What do you think?
Rate this book
Subprime Attention Crisis
From FSGO x Logic : a revealing examination of digital advertising and the internet's precarious foundation
In Subprime Attention Crisis , Tim Hwang investigates the way big tech financializes attention. In the process, he shows us how digital advertising―the beating heart of the internet―is at risk of collapsing, and that its potential demise bears an uncanny resemblance to the housing crisis of 2008.
From the unreliability of advertising numbers and the unregulated automation of advertising bidding wars, to the simple fact that online ads mostly fail to work, Hwang demonstrates that while consumers’ attention has never been more prized, the true value of that attention itself―much like subprime mortgages―is wildly misrepresented. And if online advertising goes belly-up, the internet―and its free services―will suddenly be accessible only to those who can afford it.
Deeply researched, convincing, and alarming, Subprime Attention Crisis will change the way you look at the internet, and its precarious future.
FSG Originals × Logic dissects the way technology functions in everyday lives. The titans of Silicon Valley, for all their utopian imaginings, never really had our best interests at recent threats to democracy, truth, privacy, and safety, as a result of tech’s reckless pursuit of progress, have shown as much. We present an alternate story, one that delights in capturing technology in all its contradictions and innovation, across borders and socioeconomic divisions, from history through the future, beyond platitudes and PR hype, and past doom and gloom. Our collaboration features four brief but provocative forays into the tech industry’s many worlds, and aspires to incite fresh conversations about technology focused on nuanced and accessible explorations of the emerging tools that reorganize and redefine life today.
In Subprime Attention Crisis , Tim Hwang investigates the way big tech financializes attention. In the process, he shows us how digital advertising―the beating heart of the internet―is at risk of collapsing, and that its potential demise bears an uncanny resemblance to the housing crisis of 2008.
From the unreliability of advertising numbers and the unregulated automation of advertising bidding wars, to the simple fact that online ads mostly fail to work, Hwang demonstrates that while consumers’ attention has never been more prized, the true value of that attention itself―much like subprime mortgages―is wildly misrepresented. And if online advertising goes belly-up, the internet―and its free services―will suddenly be accessible only to those who can afford it.
Deeply researched, convincing, and alarming, Subprime Attention Crisis will change the way you look at the internet, and its precarious future.
FSG Originals × Logic dissects the way technology functions in everyday lives. The titans of Silicon Valley, for all their utopian imaginings, never really had our best interests at recent threats to democracy, truth, privacy, and safety, as a result of tech’s reckless pursuit of progress, have shown as much. We present an alternate story, one that delights in capturing technology in all its contradictions and innovation, across borders and socioeconomic divisions, from history through the future, beyond platitudes and PR hype, and past doom and gloom. Our collaboration features four brief but provocative forays into the tech industry’s many worlds, and aspires to incite fresh conversations about technology focused on nuanced and accessible explorations of the emerging tools that reorganize and redefine life today.
176 pages, Paperback
Published October 13, 2020
About the author
Tim Hwang
7 books20 followersTim Hwang is a writer and researcher. He is the former director of the Harvard-MIT Ethics and Governance of AI Initiative, and previously served as the global public policy lead for artificial intelligence and machine learning at Google. His work has appeared in The New York Times, The Washington Post, Wired, The Atlantic, and The Wall Street Journal, among other publications. He lives in New York City.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 180 reviews
April 5, 2021
I encourage you to read this book. And not just because it is on a seriously interesting topic, but also because it is on a topic that is fairly difficult to understand and this book could easily be a textbook on how to write about difficult topics in a clear and accessible way.
I found out that I needed to read this book while reading an article in The Guardian - https://www.theguardian.com/commentis...
I was thinking before that it would be all too easy to write a review of this book that would say that the author is predicting the collapse of the internet as we know it – something he actually isn’t really doing – and to make this book sound like the equivalent of a disaster movie. What he does here is much more interesting. I know we particularly like doomsday predictions – but while he does point to a series of problems with the way the internet currently runs, and discusses the implications of what could happen if things turned out badly, this book is much more interesting than just providing a prophet of doom. It forces us to think about how and why the internet is currently structured as it is. And this structuring of the internet deserves our attention, not just because it might all suddenly collapse, but because how it is currently structured might not be how we would prefer it to be structured.
A large part of this book is based in the idea that the internet we currently have is a result of people simply don’t want to pay for the internet – at least, not directly. That means that the internet is paid for by advertising. But why would advertises want to fund the internet rather than, say, newspapers, billboards, films or magazines? Well, a very large part of that reason is because the internet offers something other media can’t offer. That is, measurable audience attention. Sites like Google, Facebook and Good Reads require you to enter information about yourself before you can access the features of these sites. Then they track every damn thing you do – often not just on their site, but on every other site too. Then they sell advertising that is directed at you, according to both the wants you know you have – given you have just searched for the best drone to buy or a reasonably priced divorce lawyer or a particularly flavoursome vanilla ice cream – or wants you might have if you were nudged hard enough, given the preferences you might not even know you might have, but that seven-out-of-ten divorced drone pilots who eat vanilla ice cream share.
We are so used to the standard ideas about the internet knowing us better than we know ourselves that we might find it hard to believe that perhaps this all-seeing eye could possibly not be nearly as all-seeing as we think. I am certainly a victim of this type of prejudice. But if this book is one thing, it is a bit of a curative.
Again, the book isn’t saying that all of our fears about capitalist surveillance have been overhyped, although, that is clearly part of the message. Rather, the author discusses the problems with the way advertising is sold on the web. One of these problems is that those who sell advertising have a clear and obvious interest in hyping the potential of their ability to provide attentive eyeballs to advertisements. But sometimes they stretch credibility. For instance, at one point in this the author mentions that Facebook promised to provide more young people of a certain age in the US than the census said existed.
Another part of the problem is that an increasing number of people are using ad-blocking software. So, even if Google knows you want to kill your granny given your recent history of search and WhatsApp messages, their ad for cheap rat poison suppliers might never end up be seen by you, even though it is exactly the right ad you need to see.
Then there is the problem of bots that flit about the internet ‘generating traffic’ but, since they aren’t conscious, they generate numbers ‘views’ of ads that aren’t really anything of a sort. At one point in this book the author says that over half of all ads on the internet might never have actually been seen by a human. And then there are ‘fat fingers’ – where you mistakenly tap an ad on your phone that you actually had no intention of tapping, except scrolling and tapping can all too easily look like the same thing.
Even so, let’s pretend you are a human and you don’t have ad-blocking software on your computer – does that mean you actually see and register the banner ads on display on your favourite website? Sometimes Facebook asks me if I’ve seen an ad by some company or other they have presumably displayed for me over the last couple of days. I never answer these requests, since it is in the least bit clear what possible benefit there could be for me in answering these questions, other than in helping a company that makes a fortune out of selling my attention to find better ways to market shit I don’t need at me. The fact is that I don’t think I’ve ever remembered seeing any of the ads they mention.
