, and if you can’t pay for it, convince them to accept something other than money in exchange. Lawyers didn’t get into law because they’re good at business; bamboozle them however so long as they get cracking.

Location 2645-2647

In such a world, European states and the great Middle Eastern powers of the Ottoman and Persian empires dealt with each other mostly as equals, with all the diplomatic frippery and double-dealing this required. A key actor in this fascinating civilizational face-off was the dragoman

Location 2654-2655

So you see we have two types of mistranslation at work: the intentional kind that serves as diplomatic lubricant to get a deal done, and the more serious kind, which leaves each side thinking it agreed to a different thing.

Location 2657-2659

There are always two channels of communication at work in a deal: One, a formal one, typically features email and attached documents, from founders to a company’s corp dev team, as well as a possibly relevant product team. The second is the informal and clandestine. For lack of a better word, let’s call it scheming.

Location 2658-2659

work in a deal: One, a formal one, typically features email and attached documents, from founders to a company’s corp dev team, as well as a possibly relevant product team. The second is the informal and clandestine.

Location 2660-2663

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings are generally inadmissible in court, save for criminal trials with a warrant. So when the scheming begins, a startup world that does 99.9 percent of its communication via the asynchronous channels of email, messaging, and social media suddenly goes old school, and every communication is conducted via hushed phone call in a closed-door conference room.

Location 2660-2661

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings

Location 2660-2662

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings are generally inadmissible in court, save for criminal trials with a warrant. So when the scheming begins, a startup world that does 99.9 percent of its communication via the asynchronous channels

Location 2660-2663

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings are generally inadmissible in court, save for criminal trials with a warrant. So when the scheming begins, a startup world that does 99.9 percent of its communication via the asynchronous channels of email, messaging, and social media suddenly goes old school, and every communication is conducted

Location 2660-2663

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings are generally inadmissible in court, save for criminal trials with a warrant. So when the scheming begins, a startup world that does 99.9 percent of its communication via the asynchronous channels of email, messaging, and social media suddenly goes old school, and every communication is conducted

Location 2660-2663

scheming happens either by phone or in person, with no emails or messaging involved. As a legal note, it’s against the law to record phone calls in California, and such recordings are generally inadmissible in court, save for criminal trials with a warrant. So when the scheming begins, a startup world that does 99.9 percent of its communication via the asynchronous channels of email, messaging, and social media suddenly goes old school, and every communication is conducted

Location 2687-2690

Sadly, the startup game isn’t played according to the Marquess of Queensberry’s rules. Startup entrepreneur, you are David before Goliath, the nascent state of Israel during the ’48 war, Crockett at the Alamo, the Spartans at Thermopylae. Choose your favorite metaphor of hopelessly mismatched odds: that’s you. By any rational reckoning, startups should be dead before they even start. So if some disaffected insider at the acquiring company accosts you in a bar and starts spilling the beans, you buy him another beer and lean in close. You’ll need every advantage you can get.

Location 2695-2695

vittles. Whether you approached Epicenter from east or

Location 2695-2695

or west

Location 2702-2704

The entire SF startup world, in case you haven’t been keeping track of the geography, lives between First and Eighth Streets, and between King and Market Streets, in SF’s SoMa. A former bombed-out industrial space, dotted with a few flophouses and taken over by addicts until the late nineties, it was the eight-by-eight-block playing field of global technology.

Location 2712-2714

This was the major-league, serious shit, take-no-prisoners championship of tech entrepreneurship, and if you were going to play, you’d better show up ready to bite the ass off of a bear.

Location 2716-2718

Poker’s the only game fit for a grown man. Then your hand is against every man and every man’s hand is against yours. Team-work? Who ever made a fortune by team-work? There’s only one way to make a fortune and that’s to down the fellow who’s up against you.

Location 2781-2784

So $5 million for three hires plus intellectual property Twitter might use, with a premium for YC bling and “thought leadership” evinced via our glorious blog, was way too cheap. We hadn’t risked everything from our finances to our sanity for just over a million each that would take four years to earn.

Location 2795-2797

Russ, who was investing his own money, and who was essentially playing a market in people like a savvy gambler at a poker hall, this was a fine return. He’d take his cash (or shares in Twitter, if that’s what it came to), possibly sell them, and then invest in other companies, upping his returns, and suffering occasional burns as well.

Location 2798-2800

Which is why Sacca, or any VC really, would push you to mimic his own distorted risk profile, which prefers a tenfold return, even if it marginally increases the chance of complete failure. Twofold just ain’t cool enough for the limited partners, and not why they handed over their money to Sacca in the first place.

Location 2800-2802

risk expectations of founders and investors can often be severely misaligned. I hate sports analogies, but here’s one to explain this vital point: most VCs are playing a version of baseball in which the only way to score is to hit a home

Location 2807-2811

The fund’s total profit will be calculated from whatever those initial bets return. Unlike, say, a hedge fund portfolio manager, who rolls the winnings from one good bet into the next, compounding a series of returns into something truly huge, VCs do not take liquidity from one company’s exit and pour it into yet another’s.* This, at heart, is why the go-big-or-go-home strategy makes the Silicon Valley world turn, and why entrepreneurs push themselves to be either the next Airbnb, or nothing.

Location 2811-2814

long-term business of recurring revenue but relatively slow growth is dismissed as running a mere “lifestyle business,” which is a dirty word among VCs. Of course, the entrepreneurs are quite happy to run a revenue-generating concern that spits out cash as low-tax dividends, and dedicate their lives to skiing or guitar playing or whatever. But their investors will hate them for it, and the entrepreneurs will suffer a loss of social capital as a

Location 2869-2874

Amin who was surely poached, leaves a company for a rival, he leaves behind a stench of bad mojo. Everyone on his former team is on guard, and is unlikely, at least for the first few weeks, to engage in the sort of informal information sharing that often binds people across jobs and companies. The net of it was that Amin would have no back channel into Google, and would essentially be blind to what it was really doing (or not doing), as nobody there would be on speaking terms with him. This left him open to a barefaced bluff from a company, like us, that had built its entire product on Google. That one throwaway line of mine was probably half responsible for what followed.

Location 2914-2917

The fact that you come bundled just changes the economics. We used to have unions that would give workers a magical thing called collective bargaining. Now being part of a hot startup is your union, and the only dues required are your entire life for the time you’re in the startup. Welcome to our new collectivism, tovarish. By bundling the talent, though, you command a premium due to mere leverage—don’t like the price, we’re all off the table.

Location 2922-2926

As with any interview, whether it be a hardball technical/engineering interview, or the foofy product manager one, the challenge is very simple: either you’re incompetent and you struggle to provide even one answer you hope is right to the interview question, or you’re not, and your brain produces two or three. Each of those could be right, depending on your interviewer. Truth in the world resides only in mathematical proofs and physics labs. Everywhere else, it’s really a matter of opinion, and if it manages to become group opinion, it’s undeservedly crowned as capital-T Truth.

Location 2927-2928

And that you do by reading body language, and understanding the intellectual sauce your opponent has been marinating in (if any). With that in mind, you choose among the three alternative answers you could spout.

Location 2929-2935

These interviews serve another, more subtle purpose. If you’re offered a job that you accept, then that day spent with a slate of tormentors is in fact an initiatory hazing ritual, like the collective beating doled out to an aspiring gang member. It’s weird, but you develop a certain attachment to that experience, and it’s a jump start on the team bonding you’ll need to succeed on the eventual job. An interviewer will always remember that he or she helped bring in someone who eventually became a valued employee. Companies also stress about the experience, and try to find the subtle balance of challenge and courtship that will attract the best talent. Many and sundry are the postinterview blog posts dissecting one or another idiosyncrasy of flagship companies like Apple, Google, or Facebook, and whether they’re fair, cruel, or just random. In many ways, the interview process is the face of the company.

Location 2942-2945

His angle was analytical ability, and he asked me a variant of that legendary Enrico Fermi brainteaser about piano tuners in Chicago. His variant was to estimate the number of planes in the sky at any given moment. It required nothing more than some rough base assumptions about number of airports and flights per day, and then some dimensional analysis that got us within an order of magnitude of reality, and we were done in ten minutes.

Location 2947-2951

Doing advanced studies in any quantitative field is like surviving Marine boot camp while the rest of the world is channel surfing and inhaling Oreos; you don’t exactly need to fear the push-up test. Even when the intellectual test is being doled out as filter to the elite ranks of a globe-spanning tech company, you won’t be out of your league if you understand the problem space. So if nothing else, aspiring physicist or mathematician, rest comfortable knowing you’ll come out of that long academic tunnel thinking circles around most people.

Location 2957-2960

As a random intellectual test, he next asked me to explain how pairing apps that allow two phones to talk to each other when shaken simultaneously work. He had seen physics on the résumé and wanted to kick the intellectual tires. As always, how you sell yourself is how you’ll be bought. This one was easy, and as with Rohit, the interview ended early.*

Location 2971-2973

As the waterboarding continued, I began to notice an air of arrogance that stank like bad aftershave. I’d later realize that it was his Friend of Zuck (FoZ) status I was smelling, and to which I evidently was genetically allergic. Some allergies, it turns out, worsen on exposure.

Location 2973-2974

mentally agile, and asked the standard questions about the Gospel of Startup,

Location 2978-2984

concepts in Silicon Valley, namely that of cultural fit. Cultural fit, like the Holy Spirit in the Catholic Trinity, is that mysterious intangible, hard enough to conceive of, much less define, but critical to a job at a tech company. In theory, it’s a measure of the alignment of the candidate’s values regarding collaboration, style of product development, and overarching goals (“a more open and connected world”) with those of the company. Since, in the self-aggrandizing view of most tech companies, their corporate “culture” is unique, and as valuable as that of an uncontacted Amazon tribe, it is hailed as precisely what undergirds their sky-high valuation. Therefore, a candidate’s alignment with that culture is overwhelmingly important. Being handy with the C++ programming language isn’t good enough—you also have to be one of us in thought and spirit.

Location 3022-3025

There, we compared notes. MRM and Argyris weren’t exactly rousing in their reviews of the experience. In fact, it was clear that the fascist vibe the company gave off had very much rubbed them the wrong way. They had never really liked Facebook, as either product or company, going back to our visits to their developer events. The daylong hazing had done nothing to charm them.