Sometimes I think this is part of the trick – you know, they are trying to see if I’m being honest with them and so they ask me about a penis enlargement ad they never actually displayed. Kind of a clever double pike with a twist – where there was no penis enlargement ad, but now I’m still thinking I need to get on to enlarging my penis anyway. Of course, penis enlargement might well be so front-of-mind that I might think I must have seen an ad for it over the last few days anyway. We all like to believe advertising only works on other people – I have to admit that I’m much more likely to believe advertising might well work best when we don’t notice we are aware of ads.
I remember hearing someone from the Evil Murdoch Empire™ saying that people (although, it is unlikely he would have referred to us as people – maybe something like ‘potentially viable though currently inactive target consumer marketing units’ – should be forced to make cups of tea and to go the toilet during television program times, rather than ad breaks, as skipping the ads is a form of theft. Clearly, the only advantage of the Trump wall along the border with Mexico is that it will be sufficiently long to save the revolution from the stress of working out which of these bastards needs to be first put up against it.
The book provides more than enough information on how advertising on the internet is sold and how it is variously manipulated by players, not least so as to push up its price. The book details how dark pools exist so that the lack of reliable information and transparency means that speculation around the price of advertising becomes inevitable and that this could eventually lead to a ‘subprime’ crisis where companies realise they are not getting the value they are paying for and then pull the plug on the gushing revenue stream that funds much of what happens on the web. That this might then make a ‘user pays’ model to fund the web inevitable. Such a shift would make the internet virtually unavailable for large numbers of current users. This would result in a massive loss of faith in the current value of advertising on the internet – but a large part of the point of this book is to call into question the basis upon which that faith is based.
The book begins with a discussion on a presentation the author saw at an advertising conference in which an academic researcher provided the results of their research showing that a lot of digital advertising was worse than useless – it’s more expensive than ‘non-targeted’ advertising, and often only impacts people who were about to buy your product anyway. But when the researcher finished his talk and asked if there were any questions, there were literally none – and the following session was yet another hype-and-spin on the joys of targeted advertising.
Like I said, even if this book proves to be wildly off in everything it predicts – I would still recommend you read it. It’s short, it’s clearly written, and it makes otherwise difficult topics clear and accessible. Even if there proves to never be a subprime attention crisis, at the very least this book will make it clear what those words mean and why seeing them strung together in that order should concern us.
I found out that I needed to read this book while reading an article in The Guardian - https://www.theguardian.com/commentis...
I was thinking before that it would be all too easy to write a review of this book that would say that the author is predicting the collapse of the internet as we know it – something he actually isn’t really doing – and to make this book sound like the equivalent of a disaster movie. What he does here is much more interesting. I know we particularly like doomsday predictions – but while he does point to a series of problems with the way the internet currently runs, and discusses the implications of what could happen if things turned out badly, this book is much more interesting than just providing a prophet of doom. It forces us to think about how and why the internet is currently structured as it is. And this structuring of the internet deserves our attention, not just because it might all suddenly collapse, but because how it is currently structured might not be how we would prefer it to be structured.
A large part of this book is based in the idea that the internet we currently have is a result of people simply don’t want to pay for the internet – at least, not directly. That means that the internet is paid for by advertising. But why would advertises want to fund the internet rather than, say, newspapers, billboards, films or magazines? Well, a very large part of that reason is because the internet offers something other media can’t offer. That is, measurable audience attention. Sites like Google, Facebook and Good Reads require you to enter information about yourself before you can access the features of these sites. Then they track every damn thing you do – often not just on their site, but on every other site too. Then they sell advertising that is directed at you, according to both the wants you know you have – given you have just searched for the best drone to buy or a reasonably priced divorce lawyer or a particularly flavoursome vanilla ice cream – or wants you might have if you were nudged hard enough, given the preferences you might not even know you might have, but that seven-out-of-ten divorced drone pilots who eat vanilla ice cream share.
We are so used to the standard ideas about the internet knowing us better than we know ourselves that we might find it hard to believe that perhaps this all-seeing eye could possibly not be nearly as all-seeing as we think. I am certainly a victim of this type of prejudice. But if this book is one thing, it is a bit of a curative.
Again, the book isn’t saying that all of our fears about capitalist surveillance have been overhyped, although, that is clearly part of the message. Rather, the author discusses the problems with the way advertising is sold on the web. One of these problems is that those who sell advertising have a clear and obvious interest in hyping the potential of their ability to provide attentive eyeballs to advertisements. But sometimes they stretch credibility. For instance, at one point in this the author mentions that Facebook promised to provide more young people of a certain age in the US than the census said existed.
Another part of the problem is that an increasing number of people are using ad-blocking software. So, even if Google knows you want to kill your granny given your recent history of search and WhatsApp messages, their ad for cheap rat poison suppliers might never end up be seen by you, even though it is exactly the right ad you need to see.
Then there is the problem of bots that flit about the internet ‘generating traffic’ but, since they aren’t conscious, they generate numbers ‘views’ of ads that aren’t really anything of a sort. At one point in this book the author says that over half of all ads on the internet might never have actually been seen by a human. And then there are ‘fat fingers’ – where you mistakenly tap an ad on your phone that you actually had no intention of tapping, except scrolling and tapping can all too easily look like the same thing.
Even so, let’s pretend you are a human and you don’t have ad-blocking software on your computer – does that mean you actually see and register the banner ads on display on your favourite website? Sometimes Facebook asks me if I’ve seen an ad by some company or other they have presumably displayed for me over the last couple of days. I never answer these requests, since it is in the least bit clear what possible benefit there could be for me in answering these questions, other than in helping a company that makes a fortune out of selling my attention to find better ways to market shit I don’t need at me. The fact is that I don’t think I’ve ever remembered seeing any of the ads they mention.
Sometimes I think this is part of the trick – you know, they are trying to see if I’m being honest with them and so they ask me about a penis enlargement ad they never actually displayed. Kind of a clever double pike with a twist – where there was no penis enlargement ad, but now I’m still thinking I need to get on to enlarging my penis anyway. Of course, penis enlargement might well be so front-of-mind that I might think I must have seen an ad for it over the last few days anyway. We all like to believe advertising only works on other people – I have to admit that I’m much more likely to believe advertising might well work best when we don’t notice we are aware of ads.
I remember hearing someone from the Evil Murdoch Empire™ saying that people (although, it is unlikely he would have referred to us as people – maybe something like ‘potentially viable though currently inactive target consumer marketing units’ – should be forced to make cups of tea and to go the toilet during television program times, rather than ad breaks, as skipping the ads is a form of theft. Clearly, the only advantage of the Trump wall along the border with Mexico is that it will be sufficiently long to save the revolution from the stress of working out which of these bastards needs to be first put up against it.