Location 3037-3038

In the Aaron Sorkin cinematic adaptation of this story, this is where the violins will start scraping their tension-building wail.

Location 3080-3085

Sitting unused in a closet. Thanks to Zynga, Facebook, and other companies, old Cubans like my mother were too busy playing social games like FarmVille to gather around the table and spin the tiles that had been so painstakingly smuggled. Too busy clicking to buy $0.99 pink tractors and $1.99 spotted digital cows. Facebook got Cuban old ladies to play computer games! And pay for it! Think of that miracle for a moment. And it wasn’t just Cuban old ladies.

Location 3080-3083

Sitting unused in a closet. Thanks to Zynga, Facebook, and other companies, old Cubans like my mother were too busy playing social games like FarmVille to gather around the table and spin the tiles that had been so painstakingly smuggled. Too busy clicking to buy $0.99 pink tractors and $1.99 spotted digital

Location 3098-3101

Here’s another truth about tech life: anyone who claims the Valley is meritocratic is someone who has profited vastly from it via nonmeritocratic means like happenstance, membership in a privileged cohort, or some concealed act of absolute skulduggery. Since fortune had never been on my side, and I had no privileged cohort to fall back on, skulduggery it would have to be.

Location 3120-3125

As I would come to learn, my situation wasn’t unusual, though not generally talked about. Companies with acquisition wherewithal and the nerve to use it bid for what they wanted in deals. You came in with your team and your product; they gave it the once-over, and said, “We want person A and B, but not C, and we don’t care about the tech.” They then offered you a lump sum for what they wanted, and you were left to double-deal, buy out, or otherwise fuck over whoever in order to get the deal done. The company—and places like Facebook and Google did this commonly—cared only about net price per engineer (or product person), not the absolute cost. And they certainly didn’t care what investors got. Many an early-stage acquisition unfolded in this

Location 3131-3133

The second reason I took his advice-by-example is that I liked Mick, and in a city full of cranky, asocial, self-absorbed, narcissistic startup founders, he came off as a real guy, a mate one could trust. He stood to gain nothing from my deal, and was helping me only as a way to pay it forward to a fellow startup founder in a bind.

Location 3135-3137

Long after this AdGrok drama, I’d hear an East Coast tech guy perfectly sum up the reigning attitude in Left Coast tech: “It’s like they have no memory there. It’s the Land of the Stateless Machines.”

Location 3137-3138

A bit of context: “State” is a technical term referring to data kept in memory that a program or function needs

Location 3142-3149

As every new arrival in California comes to learn, that superficially sunny “Hi!” they get from everybody is really, “Fuck you, I don’t care.” It cuts both ways, though. They won’t hold it against you if you’re a no-show at their wedding, and they’ll step right over a homeless person on their way to a mindfulness yoga class. It’s a society in which all men and women live in their own self-contained bubble, unattached to traditional anchors like family or religion, and largely unperturbed by outside social forces like income inequality or the Syrian Civil War. “Take it light, man” elevated to life philosophy. Ultimately, the Valley attitude is an empowered anomie turbocharged by selfishness, respecting some nominal “feel-good” principals of progress or collective technological striving, but in truth pursuing a continual self-development refracted through the capitalist prism: hippies with a capitalization table and a vesting schedule. What

Location 3298-3303

Even after winning one race, there was always another, and always somebody with a faster car, wasn’t there? This old guy knew that, which is why he was racing reckless young jackasses like me. He probably had a wife, kids, property, and all that bougie shit, and yet the moment he saw a challenger in his rearview, he tossed that aside for a bit of the old screeching rubber. The fact he was risking all that he had achieved by committing felonious acts of reckless driving, reckless endangerment, and “exhibition of speed” (to use the California penal code’s name for it) was immaterial. He didn’t get to that Porsche by turning down all-in challenges. Neither will you, gentle reader.

Location 3317-3319

As I was soon to learn, this is a common tactic. You pay a pittance for the company, engineering it such that investors get little, and then pack the real value into the hiring offers for the employees. TechCrunch would carry the news that the company sold for X million dollars, but technically it would sell for 10 percent of X, while the rest went into fat signing bonuses and heavily laden vesting schedules for the founders. With no

Location 3325-3328

The acquiring company doesn’t care less how money is divided between investors and employees. As we’ve reviewed, what they care about is price per high-value person (that is, engineers and product managers). If that works out, and the form of payment is whatever admixture of cash versus equity they prefer (or are willing to put up with), the deal looks fine to them. Every acquiring company will have such a target price per person in mind when you seriously discuss a deal. Your job as deal negotiator is to get as close to that as humanly possible.

Location 3334-3335

To give you an idea of just how indifferent the acquiring company is to investors, keep in mind that Sacca was reputedly Twitter’s

Location 3393-3396

Back in the office, the boys made a show of going back to work on code, but their minds were elsewhere. I had just pulled a dicey ploy that presented only downside to them, mostly because of my bloody-minded insistence on going to Facebook. The same bloody-mindedness that had raised us money and defended us from legal enemies might now blow up the entire construct we had slaved to

Location 3458-3462

At the end of the day, AdGrok was simply a long, stressful job interview for Facebook (and ditto for the boys at Twitter). We all claim we “sold” AdGrok, but in reality, AdGrok was merely leverage to score the job offers that actually made us the real financial upside, job offers we would not have been able to score otherwise. The corporate-development teams of large companies, insofar as their small-company deals are concerned, are really glorified HR recruiters with fatter checkbooks. That’s another little detail the self-glorifying founders of acquired companies often fail to mention.

Location 3464-3465

Morality, such as it exists in the tech whorehouse, is an expensive hobby indeed.

Location 3486-3490

Cox was handsome in the way of a Gosling or Depp: a tempered masculinity encased in a cuddly package, custom-made for female desire. It was a recurring internal joke at Facebook to point out the Twitter storm of oohing and ahhing whenever he took the stage at a Facebook PR event. He had the gift of the gab, which he used to great effect, weaving a seductive narrative around Facebook and the future of media. As the first speaker, he was clearly there to instill the big-picture vision of what we had been selected to help build.

Location 3495-3499

He then embarked on a common trope among Valley types, framing a product in some historical continuum of prior technologies, the product currently discussed being the ultimate and inevitable final chapter in the triumphant procession. Radio and TV were depersonalized media of mass consumption, revolutionary for their time, but ultimately lacking. Steadily more focused and fragmented media—topical magazines like Car and Driver, your local newspaper with pull-out sections for your neighborhood—continued this trend of increasing personalization. Facebook, however, was the true teleological end goal of modern media.

Location 3504-3506

Here was the first taste for the new Facebook employee of a world interpreted not through traditional institutions like newspapers, books, or even governments or religions, but through the graph of personal relations. You and your friends would redefine celebrity, social worth, and what should be churning through that restless primate brain all day.

Location 3507-3509

Andy Warhol was wrong. In the future, we wouldn’t all be famous for fifteen minutes; we’d be famous 247 to fifteen people. That was the new paradigm, even if the outside world didn’t realize it yet. Facebook employees—we few, we happy few—knew what world was coming, and we’d help create

Location 3510-3512

day. I suspected this was a recurring biweekly event for Cox, the rousing speech to the new acolytes. Just the regular speech for king and country, which he had down to the point of studied and flawless spontaneity. Facebook certainly didn’t skimp on putting on a good show.

Location 3519-3520

As I’d later learn, weirdly pointless versions of them would be held in the regional offices where no engineers even worked, as a sort of pagan celebration of the values of do-it-yourself creation, total commitment to the company, and disruptive innovation.

Location 3527-3533

Here’s where the genius in the on-boarding, and more broadly in Facebook, really lay. People joined Facebook, and like immigrants at Ellis Island, left their old, dated cultures behind, replacing them with an all-consuming new one. The on-boarding experience was designed precisely as the sort of citizenship oath that new Americans took in front of a flag and a public official. It was almost religious, and taken absolutely sincerely and at face value. Even in a culture brimming with irreverent disdain, I never heard anyone utter a word of cynical trollery about Facebook and its values, either at on-boarding or during my years of work there. As with Americans and “our troops,” motherhood, and the Constitution, certain things were enshrined, and nobody dared ridicule them.

Location 3533-3534

In a posthistorical developed world devoid of transcendent values, whose pantheons look like North Korean grocery stores, bare shelves empty of any gods or heroes, this corporate fascism was intoxicating.

Location 3538-3543

Cynicism is the last refuge of the shiftless. I don’t cite this absolutist tendency for the cheap sardonic joke, the asshole hipster who’s too cool for school, but also too cool to believe in anything. No, I cite it because I was as seduced as the next guy sitting there in Pong, perhaps even more. The human need for immortality projects—those ends that dole out meaning and purpose beyond ourselves—hasn’t changed since the pyramids. The only difference now is the nature of the putative Holy Land, and the means for achieving it. After Cox’s rhetoric, and Pedram’s grim injunctions, we broke for a breather.

Location 3577-3598

to discuss the sensitive stuff. Lesson one from Officer HR was one of Facebook’s ever-present obsessions: secrecy. Like Jesus speaking to his apostles, Facebook often imparted nuggets of its culture in the form of parables. The parable here concerned a misguided Facebook employee who leaked news of a soon-to-be-launched product to the tech press. Zuck reacted via a to-all email with the subject line “Please resign,” an alarming presence in anybody’s inbox. The email, which was projected onto the screen in Pong and read line by line, encouraged whoever had leaked to resign immediately, and excoriated the perpetrator for his or her base moral nature, highlighting how he or she had betrayed the team. The moral to this story, a parable of the prodigal son but with an unforgiving father, was clear: fuck with Facebook and security guards would be hustling you out the door like a rowdy drunk at the late-night Taco Bell. Lesson imparted, time for that close second in Facebook priority: discretion. As the curator of the largest collection of personal data outside of the NSA, Facebook was ripe for unprincipled internal abuse. Not only was such abuse unethical, but the PR hit from a story getting out about a jealous employee who stalked his wife, or immature interns checking out a celebrity’s messages, would be monumental and hugely embarrassing. As it was, people were wary of this druglike service called Facebook, with which they shared their most intimate human experiences, but also subconsciously resented and feared. Anything less than the strictest discretion would threaten to revoke the tenuous pass that hundreds of millions of users had granted that dark-blue box to mediate their lives. I would personally know at least one person who’d get bitten with this, and be terminated after “the Sec,” as internal Facebook security was called, found out they were looking at profiles without an official reason. It was that simple: even try, and we’ll catch you, and you’ll be out of here so fast we’ll need to send a cleanup crew to remove the still-warm coffee mug from your desk. The crowd took in all of this silently, with at most the occasional murmur or hint of furtive chatting between neighbors. The rousing buzz of the first speakers lingered; this HR session was like the DUI check you had to negotiate while driving home from a particularly convivial party. The cops seemed amenable, if a bit stern, as they went through their script. Then, we came to the juicy frivolities. Here, the male (oddly) took the lead and stood up to address the new employees. Picture the Facebook corporate scene for a moment: buildings full of young, emotionally inept male geeks, and sprinkled throughout them, maybe a 10 percent population of young women.