The book provides more than enough information on how advertising on the internet is sold and how it is variously manipulated by players, not least so as to push up its price. The book details how dark pools exist so that the lack of reliable information and transparency means that speculation around the price of advertising becomes inevitable and that this could eventually lead to a ‘subprime’ crisis where companies realise they are not getting the value they are paying for and then pull the plug on the gushing revenue stream that funds much of what happens on the web. That this might then make a ‘user pays’ model to fund the web inevitable. Such a shift would make the internet virtually unavailable for large numbers of current users. This would result in a massive loss of faith in the current value of advertising on the internet – but a large part of the point of this book is to call into question the basis upon which that faith is based.
The book begins with a discussion on a presentation the author saw at an advertising conference in which an academic researcher provided the results of their research showing that a lot of digital advertising was worse than useless – it’s more expensive than ‘non-targeted’ advertising, and often only impacts people who were about to buy your product anyway. But when the researcher finished his talk and asked if there were any questions, there were literally none – and the following session was yet another hype-and-spin on the joys of targeted advertising.
Like I said, even if this book proves to be wildly off in everything it predicts – I would still recommend you read it. It’s short, it’s clearly written, and it makes otherwise difficult topics clear and accessible. Even if there proves to never be a subprime attention crisis, at the very least this book will make it clear what those words mean and why seeing them strung together in that order should concern us.
March 10, 2021
When I first heard Tim Hwang speak about Subprime Attention Crisis at Logic Book Festival, he began by saying that the problem with the discourse about Big Tech is that both its champions and its critics give it too much credit. That is, the marketers overhype what the technology can do, and the rest of us are eating it right up. (Anyone who's worked at a startup gets it: Engineering is always playing catch-up to Sales's promises.) His book, then, is a case study about why taking these claims at face value is not only gullible, but a threat to the health and stability of our digital economy.
Hwang argues that the modern digital ad market is on its way to becoming a bubble, much like the subprime mortgage crisis in 2008. Today, consumers ignore ads, Facebook and Google hide behind opaque, self-serving definitions of "views" and "conversions," and everything from overt fraud (click farms) to clever accounting (Facebook's overstatement of video views) runs rampant. It feels like everyone, from the digital marketers to the industry researchers, has a perverse incentive to act like digital ads work better than they do, despite study after peer-reviewed study questioning this assumption.
Hwang is also a fantastic econ teacher—and I'm saying this as someone who had no idea how the 2008 financial crisis worked until last summer. He breaks down the programmatic ad market step by step, walks you through its similarities to the mortgage market of the late aughts, then provides numbered suggestions to reform the system—all in a tight little 178-page volume. With clever analogies (e.g. one unit of attention: one fictionalized Standardized Chicken Lot) and a clear-cut narrative structure, you won't need a pre-existing passion for two-sided market design to get something out of this read.
By the end of the book, Hwang starts describing the social consequences of having the whole Internet—from journalism to Google Maps—run on ad dollars: surveillance capitalism, addictive feeds, political polarization. Each of his carefully numbered arguments builds toward a picture of the ad industry as an increasingly precarious house of cards, so it's no surprise that he concludes that "Rather than trying to fix a broken market, we should work toward a controlled demolition that reduces its influence in the long run." Hwang ends by proposing a few of these possible reforms, such as industry-independent research organizations, mandatory disclosures for vendors of programmatic ads, and legal teeth to punish frauds.
Overall, I found Subprime Attention Crisis persuasive, if mildly hyperbolic: the narrow reforms proposed in the last few pages don't seem to match the urgency of the the crisis-like descriptions of the first three-quarters. In my opinion, the bigger challenge of the next 5 or 10 years—and what I'm most interested in asking Hwang about next Thursday—is how we'll actually wean the web off ads and toward something more sustainable.
reviewed for the reboot newsletter - come to our q&a on the book on 3/18!
Hwang argues that the modern digital ad market is on its way to becoming a bubble, much like the subprime mortgage crisis in 2008. Today, consumers ignore ads, Facebook and Google hide behind opaque, self-serving definitions of "views" and "conversions," and everything from overt fraud (click farms) to clever accounting (Facebook's overstatement of video views) runs rampant. It feels like everyone, from the digital marketers to the industry researchers, has a perverse incentive to act like digital ads work better than they do, despite study after peer-reviewed study questioning this assumption.
Hwang is also a fantastic econ teacher—and I'm saying this as someone who had no idea how the 2008 financial crisis worked until last summer. He breaks down the programmatic ad market step by step, walks you through its similarities to the mortgage market of the late aughts, then provides numbered suggestions to reform the system—all in a tight little 178-page volume. With clever analogies (e.g. one unit of attention: one fictionalized Standardized Chicken Lot) and a clear-cut narrative structure, you won't need a pre-existing passion for two-sided market design to get something out of this read.
By the end of the book, Hwang starts describing the social consequences of having the whole Internet—from journalism to Google Maps—run on ad dollars: surveillance capitalism, addictive feeds, political polarization. Each of his carefully numbered arguments builds toward a picture of the ad industry as an increasingly precarious house of cards, so it's no surprise that he concludes that "Rather than trying to fix a broken market, we should work toward a controlled demolition that reduces its influence in the long run." Hwang ends by proposing a few of these possible reforms, such as industry-independent research organizations, mandatory disclosures for vendors of programmatic ads, and legal teeth to punish frauds.
Overall, I found Subprime Attention Crisis persuasive, if mildly hyperbolic: the narrow reforms proposed in the last few pages don't seem to match the urgency of the the crisis-like descriptions of the first three-quarters. In my opinion, the bigger challenge of the next 5 or 10 years—and what I'm most interested in asking Hwang about next Thursday—is how we'll actually wean the web off ads and toward something more sustainable.
reviewed for the reboot newsletter - come to our q&a on the book on 3/18!
July 14, 2020
The internet has morphed through numerous stages to get to the current advertising model. Tim Hwang says that global model is so corrupt and corrupted it looks like the stock markets just before the Financial Crisis and recession of 2008. “The whole edifice of online advertising…is bunk,” he says. In his Subprime Attention Crisis, he explores the many ways it doesn’t work.
The whole raison d’être of the world wide web seems to be personal data collection and surveillance. Hwang says the data collected is laughable. Ten percent of the records have some data that is wrong, and as many as 85% are just plain wrong altogether. In a world where garbage in means garbage out, decisions made on this data are worthless at best. Yet they are the pride of the internet: it’s shining difference.
It was supposed to be so highly targeted there would be no waste. For users, it meant ads speaking directly to them, based on sites they had visited, things they had purchased, and chats they had posted. For marketers, it meant finally knowing which half of the money they were spending on advertising was wasted, while improving the closing of sales because no effort was being inefficient. Turns out almost all the money spent on website ads is wasted. Users are not clicking or buying, and pinpoint targeting fails repeatedly.
Compounding this disaster is the structure of ad placement itself, which has evolved from direct negotiation with various sites, to middlemen placing the ads for a fee. The problem is those middlemen are raking off a 70% cut, leaving the site with just 30% of the revenue from showing ads on their services. Sooner or later, Hwang says, both advertisers and their host sites will wake up to the ripoff. The result might be the complete destruction of the model and a whole new beginning for the internet.
The way it works now is that two kinds of middlemen meet to place ads. One aggregator represents the sell side, with an inventory of empty slots on various sites. On the buy side are the advertisers who want pinpoint placement to take advantage of the internet’s supposed power to target individuals precisely.
When someone lands on a page, the buy side and the sell side get together in an instant auction, placing some client firm’s ad on that page. The firm bought into that placement being precisely what it wanted: the right user landing in the right spot, interested and ready to purchase the firm’s product.
Nothing could be further from the truth, and sales results prove it daily.
Not only is the personal data wrong, but far from being of interest to the user, ads are all but totally ignored. When they began in the id-90s, ads were so new, people clicked on them out of curiosity. The clickthrough rate was an astonishing 44%, convincing many firms that the internet was where to place their marketing dollars. Today, the clickthrough rate less than one half of one percent, making it a worse buy that most other media. People are sick of ads, don’t want the distraction of clicking away from what they were doing and often resent the ads that come closest to actually being of importance to them. It is too creepy to have an ad know exactly what you want.
Meanwhile, back in the process itself, marketers’ ads end up costing far more than they thought they would in the instant auction process. They pay more and get less exposure than they paid for. Some ads get placed at the bottom of the page, where the user might never scroll to before clicking away. As many as 56% of online ads are never seen by a human, just because of bad placement. Some end up on pages that are merely there to collect ad revenue and have no content. Some are fraudulent clicks made by bots. And of course, both bots and paid humans click away on the ads themselves to boost the (fake) results and thereby convince the sucker, er, marketer, to keep buying that space on that page. More than half of inventory is lost to click fraud or unviewable inventory. Hwang says it is “a massive global economy of fraud in the programmatic advertising marketplace. “ 51
But wait! There’s more. Users have taken matters into their own hands too. Free ad-blocking software has been installed on hundreds of millions of devices. Browsers themselves now come equipped with automatic ad blockers. Estimates are that 75% of North Americans use some sort of ad blocking, so they never see an ad at all.
How long before the marketers wake up? How long before they admit their ads are not being seen, never mind being clicked by a highly reliable and curated potential customer? How long before the marketers rebel at the absurdly high cost per ad in the totally opaque and untraceable process of purchasing an ad space? What could possibly go wrong with this infrastructure setup?
While consumers might rejoice at the ripoff of advertisers and their failure to increase sales or even be noticed at all, there is a very dark side to the joy. The whole web has arisen on the back of the ad model. Everything that is free on the internet is supported by revenue from ads. Without the marketers ‘ money greasing their engines – even if it is just 30% of the final auction price – millions of services worldwide will quickly starve to death.
Hwang uses the buy and sell siders’ own words to show how closely the ad system tacks to the stock markets. He quotes numerous firms’ executives describing their auction systems as paralleling the New York Stock Exchange or the NASDAQ markets. It seems they are just as blind and vulnerable as those markets, which took a decade to recover from the Financial Crisis they allowed to overtake them in 2008. But in this case, they might not come back at all. They are not institutions, but weak new companies with shallow pockets and no roots.
It will mean a completely different kind of World Wide Web going forward. It might be subscription based, or perhaps services will be entirely sponsored by one firm. Or something completely different. The main point is a huge disaster is looming. No one is preparing for it, and no one knows what will replace the current system. This portends turmoil like the web has never experienced.
If you had to pay for twitter, would you? If you had to pay for every Google search, would you? Of the dozens of free services you now use, how many could you even afford to pay for?
So don’t get comfortable with your lovely set of highly focused bookmarks. All those sites are as fragile as the newspapers and magazines they so easily displaced. The internet as we know it might have the shortest lifespan of any medium to date.
David Wineberg
Author, The Straight Dope, or What I learned from my first thousand nonfiction reviews
The whole raison d’être of the world wide web seems to be personal data collection and surveillance. Hwang says the data collected is laughable. Ten percent of the records have some data that is wrong, and as many as 85% are just plain wrong altogether. In a world where garbage in means garbage out, decisions made on this data are worthless at best. Yet they are the pride of the internet: it’s shining difference.
It was supposed to be so highly targeted there would be no waste. For users, it meant ads speaking directly to them, based on sites they had visited, things they had purchased, and chats they had posted. For marketers, it meant finally knowing which half of the money they were spending on advertising was wasted, while improving the closing of sales because no effort was being inefficient. Turns out almost all the money spent on website ads is wasted. Users are not clicking or buying, and pinpoint targeting fails repeatedly.
Compounding this disaster is the structure of ad placement itself, which has evolved from direct negotiation with various sites, to middlemen placing the ads for a fee. The problem is those middlemen are raking off a 70% cut, leaving the site with just 30% of the revenue from showing ads on their services. Sooner or later, Hwang says, both advertisers and their host sites will wake up to the ripoff. The result might be the complete destruction of the model and a whole new beginning for the internet.
The way it works now is that two kinds of middlemen meet to place ads. One aggregator represents the sell side, with an inventory of empty slots on various sites. On the buy side are the advertisers who want pinpoint placement to take advantage of the internet’s supposed power to target individuals precisely.
When someone lands on a page, the buy side and the sell side get together in an instant auction, placing some client firm’s ad on that page. The firm bought into that placement being precisely what it wanted: the right user landing in the right spot, interested and ready to purchase the firm’s product.
Nothing could be further from the truth, and sales results prove it daily.
Not only is the personal data wrong, but far from being of interest to the user, ads are all but totally ignored. When they began in the id-90s, ads were so new, people clicked on them out of curiosity. The clickthrough rate was an astonishing 44%, convincing many firms that the internet was where to place their marketing dollars. Today, the clickthrough rate less than one half of one percent, making it a worse buy that most other media. People are sick of ads, don’t want the distraction of clicking away from what they were doing and often resent the ads that come closest to actually being of importance to them. It is too creepy to have an ad know exactly what you want.
Meanwhile, back in the process itself, marketers’ ads end up costing far more than they thought they would in the instant auction process. They pay more and get less exposure than they paid for. Some ads get placed at the bottom of the page, where the user might never scroll to before clicking away. As many as 56% of online ads are never seen by a human, just because of bad placement. Some end up on pages that are merely there to collect ad revenue and have no content. Some are fraudulent clicks made by bots. And of course, both bots and paid humans click away on the ads themselves to boost the (fake) results and thereby convince the sucker, er, marketer, to keep buying that space on that page. More than half of inventory is lost to click fraud or unviewable inventory. Hwang says it is “a massive global economy of fraud in the programmatic advertising marketplace. “ 51
But wait! There’s more. Users have taken matters into their own hands too. Free ad-blocking software has been installed on hundreds of millions of devices. Browsers themselves now come equipped with automatic ad blockers. Estimates are that 75% of North Americans use some sort of ad blocking, so they never see an ad at all.
How long before the marketers wake up? How long before they admit their ads are not being seen, never mind being clicked by a highly reliable and curated potential customer? How long before the marketers rebel at the absurdly high cost per ad in the totally opaque and untraceable process of purchasing an ad space? What could possibly go wrong with this infrastructure setup?