Location 3623-3625

I had five bugs to fix. I didn’t even know how to code PHP, the front-end language Facebook was written in. It was a famously crappy language and development environment with few users those days, chosen merely because that’s what Zuck knew as a hacker at Harvard.

Location 3638-3641

While the role of product manager is near universal in tech companies of any size, the de facto or de jure reality of it varies widely. What the PM does is in many ways representative of how the company itself develops product. Some companies have opted for different titles. At Microsoft, they are known as “program managers.” At Palantir, the secretive defense intelligence software company founded by the billionaire investor Peter Thiel, they are known as “product navigators,” which sounds terribly romantic.

Location 3644-3648

A more illustrative description is “shit umbrella.” If you imagine a bursting monsoon of clumpy diarrhea pouring down like God’s own biblical vengeance, that is more or less where you find yourself in either a startup or a large, high-profile, and complex organization like Facebook. You, my dear product manager, are the communal manservant to your engineering team, holding a large, cumbersome paramerde above their heads bent over keyboards on which they furiously type.

Location 3648-3654

By definition, what you do is everything that needs to be done, other than hands-on-keyboard type code. So that means sitting in endless meetings with the privacy legal team, giving highly selective and edited versions of what your product is going to do, and explaining how it fits into some antediluvian legal rubric. It means pitching a roomful of smiling and empty-headed salespeople, so they can start priming the client pump and bring spend into your new product baby. It means wheeling and dealing with other PMs to either cajole a product change or cadge some engineering resource. It means fronting for the product in high-level meetings with senior management, and trying to place it, like a quickly falling Tetris piece, into the rudimentary blueprint of their world. It means defending your team from the depredations of other PMs when they come around asking for favors, or arguing for a deprioritization of some element of your product plan for one in theirs.

Location 3656-3658

doesn’t matter if you have the United Nations or the church on your side (i.e., if management has anointed you as leader), you’re ending up in front of a firing squad sooner rather than later. The most pitiful sight in the Facebook Ads team was the PMs who had lost the confidence of their engineers.

Location 3660-3666

Externally, it was a different story. As a Facebook product manager, you resembled an Afghan warlord or a pirate captain: fearsome in appearance to any outsiders, the scourge of entire companies and industries, but actually barely in control of your small band of engineer-hooligans, and always one step from mutiny. To the outside world, your job was easy: a two-line email would have the senior management of any company waiting eagerly in the Facebook reception area almost instantaneously. Many were the startups I conjured thusly, they sputtering in flattery despite my showing up late and surly, demanding and getting a full walk-through of their entire product and business model, then dismissing them after a forty-five-minute

Location 3672-3675

Targeting is where the data rubber meets the money road, so to speak. To use a physics analogy, pressure is force per unit area, and similarly, monetization is amount of data per pixel: the more data you bring to bear for every square inch of screen real estate, then the more that ad will be worth. Targeting is how that data gets applied to the screen real estate, it’s the alchemy that turns pure data into real-world

Location 3694-3696

The reality was otherwise. Before 2013, if you wanted to know how Facebook made money, the answer was very simple: a billion times any number is still a big fucking number. Facebook monetization was laughable compared with Google’s on a purely CPM basis. But usage was ungodly. Up there with heroin, carbohydrates, or a weekly paycheck: that’s how addictive and rewarding Facebook

Location 3707-3709

Rather than an overall ads strategy, what we had was a general feeling in the air, a sort of collective mood that, like a slanted floor, tended to send more people one way than the

Location 3758-3760

Any culture able to shut itself off from the outside world goes insane in its own unique way, and Facebook had essentially done that with its Ads team. But as on Wall Street, where even someone who knew the correct price of a security couldn’t go against the will of the market, you couldn’t question the reigning insanity. And so one went along.

Location 3761-3762

magnificent apotheosis in the rise and fall of a bet-the-company product called Open Graph, and its monetization twin, Sponsored Stories, and in later chapters you’ll hear plenty about

Location 3764-3765

beasts. They change quickly. Facebook’s redeeming quality—the magical tendency that has saved it in the past, and will save it in the future—is its ability to rapidly adapt to changing circumstances, or to the results of its own bold product bets.

Location 3773-3775

By itself, genius can produce original thoughts just as little as a woman by herself can bear children. Outward circumstances must come to fructify genius, and be, as it were, a father to its progeny.

Location 3780-3783

“Home,” the Facebook-enabled home screen for Android phones, launched with much fanfare in 2013, Zuck appearing alongside the CEO of the soon-to-be-disappointed smartphone maker HTC? Or Facebook’s misguided bet on HTML5 in 2012, which slowed the mobile app to a frustrating crawl? How about Facebook’s first version of search, available in English only, mostly useful for checking out your friends’ single female friends, and since discontinued?

Location 3787-3791

No. I submit he was an old-school genius, the fiery force of nature possessed by a tutelary spirit of seemingly supernatural provenance that fuels and guides him, intoxicates his circle, and compels his retinue to be great as well. The Jefferson, the Napoleon, the Alexander . . . the Jim Jones, the L. Ron Hubbard, the Joseph Smith. Keeper of a messianic vision that, though mercurial and stinting on specifics, presents an overwhelming and all-consuming picture of a new and different world. Have a mad vision, and you’re a kook. Get a crowd to believe in it as well, and you’re a leader.

Location 3795-3799

Many cool Valley companies have engineering-first cultures, but Facebook took it to a different level. The engineers ran the place, and so long as you shipped code and didn’t break anything (too often), you were golden. The spirit of subversive hackery guided everything. In the early days, a Georgia college kid named Chris Putnam created a virus that made your Facebook profile resemble MySpace, then the social media incumbent. It went rampant and started deleting user data as well. Instead of siccing the FBI dogs on Putnam, Facebook cofounder Dustin Moskovitz invited him for an interview, and offered him a job.

Location 3802-3806

This culture is what kept twenty-three-year-old kids who were making half a million a year, in a city where there was lots of fun on offer if you had the cash, tethered to a corporate campus for fourteen-hour days. They ate three meals a day there, sometimes slept there, and did nothing but write code, review code, or comment on new features in internal Facebook groups. On the day of the IPO—Facebook’s victory rally—the Ads area was full of busily working engineers at eight p.m. on a Friday. All were at that point worth real money—even fuck-you money for some—and all were writing code on the very day their paper turned to hard cash.

Location 3818-3822

Pause and consider all this for a lingering moment: the militant engineering culture, the all-consuming work identity, the apostolic sense of devotion to a great cause. The cynics will read statements from Zuckerberg or some other senior exec about creating “a more open and connected world” and think, “Oh, what sentimental drivel.” The critics will read of a new product tweak or partnership, and think Facebook is doing it only to make more money. They’re wrong.

Location 3822-3825

true believers who really, really, really are not doing it for the money, and really, really will not stop until every man, woman, and child on earth is staring into a blue-framed window with a Facebook logo. Which, if you think about it, is much scarier than simple greed. The greedy man can always be bought at some price or another, and his behavior is predictable. But the true zealot? He can’t be had at any price, and there’s no telling what his mad visions will have him and his followers do.

Location 3831-3834

additional plus for Google Plus: it had no ads, as Google could subsidize it with AdWords, its paid search gold mine. This was the classic one-hand-washing-the-other tactic of the ruthless monopolist, like Microsoft using the revenue from Windows to crush Netscape Navigator with Explorer back in the nineties. By owning search, Google would bankroll taking over social media as well.

Location 3835-3839

But as the one-way parade of expensive talent from Google to Facebook continued with no end in sight, Google got nervous. Companies are like countries: the populations really vote only with their feet, either coming or going. Google instituted a policy whereby any desirable Googler who got a Facebook offer would have it beaten instantly by a heaping Google counteroffer. This, of course, caused a rush of Googlers to interview at Facebook, only to use the resulting offer as a bargaining chip to improve their Google pay.

Location 3849-3852

Zuckerberg was usually a poor speaker. His speech came at the rapid clip of someone accustomed to analyzing language for content only, and at the speed of a very agile mind that didn’t have time for rhetorical flourishes. It was geek-speak basically, the English language as spoken by people who had four screens of computer code open at

Location 3863-3866

In a company whose overarching mantras were DONE IS BETTER THAN PERFECT and PERFECT IS THE ENEMY OF THE GOOD, this represented a course correction, a shift to the concern for quality that typically lost out to the drive to ship. It was the sort of nagging paternal reminder to keep your room clean that Zuck occasionally dished out after Facebook had suffered some embarrassing bug or outage.

Location 3873-3875

Zuckerberg’s tone went from paternal lecture to martial exhortation, the drama mounting with every mention of the threat Google represented. The speech ended to a roar of cheering and applause. Everyone walked out of there ready to invade Poland if need be. It was a rousing performance. Carthage must be destroyed!

Location 3907-3911

Topic extraction is one of those critical but unsexy artificial-intelligence challenges that underlie huge pieces of Internet technology (e.g., Google Search), but never receive the attention of sexy initiatives like self-driving cars. In essence, it’s a programmatic way of mapping the convoluted parlance of human texts like messages, webpages, or social media posts into a dictionary of semantic categories. For example, your status update of “Tiger really managed to hit that birdie in the US Open” would be automatically mapped to the categories “Tiger Woods,” “Golf,” and “US Open.” In a world where human speech, rife with sarcasm,

Location 3921-3924

Part of the team wanted to do away with the original keywords altogether, forcing all advertisers to adopt the new, untested technology, creating a major operational challenge as advertisers struggled to adopt the new feature. Another faction wanted to simply launch the new features in addition to existing keywords, and specially designate them somehow, either in the design of the user interface, or via other means.