While consumers might rejoice at the ripoff of advertisers and their failure to increase sales or even be noticed at all, there is a very dark side to the joy. The whole web has arisen on the back of the ad model. Everything that is free on the internet is supported by revenue from ads. Without the marketers ‘ money greasing their engines – even if it is just 30% of the final auction price – millions of services worldwide will quickly starve to death.
Hwang uses the buy and sell siders’ own words to show how closely the ad system tacks to the stock markets. He quotes numerous firms’ executives describing their auction systems as paralleling the New York Stock Exchange or the NASDAQ markets. It seems they are just as blind and vulnerable as those markets, which took a decade to recover from the Financial Crisis they allowed to overtake them in 2008. But in this case, they might not come back at all. They are not institutions, but weak new companies with shallow pockets and no roots.
It will mean a completely different kind of World Wide Web going forward. It might be subscription based, or perhaps services will be entirely sponsored by one firm. Or something completely different. The main point is a huge disaster is looming. No one is preparing for it, and no one knows what will replace the current system. This portends turmoil like the web has never experienced.
If you had to pay for twitter, would you? If you had to pay for every Google search, would you? Of the dozens of free services you now use, how many could you even afford to pay for?
So don’t get comfortable with your lovely set of highly focused bookmarks. All those sites are as fragile as the newspapers and magazines they so easily displaced. The internet as we know it might have the shortest lifespan of any medium to date.
David Wineberg
Author, The Straight Dope, or What I learned from my first thousand nonfiction reviews
December 25, 2023
really interesting book
October 31, 2020
I have an insider perspective on many of the things this book discusses. I run the data science department for a digital publisher. Among other things, I'm in charge of evaluating programmatic networks that we might want to use. I've come to the conclusion that, while there are good actors out there, much of the inventory on programmatic exchanges is rubbish. We're careful to work only with the good guys, but many publishers and marketers don't have the resources or knowhow to navigate this complicated landscape. As a result, enormous amounts of ad spend are wasted on inaccurate targeting, opaque pricing structures, and outright fraud.
Subprime Attention Crisis is the most complete and coherent argument I've seen about the state of advertising today. What would happen if the advertising marketplace that supports the internet's culture of free services suddenly imploded? The answer is chilling.
Subprime Attention Crisis is the most complete and coherent argument I've seen about the state of advertising today. What would happen if the advertising marketplace that supports the internet's culture of free services suddenly imploded? The answer is chilling.
March 11, 2023
I thought this book would be about how social media destroys focus and attention. Actually, it’s about how weak the advertising system of surveillance capitalism is and how it could risk destroying the internet. Not entirely convinced and frankly, not sure if I care.
January 27, 2021
Tim took a 1,200 word Forbes guest contributor article and stretched it out into 100+ pages by adopting a writing strategy of repeating everything else he's said before introducing a new idea and them summarizing everything else he's said plus the new idea before going onto repeat everything again at the start of the next chapter before then adding another new idea.
Also, let's pretend he just wrote the 1,200 word essay. As someone who works in the programmatic world I found this essay useless. Yes, the digital ads world is far from perfect but stocks and programmatic ads should not be compared. Hiring a sales person with an opaque track record to get you new customers is not akin to storing your extra capital in hopes for a return.
Anyways... I would keep ranting but this shit book isn't worth my time. Tim & FSG should both be embarrassed by the quality of this work.
Also, let's pretend he just wrote the 1,200 word essay. As someone who works in the programmatic world I found this essay useless. Yes, the digital ads world is far from perfect but stocks and programmatic ads should not be compared. Hiring a sales person with an opaque track record to get you new customers is not akin to storing your extra capital in hopes for a return.
Anyways... I would keep ranting but this shit book isn't worth my time. Tim & FSG should both be embarrassed by the quality of this work.
December 4, 2020
There is nothing particularly wrong with this book, but it is essentially a 3.5 star online long-form news article stretched over 120+ pages. The idea is intriguing and although I have long expected something fishy is afoot with the profitability of companies like Facebook and Google, this was the first time I have seen it approached in the public sphere. I would be more than happy to pay a monthly subscription for general internet use + email in exchange for no (or greatly reduced) ads or spam, but I think the author's other suggested conclusion is more likely - that an implosion of the digital ad economy will just result in a rebuilding of the same type of digital ad economy once the dust settles.
December 11, 2020
I love a book that makes a claim that in five years time will be either eerily spot on or completely off the mark.
October 4, 2020
Review originally published 4 Oct. 2020 at Falling Letters. I receivd a free copy from the publisher via Netgalley.
Subprime Attention Crisis is one of four titles launching on Tuesday October 13 that comprise a collaboration between publishing imprint FSG Originals and tech magazine Logic. To quote the blog post announcement, “the four books in the FSGO/Logic series are brief but provocative forays into the tech industry’s many worlds, inciting fresh conversations focused on nuanced and accessible explorations of the emerging tools reorganizing and redefining life today”.
Regular readers know such topics are a far cry from the usual content reviewed on my blog. But this title caught my eye as I browsed Netgalley. Internet advertising has a significant impact on anyone reading this. We all see it (even when we diligently use an adblocker). Maybe you have advertising on your blog. Most importantly: we all depend on it as the foundation that supports how the Internet currently functions.
Hwang argues that ineffective internet advertising has built up around inaccurate and unreliable metrics. I thought I knew the basics of how internet advertising works. I learnt a lot more about it from this book. These metrics, along with other factors that Hwang explores, mean the system is poised to fail as the markets did in 2008. As the title implies, Hwang draws comparisons throughout the book to the subprime mortgage crisis. Internet advertising faces a ‘subprime attention crisis’. Why should we care about the potential failure of online advertising, which for most of us is little more than an irritant? Because it’s currently the key that allows us to freely access websites and the information contained within. Hwang concludes with broad suggestions of how the industry might anticipate such a failure and begin to mitigate it.
The nature of Hwang’s subject lends itself to mildly technical language. It reminded me of reading textbooks in university. I can understand the topic generally, but I need to focus to really comprehend what he’s telling me. Finally, I understand the 2008 mortgage crisis, haha. I had to look it up in order to understand Hwang’s comparison.
💭 The Bottom Line: Subprime Attention Crisis may require focused attention from a layperson to comprehend, but its relatively short length and important topic make it a good read for anyone who cares about the long term sustainability of the Internet as we know it.
Subprime Attention Crisis is one of four titles launching on Tuesday October 13 that comprise a collaboration between publishing imprint FSG Originals and tech magazine Logic. To quote the blog post announcement, “the four books in the FSGO/Logic series are brief but provocative forays into the tech industry’s many worlds, inciting fresh conversations focused on nuanced and accessible explorations of the emerging tools reorganizing and redefining life today”.
Regular readers know such topics are a far cry from the usual content reviewed on my blog. But this title caught my eye as I browsed Netgalley. Internet advertising has a significant impact on anyone reading this. We all see it (even when we diligently use an adblocker). Maybe you have advertising on your blog. Most importantly: we all depend on it as the foundation that supports how the Internet currently functions.
Hwang argues that ineffective internet advertising has built up around inaccurate and unreliable metrics. I thought I knew the basics of how internet advertising works. I learnt a lot more about it from this book. These metrics, along with other factors that Hwang explores, mean the system is poised to fail as the markets did in 2008. As the title implies, Hwang draws comparisons throughout the book to the subprime mortgage crisis. Internet advertising faces a ‘subprime attention crisis’. Why should we care about the potential failure of online advertising, which for most of us is little more than an irritant? Because it’s currently the key that allows us to freely access websites and the information contained within. Hwang concludes with broad suggestions of how the industry might anticipate such a failure and begin to mitigate it.
The nature of Hwang’s subject lends itself to mildly technical language. It reminded me of reading textbooks in university. I can understand the topic generally, but I need to focus to really comprehend what he’s telling me. Finally, I understand the 2008 mortgage crisis, haha. I had to look it up in order to understand Hwang’s comparison.
💭 The Bottom Line: Subprime Attention Crisis may require focused attention from a layperson to comprehend, but its relatively short length and important topic make it a good read for anyone who cares about the long term sustainability of the Internet as we know it.
December 30, 2023
I knew this wasn’t going to go well on page 3 when the Author said the reason social media is free is advertising. Wrong. Social media is free because we are the product and they’re data mining and selling our information.
October 3, 2021
I will never forget the new hire orientation on my first day of work at [An Unnamed Major Tech Company] back in 2014. During orientation, a popular executive extolled the virtues of targeted online advertising, both for businesses and for the people lucky enough to see these amazing targeted ads.
“We can show people ads for things that are actually relevant to them!” the executive declared. “Wouldn’t you prefer to see ads for things you actually want?”
“No,” I remember thinking, “I would rather just not see ads at all.”
But this idea, that people hate ads and don’t want them invading every moment of their consciousness, is never part of the discourse at places like [An Unnamed Major Tech Company]. It is taken as gospel that advertising is good, and that carefully targeted advertising is even better.
This book tries to dispel those assumptions–and many more that dominate the world of online advertising. And if what this book says is true, these are some very troubling assumptions indeed. For one thing, online advertising apparently reached peak engagement with banner ads in the 1990s and hasn’t gotten anywhere close to that since. Turns out most people just don’t want to be bothered by ads. Shocking, right?
For another thing, highly targeted advertising may not even be very effective at two of its main objectives: (1) getting eyeballs on it and (2) getting people to buy stuff. From the ubiquity of ad blocking software to the rise of click farms and ad fraud, billions of dollars in ads are bought, paid for, and seen by no one at all. This coupled with the pervasive opacity of how targeted online ads work, how they are priced, and where they appear means that the whole digital ads ecosystem, according to this book, is rife with abuse and poised to pop in a big (and very bad) way.
The author draws compelling parallels between the state of online ads today and the state of mortgages in 2008, another bubble that popped in a big and bad way. I don’t know enough about this topic to dispute any of the author’s claims. Perhaps he is wrong and things aren���t as calamitous as they seem. I can’t really argue this. But I read a similarly engaging and well-researched article on this topic by Cory Doctorow a few years ago. There is a growing chorus of voices warning that online ads are a precarious bubble waiting to burst—and take a good chunk of the economy down with them. This book definitely felt like another “oh shit” moment in a steady stream of “oh shit” moments that made me want to sell my tech stock and diversify my portfolio before stuff hits the fan.
If we accept that online ads are not as effective as they seem, and in fact their adoption and pervasiveness are mostly driven by hot air and hype, how do we prevent them from tanking and causing another recession a la 2008? The author calls for better regulations and transparency around the cost and effectiveness of online ads, so that advertisers aren’t paying unpredictable prices for some vague thing. He also advocates for more independent research into the ads industry, again aimed at increasing transparency and knowledge about the actual value of online ads so they’re priced in accordance with what they are actually worth.
These are both perfectly okay solutions, I suppose, but they didn’t get me super fired up about the possibility of actual change. I have a feeling this is a bubble that is going to explode before regulators get around to doing anything about it, as so many people have a huge economic stake in pretending that targeted advertising is the best thing since antibiotics, even if it performs markedly worse than a banner ad from 1997. For my own part, I would prefer an internet that is completely free of ads, even though, as the author points out, that would make many services unaffordable for most people. I’d like to see more of the internet be publicly funded as a public good, rather than propped up by hungry capitalists looking to sell things. Perhaps the author is right that a “controlled demolition” of the digital ads market can eventually give rise to a more equitable and transparent internet that isn’t choked by ads. But I’m skeptical this will happen, and I would have liked the author to explore in more detail what an internet without advertising could look like.
Overall, this was a great book and a very quick read. Short, sweet, compellingly argued, and gave me a lot to think about.
“We can show people ads for things that are actually relevant to them!” the executive declared. “Wouldn’t you prefer to see ads for things you actually want?”
“No,” I remember thinking, “I would rather just not see ads at all.”
But this idea, that people hate ads and don’t want them invading every moment of their consciousness, is never part of the discourse at places like [An Unnamed Major Tech Company]. It is taken as gospel that advertising is good, and that carefully targeted advertising is even better.
This book tries to dispel those assumptions–and many more that dominate the world of online advertising. And if what this book says is true, these are some very troubling assumptions indeed. For one thing, online advertising apparently reached peak engagement with banner ads in the 1990s and hasn’t gotten anywhere close to that since. Turns out most people just don’t want to be bothered by ads. Shocking, right?
For another thing, highly targeted advertising may not even be very effective at two of its main objectives: (1) getting eyeballs on it and (2) getting people to buy stuff. From the ubiquity of ad blocking software to the rise of click farms and ad fraud, billions of dollars in ads are bought, paid for, and seen by no one at all. This coupled with the pervasive opacity of how targeted online ads work, how they are priced, and where they appear means that the whole digital ads ecosystem, according to this book, is rife with abuse and poised to pop in a big (and very bad) way.
The author draws compelling parallels between the state of online ads today and the state of mortgages in 2008, another bubble that popped in a big and bad way. I don’t know enough about this topic to dispute any of the author’s claims. Perhaps he is wrong and things aren���t as calamitous as they seem. I can’t really argue this. But I read a similarly engaging and well-researched article on this topic by Cory Doctorow a few years ago. There is a growing chorus of voices warning that online ads are a precarious bubble waiting to burst—and take a good chunk of the economy down with them. This book definitely felt like another “oh shit” moment in a steady stream of “oh shit” moments that made me want to sell my tech stock and diversify my portfolio before stuff hits the fan.