Location 3924-3926

As I’d soon learn, the tiebreaker role of the product manager, mediating between the internal dueling factions of a product team, or channeling the voice of the eventual user among engineers lost in low-level technical detail, was

Location 3940-3947

This may sound trivial, and in the scheme of things, it sort of was. But the mundane trade-offs of technical difficulty among various implementations, the prioritization of engineering time, the vagaries of user perception, and the demands of marketing a new product are the bread and butter of everyday product manager work. Flashy new flagship-product launches, like a new iPhone birthed on a conference room stage by a strutting Steve Jobs, are what draw mainstream headlines and the attention of nontechie normals, like swine hearing the clanging of a swill bucket. But the beat to which the Valley really marches, the actual workaday cadence of technological progress, is that line product manager in a conference room or an engineering bullpen alongside a team, like the first lieutenant leading a platoon in every great war film ever made, figuring out what to build, how to build it, and how to sell it once built. Thousands of such teams dot Silicon Valley, and it’s how the work of technology development really gets done.

Location 3950-3953

Product managers must either apply the brakes to impatient engineers who want their creations to see the light of day, or conversely, whip and drive to get perfectionist engineers to stop mucking with the code and procure some real users already. In general, be it at startups or aggressive companies like Facebook, there should be a cultural bias for launching. The perfect is very often the enemy of the good, and as the Facebook poster screamed from every wall: DONE IS BETTER THAN PERFECT. Very few companies have died due to launching early;

Location 3963-3966

In essence, I had earned my spurs: I had safely shipped new product the market seemed to want, leveraging an engineering team that respected my product guidance, and with a plan to iterate the product going forward, with metrics to mark our progress. In a nutshell, that’s what Facebook expected of its product managers.

Location 3972-3973

whatever it is, a good product manager will slave away until it’s going up and to the right, whatever that figure represents. You make what you measure, so measure carefully.

Location 3984-3985

It was incomprehensible, and it tested my faith (which, believe it or not, I certainly had at that time) in Facebook’s claim to unique primacy in the realm of user data. As with many mysteries of my newfound religion, experience

Location 4004-4011

And when it does, things aren’t as simple as they seem. I can almost guarantee you that when the text string “Obama” appears, in either a message or a post, in the Facebook equivalent of a bar in Alabama on a Friday night, half the time it is preceded by the word “fucking,” and you probably shouldn’t add that person to your “#Democratic Party” targeting cluster, unless you want him to X-out every goddamn smiling Obama photo that’s going to take over his Facebook experience.* Which, incidentally, is going to make Obama ads’ delivery tank due to all the negative feedback, which means their account manager will soon be emailing you wondering why they can’t spend their budget, and fuck you, it’s your problem, and you’ll grow to hate every goddamned redneck in Alabama, and the Obama campaign for monitoring its ads stats so closely.

Location 4022-4025

That scale worked against you when you were on the ads side, though. It was clear from Gokul’s management ethos, which was essentially “Go big or go home,” that we were being tasked with finding the one breakout product, à la Google AdWords, that would completely change Facebook’s fortunes. Given Facebook’s revenues, which were already huge (remember: a billion times anything . . .), it was very difficult to make a dent in the income statement with any product.

Location 4029-4031

Bucketing users in a congressional district, and allowing political advertisers to target specific voting precincts? Nope. Sounds good, but political budgets on Facebook were relatively small then, and materialized only every four years for an election.

Location 4041-4043

To earn your beer money as product manager you had to move that revenue needle a good 5 percent or so. That was about $100 million in revenue per year you had to generate, the equivalent of two or three Wall Street traders at a bank like Goldman. Between

Location 4069-4070

chose Pelayo for the middle name, after the eighth-century Visigoth nobleman who initiated the seven-century-long struggle to free Spain from Islamic rule. Thus did Noah Pelayo come into the world, and I was the father to a son as well as a daughter.

Location 4082-4085

But then, that would be doing the bourgeois family thing full-on, with combined finances, expectations of pricey schools, mortgages, the entire needy contraption of settled-down life. To me, only the man who needed nothing was truly free. Until I was financially independent (e.g., fuck-you money), or the captain of a profitable enterprise, I was merely a slave whose bondage was worth one or another price, locked in as much by diapers and tuition costs as by a vesting schedule.

Location 4092-4095

All my Facebook colleagues in the over-thirty cohort were in that dependent boat. Facebook said jump and they could only ask “How high?” And jump they did, reciting whatever corporate script and leaping through whatever hoops their paymaster required, down to the logo-ed onesie they’d slap on their newborns, photos posted on Facebook to the online applause of their equally enslaved colleagues.

Location 4098-4100

child or parent, ever again, for the same reasons of personal history. I couldn’t do it, no matter the paternal draw of vulnerable newborn Noah, nor the charms and wiles of always-clever Zoë.

Location 4098-4100

family situation, as either child or parent, ever again, for the same reasons of personal history. I couldn’t do it, no matter the paternal draw of vulnerable newborn Noah, nor the charms and wiles of always-clever Zoë.

Location 4123-4124

Feed was still sacrilegious, and not mentioned in polite company. The mere thought of using outside data in Facebook ads delivery was similarly heretical, and not even considered.

Location 4155-4160

Every quarter Sheryl scheduled a gargantuan meeting meant to show off the wonderful tools engineering was building for Sales, and how well the hybrid engineering-operations teams were collaborating. Sheryl’s managerial prowess was on full display in these powwows, as she skillfully held court among the assembled lieutenants of the various subrealms. Picking up subtle psychological cues from an off-the-cuff remark here, unearthing some lingering issue that lay dormant over there, making sure every voice was heard but no voice heard too much, tamping down some burst of irrelevant drama to keep the action moving, the woman knew how to run a roomful of big names and even bigger egos.

Location 4200-4201

You moan about your friend’s breast-feeding photo being flagged, but fail to notice the complete lack of porn in your News Feed.

Location 4203-4206

Our social shadow warriors did have one showcase: there was an internal Facebook group with the provocative name of “Scalps@Facebook.” It was essentially an online trophy case of taxidermied delinquents, the sexual predators, stalkers, and wife-beaters whom the FB security team, in conjunction with law enforcement, had managed to catch. The weird thing about it was that the posts featured

Location 4210-4212

don’t know the official answer, but I can speculate. If Facebook were to publicize its very real efforts at stopping crime, people might associate the blue-framed window with the thought of some predatory creeper. As is, many people have an ambiguous relationship with Facebook.

Location 4222-4224

Perhaps just once, marvel at your Facebook experience, at its almost total lack of pornography, spam, hate speech, and general human detritus, and consider what spectacular systems and expertise must exist for a few hundred people to safeguard the online experience of 1.5 billion users, fully a fifth of humanity, on a continual, 247 basis.

Location 4249-4252

As a geographic tangent: New Zealand was commonly used as a test bed for new user-facing products. It was perfect due to its English-language usage, its relative isolation in terms of the social graph (i.e., most friend links were internal to the country), and, frankly, its lack of newsworthiness, so any gossip or reporting of new Facebook features ran a low risk of leaking back to the real target markets of the United States and Europe.

Location 4261-4262

More convincing is the actual downstream monetization resulting from someone clicking through and buying something—assuming Facebook got the conversion data, which it often didn’t, given that Facebook didn’t have a conversion-tracking system.

Location 4280-4282

evidently the one-man data protection office there. How the land of suffocating Catholicism and potato famines came to be the regulator of the biggest accumulation of personal data since DNA is a curious story. Ireland,

Location 4281-4289

How the land of suffocating Catholicism and potato famines came to be the regulator of the biggest accumulation of personal data since DNA is a curious story. Ireland, following its rise as the “Celtic Tiger” in the mid-aughts and subsequent property crash, actively encouraged American companies to choose Dublin as their European base. Ireland offered tax incentives (the corporate tax rate was very low), an educated (and desperate for work) talent pool, and a rational legal framework. The net result was that the waterfront docks east of Dublin became colonized by US tech companies like Facebook, Google, and Airbnb, each running (mostly) its sales and operations teams out of gleaming new high-rise offices. Facebook occupied several floors, each decorated with a European flag marking the spot where that country’s sales and ops team lived. The joke was that the Facebook Dublin office was the Noah’s ark of the EU: if continental Europe was consumed in some cataclysm, it could be repopulated with the breeding pairs located under each flag in the office. And yes, we had breeding pairs, as the office’s gender ratio was true to life, unlike that at the corporate mother ship, which was a never-ending sausage party.

Location 4301-4306

The principal reason for you to be technical is not to help technically design the system under development; if you’re doing that, then you’re PMing wrong. No, you’re technical so you can tell when engineers are bullshitting you, which will be often. At times it’s accidental (as it was with Rong), due to either miscommunication, bad memory, or wishful thinking (engineers are as inclined to it as anybody). Sometimes it’s more stealthy, their passive-aggressive way to disagree with the product direction (“That’ll eat up all our servers”), or laziness (“It’s impossible to build that”). The PM is there to give a sniff test to any such product-killing assertion.

Location 4312-4315

knew enough about how the back-end Ads system worked to log in and poke around the various machines. The targeting tables defined what targeting segments the Facebook Ads system used, and formed the logical glue connecting the eyeballs to the buckets of money. If the database tables underlying the targeting logic were inconsistent (e.g., certain table rows were missing when they shouldn’t have been), ads targeting would cease to function, and

Location 4341-4342

How many times have you heard some variant of these questions? The reality is that Facebook isn’t showing you anything.

Location 4342-4352

Here’s what people don’t understand about advertising. Facebook is simply a routing system, almost like an old-time telephone exchange, that delivers a message for money. The address on that message can be approximate (e.g., males aged thirty-five in Ohio), or it can be specific (e.g., the person who just shopped for a specific pair of shoes on Zappos). But either way, Facebook didn’t make the match of user and message, and at most decides secondary things like how often the ad is seen in general, or which of two ads addressed to you is seen that particular instant. In this sense, ads on Facebook are no different from phone calls or emails. We receive commercial versions of both in the form of spam and telemarketing calls. And yet, when we get a penis-enlargement email, nobody blames Google for providing Gmail, does he? Nor do you blame AT&T for the marketing call that distracted you from Game of Thrones. The only difference is that while people commonly make phone calls and write emails, few if any people address and post an ad. Like infants who haven’t learned object permanence yet, the Facebook whiners see an ad, the Facebook logo, and assume it’s all connected. Make the ad go away, and they don’t even think about it. Of course, what they should really be thinking about is how that ad got addressed, and what the advertiser, and not Facebook, knows about them.

Location 4355-4357

If they were to publish content themselves, or work in the business of delivering all of humanity’s digitized social life 247 all over the world, they’d realize there’s a human cost to that blue-framed browser tab, and it most certainly is not free. Ad blocking is tantamount to theft, or at the very least running a toll booth without paying.

Location 4359-4361

Setting a usage fee would be like IBM declaring it’s going to pick an individual price for every investor who wants to buy a share, rather than leaving the price discovery to the open market of a stock exchange.

Location 4368-4370

Data policy changes are an orchestrated production on the order of the Academy Awards, and not to be taken lightly. The fact that a billion users have never even read the user agreement they’ve consented to is immaterial. The world’s lawyers certainly have, and that’s the real intended audience of that document.

Location 4385-4388

We needn’t have worried. The adoption of the new policy lost by a landslide: 90 percent of voters were against the new data policy that Facebook needed to survive. But . . . almost nobody voted. Not even close to 30 percent of users. As such, the voting result was “taken under advisement”—by which we mean “ignored.” And be grateful we did; Facebook would be in trouble right now otherwise, as the company-saving products that launched later would have been impossible under the old data policy.

Location 4396-4400

Think about it for a moment. Picking out a football jersey on some crappy cellphone photo and determining the accurate associated and commercially interesting topic (e.g., the 49ers), and assigning it to the right member of your family/social tree who shares that interest, based on reams of random drivel you’ve shared on Facebook, would be an image-recognition and machine-learning feat on the order of the moon landing. Facebook ads targeting is so simple, this was a preposterous thought, but no number of denials kept the journalists with their ah-ha! anecdotes from pressing their

Location 4403-4410

Look: Facebook could have a video of you fornicating wildly with a frisky German shepherd, with your Social Security number and bank details written in lipstick on the dog’s back, while some deep voice in the background read out your deepest, darkest secrets from childhood and adolescence, and you know what? No advertiser would give a shit. They would, however, love to know what movie you saw on Netflix last night, what’s in your Amazon shopping cart, every item you scrutinized when you last visited Best Buy, and how long it’s been since you bought a car (and which car). They also want to know what mobile devices and browsers you use and every website you visit, so they can serve and track media on each aforementioned device. Almost nothing of what you share on Facebook—sweet nothings to your mistress, photos of you passed out on a couch, your secret brownie recipe—is worth anything in commercial terms. So even assuming perfectly evil behavior on Facebook’s part, the company would have no use for

Location 4411-4413

Facebook doesn’t sell your data; it buys it. It does this by providing services to advertisers that incentivize them to let Facebook ingest the data you’ve generated outside Facebook. In fact, as we’ll soon see, Facebook is one of the most jealous guardians of user data known to man. It is a black hole of data that can never leave.

Location 4552-4554

ones. Your Microsoft Calendar, which ruled your life via audio reminders from your corporate iPhone, as well as those of your colleagues, was fought over like a hundred yards of no-man’s-land during World War I. The moment you would clear a meeting slot on your calendar, someone would likely send you an invite to fill it (there was an internal

Location 4570-4573

But there is another motivation: by hybridizing their corporate DNA with the pluck and daring of the startup entrepreneur, they revitalize their internal cultures and add traits not typically found among their recruitment fodder (i.e., smart but obedient engineering grads). It’s like the intentional mixing of refined European breeds with wild dingoes in Australia that produced the smart and rangy Australian cattle dog.

Location 4592-4595

If we don’t create the thing that kills Facebook, someone else will. “Embracing change” isn’t enough. It has to be so hardwired into who we are that even talking about it seems redundant. The Internet is not a friendly place. Things that don’t stay relevant don’t even get the luxury of leaving ruins. They disappear.

Location 4603-4604

Like the naive new recruit, I took those values to heart. And like the new recruit, I’d realize only later that the Facebook reality was rather more complicated.

Location 4605-4606

To fill the hour—that is happiness; to fill the hour, and leave no crevice for a repentance or an approval.

Location 4609-4610

camber, and you’re suddenly on a flawless surface winding

Location 4615-4618

Add in two dozen other gearhead degenerates going apeshit along with you on the track, and you’ve got yourself a ten-lap ticket straight to your own mental redline. When you finally pull off the track into the pits, your overheated car will reek of burnt brake and clutch, your damp clothes are glued with sweat to your skin, and you’ll remove your driving gloves with shaking hands. The adrenaline crash will hit, and you’ll be reminded of the moments after your first fight, or your first real

Location 4651-4652

Nonetheless, I’d have to patiently wait out my vesting schedule; the joke term for doing so in less spirited companies than Facebook was being a “VIP”—that is, “vesting in peace.”

Location 4654-4656

You were really going to spend at best two years there, so halve the number on your offer or term sheet, and then halve it again if it’s in the form of stock rather than options (thanks to ass-backward IRS treatment of Valley compensation). Doesn’t look like such a great deal now, does it?

Location 4741-4744

That feeling that the universe had conspired to make something big happen right there in front of you, and you were a part of it by being present. That’s what stared out at you from photos by Glinn and Korda in those heady days of early 1959, which is why they transfixed the world, and still decorate some confused malcontent’s bedroom in Berlin or La Paz.

Location 4750-4755

At their extremes, capitalism and communism become equivalent: Endless toil motivated by lapidary ideals handed down by a revered and unquestioned leader, and put into practice by a leadership caste selected for its adherence to aforementioned principles, and richly rewarded for its willingness to grind whatever human grist the mill required? Same in both. A (mostly) pliant media that flatters the existing system of production, framing it as the only such system possible? Check!

Location 4763-4765

As the ever-sagacious @gselevator quoted: in communism people made lines for bread, while in capitalism they make lines for iPhones. Sure, iPhones are better than bread, and the standard of living in capitalist countries is clearly higher, but the lived experiences of either, from the point of view of the working proles, bears more than a passing resemblance.

Location 4766-4771

reality is that capitalism, communism, and every other sweeping ideology feed off the same human drives—the founder’s or revolutionary’s narcissistic will to power, and the mass man’s desire to be part of something bigger than himself—even if with very different outcomes. National Socialism, Technofuturism, Bolshevism, the Islamic State, Pan-Arabism, la Commune, Jonestown, the Crusades, la mission civilisatrice, the white man’s burden, evangelical Christianity, Manifest Destiny, Spanish Falangism, the Church of Latter-day Saints, the Cuban Revolution—the villain with a thousand faces, yoking together the monomaniac’s twitchy urge and the follower’s hunger for a role in some captivating story.

Location 4793-4796

In day-to-day terms, it was something like what the famous Google masseuse Bonnie Brown wrote in her autobiography about that company’s encounter with the bipartite wealth split: A sharp contrast developed between Googlers working side by side. While one was looking at local movie times on his monitor, the other was booking a flight to Belize for the weekend. How was the conversation Monday morning going to sound now?

Location 4814-4815

And yet the managers sided with their overseers against the very people they worked alongside every day. Say what you like about Marxism in practice, but it describes our contemporary technobourgeois society exactly.

Location 4819-4821

A man with a conviction is a hard man to change. Tell him you disagree and he turns away. Show him facts or figures and he questions your sources. Appeal to logic and he fails to see your point.

Location 4834-4836

it. It’s a case study in how even very smart companies can go temporarily mad and believe in fairies or flying saucers, under the twin pressures of market expectation and blinkered arrogance. Every large company has languished in the delirious trance of some product folly, waking

Location 4839-4844

Those were simple times in Facebook-land: Mobile usage was low and the app was buggy and slow. Pages were the only sanctioned commercial expression on Facebook, each being a sort of simplified personal profile for a brand or business (remember, the data-rich and expansive Timeline we all now enjoy hadn’t yet been launched). The only ways for someone else to get a piece of content into your feed were a friend posting it, a nonfriend tagging you in a photo, and a page you had liked posting something popular. There was no way to pay and get into users’ feeds directly. The few ads were small, postage-stamp-sized affairs in the far right-hand side of the page,

Location 4847-4849

Facebook—or, really, Zuck—saw News Feed as the untouchable magic real estate of the Internet, in which the dirty mitts grasping at filthy lucre had no place. How did Facebook make money then? Two ways, which each reflected the high-level bipartite split in the advertising world.

Location 4853-4855

However, it was never quite explained how Likes would transubstantiate into actual sales dollars. This was one of the mysteries of the Facebook faith that believers had to simply accept as beyond human understanding. To attempt to measure any of this, as with radiocarbon dating the Shroud of Turin, would have been spiritual

Location 4856-4861

The direct-response advertisers, who, you’ll remember, are those actively selling you a sweater or a flight to Boston right now, were barely represented on Facebook at the time. Facebook’s targeting system was so weak that nobody could actually directly sell anything on Facebook (other than the aforementioned Likes). The only direct marketers who managed to make Facebook advertising work were developers on the FB gaming platform (e.g., Zynga). There was a bit of signal amid the noise in Facebook’s “interests” targeting, which crafty marketers teased out over thousands of ads and millions of dollars spent. For example, a large gaming company discovered that people who had liked various energy drinks converted very well for Mafia Wars–style games.