If we accept that online ads are not as effective as they seem, and in fact their adoption and pervasiveness are mostly driven by hot air and hype, how do we prevent them from tanking and causing another recession a la 2008? The author calls for better regulations and transparency around the cost and effectiveness of online ads, so that advertisers aren’t paying unpredictable prices for some vague thing. He also advocates for more independent research into the ads industry, again aimed at increasing transparency and knowledge about the actual value of online ads so they’re priced in accordance with what they are actually worth.
These are both perfectly okay solutions, I suppose, but they didn’t get me super fired up about the possibility of actual change. I have a feeling this is a bubble that is going to explode before regulators get around to doing anything about it, as so many people have a huge economic stake in pretending that targeted advertising is the best thing since antibiotics, even if it performs markedly worse than a banner ad from 1997. For my own part, I would prefer an internet that is completely free of ads, even though, as the author points out, that would make many services unaffordable for most people. I’d like to see more of the internet be publicly funded as a public good, rather than propped up by hungry capitalists looking to sell things. Perhaps the author is right that a “controlled demolition” of the digital ads market can eventually give rise to a more equitable and transparent internet that isn’t choked by ads. But I’m skeptical this will happen, and I would have liked the author to explore in more detail what an internet without advertising could look like.
Overall, this was a great book and a very quick read. Short, sweet, compellingly argued, and gave me a lot to think about.
June 17, 2020
In Subprime Attention Crisis, Tim Hwang initiates an upsetting conversation about not only the relationship of social media to advertising (which latter supports social media and causes it to remain free) but about the possibility that the many free services that enable internet users to explore products and news were suddenly behind a paywall. It could happen. If advertising is contingent upon the public having free access, how far down would sales be driven by the abrupt necessity to pay for each service, much as one pays for TV programming services? It does not take too many pages--or paragraphs--for the author to draw the reader's attention to the fact that the online advertising edifice is a wobbly thing. The description of the algorithm-driven process by which programmatic advertising takes place is very damaging to the confidence of anyone counting on such advertising to boost sales.. Hwang's description, best suited to understanding by familiarized online advertising experts, goes some distance in helping readers grasp the (lack of) logic behind the advertisements they see. This book seems to be an important study in online marketing and its interconnection with the rest of the global economy. Marketing teachers would do well to consider choosing Subprime Attention Crisis for their classes, if for nothing else than critical thinking and planning. #SubprimeAttentionCrisis #NetGalley
January 3, 2022
Read this book if you are interested in programmatic advertising and the business models of the internet.
Hwang delivers a clear and concise argument for the notion that the programmatic advertising economy is in a growing bubble, comparing it to the financial crashes of 1929 and 2008.
Ads are often mismeasured, not viewed or not effective. Fraud happens all the time. Everyone knows this, but the market is so intransparent that advertisers still pay inflated prices, while middlemen and platforms profit.
The ad market is complex, so the book would have benefitted from a few diagrams here and there, but overall it's understandable and easy to read. It's written from a clear Western perspective, I wonder if things are similar in China.
If the ad bubble bursts, this will have ripple effects across the entire digital economy. We should prevent a destructive burst by introducing smart regulation and developing alternative business models. Read this book if you care about that kind of stuff!
Hwang delivers a clear and concise argument for the notion that the programmatic advertising economy is in a growing bubble, comparing it to the financial crashes of 1929 and 2008.
Ads are often mismeasured, not viewed or not effective. Fraud happens all the time. Everyone knows this, but the market is so intransparent that advertisers still pay inflated prices, while middlemen and platforms profit.
The ad market is complex, so the book would have benefitted from a few diagrams here and there, but overall it's understandable and easy to read. It's written from a clear Western perspective, I wonder if things are similar in China.
If the ad bubble bursts, this will have ripple effects across the entire digital economy. We should prevent a destructive burst by introducing smart regulation and developing alternative business models. Read this book if you care about that kind of stuff!
February 13, 2021
4-star writing, 6-star creative idea for basing the book entirely on an extended metaphor (the 2008 subprime mortgage crisis as an analogue to the coming reckoning of “Hey, paid digital ads aren’t even CLOSE to as valuable as we thought they were!”)
Bonus points for an awesome book cover and the decision “let’s keep this to 150 pages instead of beating a dead horse for 350”
Bonus points for an awesome book cover and the decision “let’s keep this to 150 pages instead of beating a dead horse for 350”
October 5, 2023
Programmatic advertising is the main funding model of the modern internet, and author Tim Hwang thinks it’s headed for a crisis.
Although, the web could have developed in different ways, most of the services we take for granted are funded primarily through some type of ad-driven model. From Google’s garden of services, to Facebook, Twitter, Reddit, blogs, recipe sites, news, and on and on; all of these exist, mostly for free, due to advertising.
Every time you open a new web page, the gears of the programmatic advertising machine start turning. In the span of milliseconds, an auction is held through digital exchanges, multiple advertisers bid for the right to display something, and the winning bid has their ad delivered from their servers to your screen. All in the hope of capturing a bit of your attention.
However, the programmatic advertising market has inherent flaws, similar to the financial markets they were modelled after. There is: a large amount of fraud, where people don’t have clarity on whether or not what they purchased is what they got; a large amount of opacity, where participants have little clarity on the pricing of assets (ad space), and that the value of those assets (i.e. the ability to capture attention) are both not as effective as we currently think (due to metrics being inflated) and eroding in value over time (due to ad blocking).
While all this is happening, the global advertising market funnels greater shares of their ad dollars into programmatic advertising due to the declining attention towards traditional media.
This is creating a bubble. A subprime bubble that could burst, similar to what happened in 2008 with subprime mortgages, and radically change the face of the internet. What happens if we can no longer rely on advertising to fund a mostly free, mostly open internet?
The argument Hwang spells out here is compelling, and highly readable. He has no issues bringing to light the highly technical world of programmatic advertising in a way that feels approachable. And he convincingly builds his argument with comparisons to the structure of our existing financial markets and their own flaws.
But it’s interesting reading this book a few years later. It’s not that programmatic advertising has seen a dramatic decline in the way that Hwang warns about, but the soft-landing he describes, the one in which the value of programmatic advertising slowly declines and replacement models are found, is something that I think has been happening.
We can see it in the fact that the market for SaaS has exploded over the last decade; in the growth in online subscription services like Spotify, Netflix et al; in the increasing effort to put news behind paywalls from the NYTimes to newer platforms like Substack, where it feels like almost every news site is unreadable unless you’re a paid subscriber; and even in the fact that some marketers are shifting dollars back to traditional ads (putting dollars in things like podcast ads, product placements etc.)
Going more technical, having worked on data teams adjacent or directly working on these problems, there’s an increasing emphasis on counterfactual and uplift estimation (would X have purchased if I hadn’t spent this ad dollar anyway?) in the industry to more accurately understand the value of individual ads. Creating tech that funnels a narrower, but more effective, slice of ad dollars into campaigns that actually work.
All of these are signals that Hwang was right, the long standing funding model of the internet doesn’t work, but we’re already seeing the shift in almost every corner of the internet. More services are going behind paywalls, more dollars are going back into traditional marketing channels, and the dollars that are going into programmatic advertising are going down in aggregate but increasing in effectiveness.