Location 4871-4876

A later version added the verbal dictionary that accompanied “Like,” expanding the Facebook vocabulary to things like “play,” “listen,” “watch,” or “buy.” It was the new subject-verb-object language of everything you did online. “Antonio García Martínez listened to Wax Tailor’s ‘Only Once’ on Spotify.” Rather than merely express some vague approval via Like, Facebook users could now broadcast everything they were doing, with the aid of outside developers who built Facebook’s new grammar into their products. By doing so, these developers made their products “social,” and potentially viral. In exchange for pumping their data full-throttle into Facebook, those outside developers—music players like Spotify, or publishers like the Washington Post—got News Feed distribution,

Location 4889-4892

Sponsored Stories was the classic indefensible two-miracle startup idea. Miracle one was all the companies in the world doing work to integrate with Facebook and give away their data, at dubious benefit to themselves. Miracle two was those companies’ marketers abandoning their old workflows and success myths for some newfangled paradigm, one with unproved and unprovable efficacy, just because Facebook said so. And so, like any two-miracle startup idea, it was almost certain to

Location 4931-4935

this was terrible news. The clickthrough rates on Facebook ads were abysmal, mostly due to their crappy targeting. A 0.05 percent clickthrough rate was average; achieve a rate of 0.11 percent and you were ordering the truffled steak at Alexander’s in SoMa (on the client’s dime, of course). Compared with regular display advertising, Facebook ads didn’t even register; even the worst-targeted ads (e.g., those moronic “punch the monkey” pseudogame banners) got clickthrough rates of at least 0.1 percent, and well-targeted ads using first-party data and rich, dynamic creative got click rates as high as 1 percent or more. It meant Facebook ads underperformed by as much as twenty times, if you were actually doing

Location 4944-4945

Since Facebook Ads didn’t ship products that people actually asked for, launching always had a certain foie-gras-duck-undergoing-gavage quality to it: open up and pump it in.*

Location 4960-4962

To the salespeople, who led shitty, underpaid lives spent rehearsing the script of a movie in which they never starred, this was one of the very reasons for working at Facebook. Everyone got his show in the end.

Location 4980-4981

Those fairy tales had been believed for a while, with the biggest brand budgets in the world—Burberry, Ford, Starbucks, BMW—spending money on Likes they didn’t know what to do with, and that at best were a license to spam users’ News Feeds.

Location 4986-4986

Growth for the sake of growth is the ideology of the cancer cell.

Location 4990-4993

This idea was of particular interest, since it brought me into contact and conflict with the single most important group at Facebook, the one team that above all others can be credited with Facebook’s humanity-spanning success. That team was known simply as “Growth.” There are three certainties in life: death, taxes, and Facebook user growth.

Location 5030-5032

Eventually Russia, Iran, India, Brazil, and parts of Africa will fall to the Growth team’s patient ministrations. Then, Mark Zuckerberg, like a young Alexander the Great at the Indus River, will weep for having no more world to conquer.

Location 5037-5040

The idea was to sell this space to the richest and stupidest advertiser of all, the legacy-admissions student of digital media: the brand marketer. Brand marketers, like politicians, are people paid to spend other people’s money. In this case, their putative goal is to raise “brand awareness,” that diaphanous and quasi-intangible substance that makes you covet a $10,000 Rolex Submariner when you finally get that bonus or promotion.

Location 5037-5040

The idea was to sell this space to the richest and stupidest advertiser of all, the legacy-admissions student of digital media: the brand marketer. Brand marketers, like politicians, are people paid to spend other people’s money. In this case, their putative goal is to raise “brand awareness,” that diaphanous and quasi-intangible substance that makes you covet a $10,000 Rolex Submariner when you finally get that bonus or promotion.

Location 5037-5040

The idea was to sell this space to the richest and stupidest advertiser of all, the legacy-admissions student of digital media: the brand marketer. Brand marketers, like politicians, are people paid to spend other people’s money. In this case, their putative goal is to raise “brand awareness,” that diaphanous and quasi-intangible substance that makes you covet a $10,000 Rolex Submariner when you finally get that bonus or promotion.

Location 5038-5040

paid to spend other people’s money. In this case, their putative goal is to raise “brand awareness,” that diaphanous and quasi-intangible substance that makes you covet a $10,000 Rolex Submariner when you finally get that bonus or promotion.

Location 5095-5097

The longevity of these keys and their associated data is important, however. We move our physical address less often than we clear the cookies in our browser, or than McAfee does for us. Changing devices is somewhere in between, which is why data-privacy issues on mobile are so much more challenging than on desktop.

Location 5098-5101

thing going on in marketing right now, what is generating tens of billions of dollars in investment and endless scheming inside the bowels of Facebook, Google, Amazon, and Apple, is the puzzle of how to tie these different sets of names together, and who controls the links. That’s it. Other than this Game of Thrones power struggle among the great digital powers to control identity, targeting, and attribution, everything else is a parasitic sideshow scarcely worth the hassle of following.

Location 5126-5131

submit that the role of modern-day Big Brother is actually played by companies you’ve likely never heard about. Companies with names like Axciom, Experian, Epsilon, Merkle, and Neustar (among others). These are the companies that since the dawn of the direct-marketing age in the sixties and seventies have been tracking all of consumer America. They know your name, address, phone number, email address, education level, rough income, who else is in your household and their ages and spending patterns, and which consumer segment you fit into—and they’ve been accumulating this information since before the Internet even existed.

Location 5136-5140

We live in a society in which your cultural, religious, or political identity has been supplanted by your consumption patterns (“soccer mom”; “iPad generation”; “Nascar dad”). In fact, companies like Nielsen have constructed detailed menageries of consumers, organized along multiple dimensions like family size, education, income, dwelling, and so forth, and with amusing monikers like “Beltway Boomers” and “Country Casuals.” Marketed as the “PRIZM Segments,” they form a sort of predictive zodiac of consumption, purportedly allowing marketers to tailor their inducements to demographic archetypes.

Location 5153-5155

Do the math. It works out to about $1,000 CPMs (cost per thousand views). Most online media sells for somewhere around $1 CPM, perhaps a bit more. That means online advertising can get away with being a thousand times less effective than mail, and still come out even in terms of your marketing return.

Location 5159-5168

The term of art to describe this witchcraft is “data on-boarding,” and it works as follows: Companies like Datalogix, Neustar, and LiveRamp buy Web real estate on second-tier social networks (Hi5 or Orkut, anyone?), email newsletters, dating sites (hint: Match.com doesn’t just make money on subscriptions), or anywhere that personal information like name and address meets a browser. That means these on-boarding companies literally have a tiny image (usually just white space) on whatever your browser is loading when you’re interacting with an email or with a website where you’ve got an account—anyplace that knows offline details about you. That’s enough for them to either drop a cookie or read one that’s already there. Since they know from the newsletter you’re reading that your email is agm@gmail.com, or from Match.com that your name is Antonio García Martínez, they know to associate that browser cookie with various pieces of your personal information. That personal information is stored in a database, along with the browser cookie that corresponds to it, forming a bridge from real-world you to the browser version of you. It’s probably in hashed form, but that’s just privacy theater; if everyone agrees on the same hash function, it doesn’t matter how it’s stored.

Location 5172-5177

Think about it: Media publishers like Facebook and Google are just more efficient versions of the post office. They deliver a message for money. They’ll even give you a return receipt if you ask for it. What’s different is the address they use to send it. In our mediated age, we’ve gotten to the point that a Facebook or Google user ID is a better way of reaching you than a name or physical address. I can scar your retinas for almost nothing (relatively) if I know your Facebook user ID; your name and address would do nothing for me—unless, of course, they provided the primary key to a database of buying behavior spanning decades. Which they do, and that’s the only reason, along with the legacy inertia of big-company business relationships, that Acxiom, Epsilon, et al. are even still in business.

Location 5180-5184

The advantage that Facebook and Google have over the regular data on-boarders is twofold: they have much more of your personal data, and they see you online all the time. The match rate (i.e., the percentage of offline personas that can be found online) for Facebook’s Custom Audiences product, introduced in 2012 (and more on that shortly), is as high as 90 percent. That means for every hundred people marketers target via Custom Audiences, Facebook will find ninety of them, a shockingly high fraction in the fuzzy world of advertising.

Location 5184-5186

grail of all marketers: a high-fidelity, persistent, and immutable pseudonym for every consumer online. Even better, they’ve joined that to your real-world persona, the one that shows up bleary-eyed at two a.m. at a Target in El Cerrito looking for tampons or a six-pack of Natural Light.

Location 5209-5211

How did we finally get on the right track? The way Facebook sussed out anything: it mooched information from potential acquirees and business partners via meetings of dubious good faith, and then figured out how to hack the outside world to its advantage. It’s what every large

Location 5209-5211

How did we finally get on the right track? The way Facebook sussed out anything: it mooched information from potential acquirees and business partners via meetings of dubious good faith, and then figured out how to hack the outside world to its advantage. It’s what every large company with incumbent leverage does, incidentally.

Location 5271-5274

In the FBX case, it would be the New York Stock Exchange of eyeballs, human desire traded for money billions of times a day in real time. With this real-time pipe from the outside world to all of Facebook, the company would go from targeting someone who had accidentally liked Jay-Z two years ago to targeting someone who had just chosen between three different shoes at Zappos, or read an article on the new Mazda Miata, or bought something on eBay, just like that!

Location 5279-5283

would be the product’s internal champion, and navigate Facebook’s internal turmoil to conjure resources and consensus. I’d also make sure all the proper people were in meetings and pitch sessions. But someone from outside had to serve as the voice of the market. Someone who knew the world of programmatic advertising inside and out, and yet had the nonsleazy salesmanship and technical chops it would take to convince Facebook engineering and product managers. Finding a knowledgeable yet trustworthy player in the advertising space was like finding a snowball in hell. But then, someone came to mind.

Location 5287-5289

Along with several other companies who became competitors, he started a “demand-side platform” (DSP): the sophisticated ads-buying technology that, like a stockbroker, interfaced with exchanges in the service of ads buyers.

Location 5334-5335

Think instead of Bluto (played by John Belushi) in Animal House, but slimmed down, and with a computer science education (and perhaps even $200 sunglasses). Code, curls, and girls, those were the priorities of the day, and in that order.

Location 5336-5338

reflected the reality that coding, far from being the loserly redoubt of the socially isolated loner, was now the avenue for social mobility and elitism of even the football captain. Whenever real money and status were on the line, as they were now, the status-craving males were never far behind, whatever the task.

Location 5339-5340

“Brogramming” Facebook page that achieved cult status, even outside Facebook. As a measure of its bro-y cheek,

Location 5340-5344

fairly serious legal document subject to constant debate and rewriting) by rephrasing it in brogrammerese and posting it publicly. (Sample line: “We give lots of fucks about your privacy, so we wrote this. Read it, so you know what the fuck we’re going to do with the shit you post.”) The tech ecosystem and its chroniclers like TechCrunch, forever agonizing about the deplorable state of women in tech, started howling about the corrosive effects of such a culture. (Some internal wags helpfully suggested starting a page for the female version of a brogrammer, “hogrammers” as they

Location 5449-5452

One of the strange hypocrisies around the “hacker way” was that the company was so big most employees didn’t have the ability to hack on anything at all, engineers being vastly outnumbered by every flavor of corporate camp follower. Typically these nonhackers would just absent themselves from the geeky nocturnal ritual, but you couldn’t miss out on the pre-IPO party, so there was every salesperson, admin, operations specialist, and IT help-desk attendant well into the wee hours, milling around without much to

Location 5486-5488

IPO! We were living and watching it; our friends can only envy, our grandkids can only imagine when we tell them the story! Feel that temporary pause in baseline human existential angst by being part of something bigger than yourself, usually ignoble like a lynch mob, sometimes heroic like D-Day, rarely profitable like an

Location 5496-5497

What’s my big beef with capitalism? That it desacralizes everything, robs the world of wonder, and leaves it as nothing more than a vulgar market.

Location 5505-5507

dances, or art would cause anybody such revelry. Now we’re driven to ecstasies of delirium because we have a price tag, and our life’s labors are validated by the fact it does. That’s the smoldering ambition of every entrepreneur: to one day create an organization that society deems worthy of a price tag.

Location 5508-5509

Clap at the clever people getting rich, and hope you’re among them. Is

Location 5508-5512

Clap at the clever people getting rich, and hope you’re among them. Is it a wonder that the inhabitants of such a world clamor for contrived rituals of artificial significance like Burning Man, given the utter bankruptcy of meaning in their corporatized culture? Should we be surprised that they cling to identities, clusters of consumption patterns, that seem lifted from the ads-targeting system at Facebook: “hipster millennials,” “urban

Location 5512-5516

Ortega y Gasset wrote: “Men play at tragedy because they do not believe in the reality of the tragedy which is actually being staged in the civilized world.” Tragedy plays like the IPO were bound to pale for those who felt the call of real tragedy, the tragedy that poets once captured in verse, and that fathers once passed on to sons. Would the inevitable descendants of that cheering courtyard crowd one day gather with their forebears, perhaps in front of a fireplace, and ask, “Hey, Grandpa, what was it like to be at the Facebook IPO?” the way previous generations asked about Normandy or the settling of the Western frontier?

Location 5549-5550

bonuses and what all to get to a few hundred K of dosh, all that changes is what I’ll call your indifference threshold to expense. If before you didn’t think about dropping

Location 5559-5563

change (even if not in a fully “fuck you” sort of way) very quickly; those of everyone around would change as well. From the slightly ridiculous nights out (I recall a double steak dinner and four-figure restaurant bills in there somewhere), to the rather rapid accumulation of Porsches, Corvettes, and even the odd Ferrari in the parking lot, things did assume a certain debauched air at Facebook, despite all the corporate clamor for austere discipline. That was all

Location 5573-5574

The news coverage surrounding the IPO, even from the supposedly savvy tech and financial press, was a reminder of that harsh lesson of life: there are those who write headlines about money for a living, and then there are those who make money.

Location 5636-5639

You’ll see that my analogy to Wall Street for programmatic media wasn’t merely historical; it was an exact comparison reflecting both business and technical realities. FBX was the Flash Boys of media, trading quanta of human attention at the speed of light. Every time you load a new page in Facebook, not to mention most of the Internet, optical signals crisscross the globe to hundreds of waiting machines, all announcing, like some phalanx of royal trumpeters, your impending arrival.

Location 5641-5645

“demand-side platforms” (DSPs), and they’re the stockbrokers of this real-time media world, working in the employ of the advertiser or agency who wants to sell you something.* The DSP quickly unpacks the bid request, and queries its data for anything it knows about you: sites you’ve browsed, shopping carts you’ve abandoned, that airfare quote you got and never acted upon. They’re all there and returned within milliseconds via state-of-the-art databases hyperoptimized for the purpose.

Location 5651-5655

The real-time, millisecond bids from Facebook Exchange then fed into the regular ads auction that Facebook was holding on behalf of non-FBX advertisers. At which point Facebook, with all of its vaunted data, became merely one more DSP in the auction, just one more broker doing its business. The difference was, the rest of the Facebook Ads system had Facebook’s set of data, while the FBX advertisers, however many there were in that auction, had their own data from the non-Facebook world. Since both the Facebook Ads system and the outside world now received a request for an ad via FBX, we had put outside advertisers on an equal footing with Facebook itself, and may the best man (or bid)

Location 5664-5668

Note: There was nothing magical about FBX. It was merely sophisticated plumbing. The point was that joining Facebook Ads, via a real-time exchange, to a world of data it didn’t previously have access to was the fastest and most sophisticated way to get that data into Facebook. Custom Audiences could also bring the exact same data into Facebook, but nobody who dealt in sophisticated targeting for a living wanted to use Facebook’s clunky infrastructure to facilitate it. To cite loose technical analogies, CA versus FBX was like faxes versus email. Surely Silicon Valley’s most aggressive company would see the historical inevitability of programmatic exchanges and their technical superiority.

Location 5672-5674

When it came to monetization, Facebook had no interest in real innovation. It liked its faxes. Like any large company, Facebook would always aim to create monopoly pricing power and maintain information asymmetry, rather than drive true innovation. If Facebook played with the outside world, it always played with loaded dice.

Location 5700-5702

To paraphrase Lord Palmerston about great nations, companies like Facebook had no eternal allies and no perpetual enemies, they merely had eternal and perpetual interests.

Location 5711-5713

(Twitter didn’t even enter this worldview, and was merely a distraction. Zuck had once famously derided the company as “a clown car that drove into a gold mine,” and that’s possibly the last time anyone at Facebook had thought about it.)

Location 5716-5718

With good reason too. These companies without question were smart, thought in decades rather than the years (maximum) of most tech companies, and had the wherewithal to project their own agenda despite Facebook’s wishes. Oh, and while we all pretended to be friends, nobody was even remotely stupid enough to believe that.

Location 5732-5737

There were lots of hard questions from both sides on how data would get used, and potentially recycled inside the obscure inner workings of the other company. Would Facebook record what Amazon ads were shown to which person, and use that data? Even in a distant, second-order way, would Amazon ad performance be used in some Facebook model around clickthrough rates? Facebook in turn wondered how far into the Amazon machine all this user-level impression data would flow, and if the product recommendations on amazon.com would suddenly reflect your interaction with ads on facebook.com. The two giants circled each other warily, neither willing to take its hand off

Location 5747-5752

By all accounts (and Google information security was clearly not as good as Facebook’s) this caused a considerable stir internally. In January 2012, at a company-wide Q&A, Google’s founder Larry Page addressed this new direction forcefully, quelling the internal dissent and issuing a Googler ultimatum. “This is the path we’re headed down—a single, unified, ‘beautiful’ product across everything. If you don’t get that, then you should probably work somewhere else.” With the gauntlet thrown down, Google products were soon ranked via one unique metric—how much did they contribute to Google’s social vision?—and either consolidated or discarded appropriately.

Location 5769-5771

Vic Gundotra was a former Microsoft exec who’d climbed the treacherous corporate ladder there before jumping to Google. It was he who had whispered a litany of fear into Larry Page’s ear, who had greenlit the project, and it was he who headed the rushed and top-down effort (unusual for Google) to ship a product within an ambitious one hundred days.

Location 5796-5799

To the extent it could produce revenue right now! it received the lukewarm support of Gokul and Rabkin, and the tolerance of the sales side, which took its cues from product anyhow. My naive assumption was that if I could only make FBX into a billion-dollar business, I’d convince the recalcitrant (to not say willfully ignorant) Facebook management of the size of the overlooked opportunity, and the tide of opinion would shift.

Location 5804-5815

To be someone or to do something, which would I choose?* A man occasionally reaches a fork in life’s path. One road leads to doing something, to making an impact on his organization and his world. To being true to his values and vision, and standing with the other men who’ve helped build that vision. He will have to trust himself when all men doubt him, and as a reward, he will have the scorn of his professional circle heaped on his head. He will not be favored by his superiors, nor win the polite praise of his conformist peers. But maybe, just maybe, he has the chance to be right, and create something of lasting value that will transcend the consensus mediocrity inherent in any organization, even supposedly disruptive ones. The other road leads to being someone. He will receive the plum products, the facile praise afforded to the organization man who checks off the canonical list of petty virtues that define moral worth in his world. He will receive the applause of his peers, though it will be striking how rarely that traffic in official praise leads to actual products anybody remembers, much less advances the overall cause of the organization. I certainly had options. I can’t say I didn’t. I had outs, and could have walked away from FBX.

Location 5819-5823

Consider the number of shares and the fact that at the time of this writing, in December 2015, Facebook is worth almost four times what it was when I struck my AdGrok deal. I had two children, whose mother’s nine-to-five corporate job provided for them, though just barely. Her London trader heyday was long gone, and now her salary was that of your standard-issue MBA inside the corporate machine. I contributed a slab of cash every month, per California support guidelines, but it would be my Facebook stock vesting yet to come that would pay for private high schools and Stanford, so Zoë and Noah wouldn’t have to sneak into this country’s elite through the back door from cattle class, as I had to do.

Location 5825-5826

But the ego drives, and honor compels, and stubbornness steels the will that moderation and judgment might sap. Faced as I was with an array of opponents, my reflex was the same as always: double down and dig

Location 5833-5834

There were four ad agencies, so-called holding companies, which served as umbrella organizations for the hundreds of smaller agencies that ruled the media world. Collectively, their revenues were in the tens of billions, and the total media spend they controlled or influenced

Location 5855-5859

app installs, offline purchases in stores, and so forth—would it build an open system that allowed advertisers to bring their data to Facebook, while preserving Facebook’s user privacy and preventing data leakage to the outside world? Or, as it had in the past, would it attempt to capture as much of the ads management stack as it could, even if it meant shipping inferior product, shunting out any and all companies other than the end advertiser?* The implications for revenue, arguably the most important issue in the debate, were severe.

Location 5920-5923

Fortunately, Facebook itself served as the society pages to its own courtly intrigue: the quasimaniacal Facebook usage among employees meant every debutante’s coming out with Zuck, or every kingly banquet at In-N-Out Burger, was there for all to see. Like a Brit on a bar stool scanning tabloids for news of the royal family, the Facebook hoi polloi friended or followed whomever they could among the Facebook nobility, and watched, nose pressed against the glass, the regal extravaganza unfold.

Location 5942-5946

In Katherine Losse’s so-so firsthand account of life at early Facebook, The Boy Kings, she describes a mafia of Harvard recruits. They were core hires, and would go on to form the backbone of future Facebook. By all reports, the atmosphere at early Facebook was bro-y and redolent with the joshing violence typical of young, hormonal males. One of the Harvard mafiosi got into the habit of threatening other engineers that he’d “punch them in the face” if they messed up. That person was Boz, as I was able to guess while discussing the matter with a Facebook old-timer who dished

Location 5965-5966

McLean was the inventor of the intermodal container, those metal boxes piled into immense heaps on the cargo ships coming from China.

Location 5971-5974

Modern digital advertising has also settled on a containerized box—or, rather, a set of them. They’re known as Internet Advertising Bureau (IAB) ad units for desktop ads, and Mobile Marketing Association (MMA) ad units for mobile. They come in the standard sizes, measured in pixels—728×90, 300×250, and so forth—of every banner ad on the Web, with other sizes for mobile.

Location 5978-5984

But there are some ships that consider themselves either too big or too important to accept containers—either because they don’t like the look of them on the deck, or because they claim that moving freight is really a side business and not their mission in life. So if you want to move anything through them, you suddenly have to repack everything in whatever arbitrary, random containers these ships accept. This situation is what we politely call “native ad formats” in the ads business. Those ships are products like Google Search, Facebook, and Twitter. Either because they started as user-focused viral plays unconcerned with monetization (Facebook, Twitter), or because they very intentionally bucked a going standard, basically because they could (Google Search), the net result is that standard ad formats get no play.

Location 5991-5997

Look: the advertiser doesn’t trust its agency, the agency doesn’t trust its trading desk, the trading desk doesn’t trust the ads-buying software it’s using, and the ads-buying technology company doesn’t trust the exchanges. The only thing that keeps this dishonest world honest is the existence of an agreed-upon source of truth. That oracle is the ad server. If a marketer wants to reach a million people in the eastern United States, showing each person no more than four ad impressions between the hours of four and ten p.m. on Thursday (a common buy for movies, incidentally, which always launch on a Friday), then that marketer will be satisfied only if the ad server report says that’s what he got. The ad server isn’t merely a data server spewing forth pixels on demand; it’s also the accounting system that decides what gets delivered when, to whom, how often, and where on the Internet.

Location 6017-6022

All Google has to do is get that person to the storefront. The value of a Google click is immense and expensive (which is why Google prints money). The value of a viewthrough to them is nil. In fact, it’s so nil that Google opts not to even show you an ad if your query doesn’t seem commercial, choosing to show you a blank billboard much of the time. Roughly half of Google queries don’t generate ads in the space next to them; test it yourself. To Google, the claims of viewthrough value seem bogus, an attempt to justify wasteful media spending. As a result, its ad server counts only clickthroughs. Its manual explicitly states, “Clicks trump impressions,” and Google drives that point home in every demo.

Location 6023-6024

it has to manufacture desire rather than exploit it. So in Facebook’s accounting, viewthrough is valuable indeed, and it would very much like credit for that splashy image (or soon, video)

Location 6103-6105

For my part, outside of physics textbooks, I found Truth to be a rather rare commodity, particularly in the tech world. I had also noticed that those who most made a big show of believing in Truth were unusually attached to whatever well-groomed pack of lies they held dear.

Location 6109-6112

I could imagine the feedback, without having been told the specific authors: Reesman, Hari, and Gary and others on the FBX team all gushing praise for my dedication and product leadership. And then Gokul and Boland decrying my uncouth insubordination, smart-alecky arrogance, and corrosive criticism of current Facebook strategy.

Location 6116-6119

Don’t be deceived by my withering treatment of Facebook in this book; inside every cynic lives a heartbroken idealist. If I’m now a mordant critic, it’s because at one point, like Lucifer once being the proudest angel before the fall, I too lived and breathed for Facebook, perhaps even more than most.

Location 6155-6159

In FBX land, it was a different story: we were competing for spend with budgets that had been earmarked mostly for Google and its real-time ad exchange AdX, which was how most retargeting was actually being done. This was good, in that FBX spend was truly “incremental” (to use Facebookese); that is, FBX revenue was new dollars that were not otherwise being spent on Facebook. The fact these dollars were formerly being spent on much-hated Google made them even yummier, of course. Incremental spend—new money!—was the holy grail for Facebook at that point. Incrementality

Location 6176-6181

Pause and think for a moment. What’s it like to be a Gokul or a Boland or a Fischer? You’ve seen excellent engineers and product people come and go, as your recruiters land the best new grads from every leading school. Your privileged position inside a powerful and market-leading organization exposes you to every industry trend and player, and pads your network with a roster of power and influence. Your ability to make any company’s senior management appear and deliver fawning and revealing product demos means you know everything that’s going on in your industry, down to the color of the clickable buttons in the products.

Location 6187-6189

You learn that what matters in a big company is to avoid falling victim to firing or layoffs, and to appear important and critical to the company’s mission. You have mastered the art of “managing up”: namely, controlling the feelings and perceptions of the management layer above you.

Location 6192-6196

You make sure to form allegiances and friendships with your peer managers, particularly in organizations like sales or business development that you’ll need to push your business agenda forward. When there’s an ineffective and incompetent member in the organization, rather than calling them an idiot to their face and firing them if possible, you channel feedback to their manager and learn to work around their incompetence. If the incompetence does not directly affect you or your team, you look the other way and focus on the levers you do control.

Location 6196-6198

middle management: you’re the necessary layer between the visionaries and risk-takers who created the organization, and the new acolytes of your religion for whom this is a job, and you are their first whiff of corporate culture and authority.

Location 6470-6473

In mobile, Web browsers generally don’t accept third-party cookies, which means that someone other than the New York Times can’t read or write data about you when you’re on nytimes.com on your mobile browser. Contrast that to the data mayhem that reigns on a desktop browser. Also, that mobile browser typically does not have access to your unique device ID, which, as we’ll see, is the real identifier that matters in the mobile advertising world.

Location 6482-6484

This is actually counterintuitive if you know how mobile data is stored: on mobile, every device has a unique ID that’s associated with the physical hardware you’re holding in your hand.* In theory, marketers could sell all that data, associating it with your device ID, and then use

Location 6492-6495

marketplace by the time Facebook showed up peddling Likes, that was emphatically not the case on mobile. On mobile, targeting data was sparse, and bad when it did exist, such that even basic targeting like age and gender was a godsend to data-starved mobile marketers who’d been mostly shooting in the dark. This was confirmed in a big way when Facebook Ads launched the only truly novel thing it’s launched since the IPO: the Facebook Audience Network.

Location 6492-6495

marketplace by the time Facebook showed up peddling Likes, that was emphatically not the case on mobile. On mobile, targeting data was sparse, and bad when it did exist, such that even basic targeting like age and gender was a godsend to data-starved mobile marketers who’d been mostly shooting in the dark. This was confirmed in a big way when Facebook Ads launched the only truly novel thing it’s launched since the IPO: the Facebook Audience Network. AN,

Location 6504-6508

In mobile, leading app publishers—mostly games companies and a few mobile commerce companies—were adept at monetizing their users, by banking both on the gatekeeping and on the ease of payment the App Store represented (e.g., “$3.99 to download,” beat it if you don’t like it), as well as a general culture of savvy monetization. The newspaper people turned digital publishers were still rubbing the newsprint off their hands as they tried to figure out “this Internet thing,” while the marketers inside leading games companies knew their CACs and LTVs out to three decimals.

Location 6504-6508

In mobile, leading app publishers—mostly games companies and a few mobile commerce companies—were adept at monetizing their users, by banking both on the gatekeeping and on the ease of payment the App Store represented (e.g., “$3.99 to download,” beat it if you don’t like it), as well as a general culture of savvy monetization. The newspaper people turned digital publishers were still rubbing the newsprint off their hands as they tried to figure out “this Internet thing,” while the marketers inside leading games companies knew their CACs and LTVs out to three decimals.

Location 6533-6536

Multiple insiders at the time leaked to me the concern around the strategy: pimping out News Feed had to end at some point, and then what would Facebook do? There were even internal projections guesstimating when the moment would arrive, when the ads pioneers would reach the Pacific Ocean on the westward fringe of this new ads continent, and realize that the land rush and the revenue boom were over.

Location 6547-6556

That’s the nature of Valley success, however: you try ten things, based mostly on random hunches, a few key product insights, and whatever internal mythologies your culture reveres. Seven of them fail miserably, are discontinued, and are soon quietly swept under the rug of “forever today” forgetfulness. Two do OK, for more or less the reasons you thought, but they don’t blow the doors off your success metric. And one, for reasons you discover only after the fact, becomes a huge, transformational success. The amnesiac tech press weaves the narrative fallacy around the proceedings, fabricating a make-believe dramatic arc from steely-eyed product ideation to flawless and unhesitating technical execution. What was an improbable bonanza at the hands of the flailing half-blind becomes the inevitable coup of the assured visionary. The world crowns you a genius, and you start acting like one. When the next usage or revenue crisis hits, you repeat the experiment, rolling your set of product dice on the big Valley table. At some point, you don’t find the crisis-solving winner, the dealer sweeps up your remaining chips, and you’re busted. The company fails and your logo is recycled as a reminder of corporate mortality. Then everyone wonders how such a confirmed genius could have possibly failed, and ruminates on the transience of talent.

Location 6631-6636

A breeze from the grave suddenly blew across dreams of youth, which lay caked in the patient dust of “someday,” and made them gleam again with their original brilliance. I withdrew from any professional obligations, sold everything, and committed to two all-consuming goals: First, to finish the book you are holding, whose initial scribbles dated back to the events described, but whose completion required a year of itinerant writing and editing. Second, to sail around the world alone on a small boat. This was a dream cultivated since childhood, when I raced my Optimist dinghy in Biscayne Bay by day, and read of Robin Lee Graham’s adventures by night. Graham was an American sailor whose solo circumnavigation in 1970, on a cramped twenty-four-foot boat named Dove, made him the youngest person to ever complete such an ocean passage.

Location 6643-6645

To quote Theodor Herzl, “If you will it, it is no dream; and if you do not will it, a dream it will remain.” Willing dreams into existence was what being an entrepreneur and a product manager was about; for once, finally, they’d be my dreams, rather than purely corporate or mercenary ones.