With all these trends occurring, it means the dramatic bubble burst he predicts seems unlikely to happen. Everyone in the market already recognize many of these problems, and are working out new models to fix them. That’s not to say this book was wrong or out of date. Much of it is still relevant. But the shifts Hwang predicts are coming about as we speak. It remains to be seen what this ultimately holds for the future of life online. I guess we’ll have to wait and see.
Although, the web could have developed in different ways, most of the services we take for granted are funded primarily through some type of ad-driven model. From Google’s garden of services, to Facebook, Twitter, Reddit, blogs, recipe sites, news, and on and on; all of these exist, mostly for free, due to advertising.
Every time you open a new web page, the gears of the programmatic advertising machine start turning. In the span of milliseconds, an auction is held through digital exchanges, multiple advertisers bid for the right to display something, and the winning bid has their ad delivered from their servers to your screen. All in the hope of capturing a bit of your attention.
However, the programmatic advertising market has inherent flaws, similar to the financial markets they were modelled after. There is: a large amount of fraud, where people don’t have clarity on whether or not what they purchased is what they got; a large amount of opacity, where participants have little clarity on the pricing of assets (ad space), and that the value of those assets (i.e. the ability to capture attention) are both not as effective as we currently think (due to metrics being inflated) and eroding in value over time (due to ad blocking).
While all this is happening, the global advertising market funnels greater shares of their ad dollars into programmatic advertising due to the declining attention towards traditional media.
This is creating a bubble. A subprime bubble that could burst, similar to what happened in 2008 with subprime mortgages, and radically change the face of the internet. What happens if we can no longer rely on advertising to fund a mostly free, mostly open internet?
The argument Hwang spells out here is compelling, and highly readable. He has no issues bringing to light the highly technical world of programmatic advertising in a way that feels approachable. And he convincingly builds his argument with comparisons to the structure of our existing financial markets and their own flaws.
But it’s interesting reading this book a few years later. It’s not that programmatic advertising has seen a dramatic decline in the way that Hwang warns about, but the soft-landing he describes, the one in which the value of programmatic advertising slowly declines and replacement models are found, is something that I think has been happening.
We can see it in the fact that the market for SaaS has exploded over the last decade; in the growth in online subscription services like Spotify, Netflix et al; in the increasing effort to put news behind paywalls from the NYTimes to newer platforms like Substack, where it feels like almost every news site is unreadable unless you’re a paid subscriber; and even in the fact that some marketers are shifting dollars back to traditional ads (putting dollars in things like podcast ads, product placements etc.)
Going more technical, having worked on data teams adjacent or directly working on these problems, there’s an increasing emphasis on counterfactual and uplift estimation (would X have purchased if I hadn’t spent this ad dollar anyway?) in the industry to more accurately understand the value of individual ads. Creating tech that funnels a narrower, but more effective, slice of ad dollars into campaigns that actually work.
All of these are signals that Hwang was right, the long standing funding model of the internet doesn’t work, but we’re already seeing the shift in almost every corner of the internet. More services are going behind paywalls, more dollars are going back into traditional marketing channels, and the dollars that are going into programmatic advertising are going down in aggregate but increasing in effectiveness.
With all these trends occurring, it means the dramatic bubble burst he predicts seems unlikely to happen. Everyone in the market already recognize many of these problems, and are working out new models to fix them. That’s not to say this book was wrong or out of date. Much of it is still relevant. But the shifts Hwang predicts are coming about as we speak. It remains to be seen what this ultimately holds for the future of life online. I guess we’ll have to wait and see.
July 23, 2023
Full review on my blog: https://blog.siddharthkannan.in/2023/...
What would happen if all the free services that you use on the Internet, which are powered by advertisements, stop being free one day? This is the premise of the Subprime Attention Crisis. If you ask yourself this, you might realize that many of the services that we think of as free are powered by advertising. For me, the most frequently used services that are ostensible free are Google Maps, WhatsApp and YouTube. I use at least one of them almost every day. If they were to become paid services, or the hurdles to using them without giving up too much data increased, it would be mildly annoying. This calculus will be very different for an Internet user who uses Gmail and uses it to receive important communication. Hwang’s argument is sensible and easy to understand. He starts from the basics of advertising on the Internet and builds up to his myriad theses: Programmatic advertising is very similar to the financial markets. Ad networks claims that targeted advertising on the Internet is better than “spray everyone” advertising on TV. This claim is a lie and that banner ads don’t really change consumer behavior. Commercial interruptions are blocked by users using ad blocking plugins or because users reliably skip ads on video-only platforms in under half a second. To back all of this, he presents a lot of industry research and anecdotal evidence. This was a convincing case for being aware that free services could stop being free any day, and there would be nothing really surprising about it. (Just the other day, YouTube took a step towards blocking ad blockers.)
What would happen if all the free services that you use on the Internet, which are powered by advertisements, stop being free one day? This is the premise of the Subprime Attention Crisis. If you ask yourself this, you might realize that many of the services that we think of as free are powered by advertising. For me, the most frequently used services that are ostensible free are Google Maps, WhatsApp and YouTube. I use at least one of them almost every day. If they were to become paid services, or the hurdles to using them without giving up too much data increased, it would be mildly annoying. This calculus will be very different for an Internet user who uses Gmail and uses it to receive important communication. Hwang’s argument is sensible and easy to understand. He starts from the basics of advertising on the Internet and builds up to his myriad theses: Programmatic advertising is very similar to the financial markets. Ad networks claims that targeted advertising on the Internet is better than “spray everyone” advertising on TV. This claim is a lie and that banner ads don’t really change consumer behavior. Commercial interruptions are blocked by users using ad blocking plugins or because users reliably skip ads on video-only platforms in under half a second. To back all of this, he presents a lot of industry research and anecdotal evidence. This was a convincing case for being aware that free services could stop being free any day, and there would be nothing really surprising about it. (Just the other day, YouTube took a step towards blocking ad blockers.)
January 3, 2021
This whole book basically argues that programmatic advertising, on which the entire internet economy is based, is a bubble growing larger and larger as it prepares to pop. The ways in which the advertising industry today mirrors Wall Street before the Great Recession are compelling. If Hwang is right, we'll be talking about this book a lot a decade from now. If he's right, the internet as it is now might be unrecognizable in a decade. If he's right, we're in for a doozy of a time. And if what he predicts does not come to pass, it'll be because efforts were made to stop it.
One thing is for certain: I'm not sure I grasped the reliance of the internet as we know it now on ads before I read this book. And another: we should decide on the future of the internet together. Instead, it increasingly feels as though it's left to Google and Facebook, and we just have to deal with it.
One thing is for certain: I'm not sure I grasped the reliance of the internet as we know it now on ads before I read this book. And another: we should decide on the future of the internet together. Instead, it increasingly feels as though it's left to Google and Facebook, and we just have to deal with it.
November 14, 2020
Får fire stjerner fordi det er en veldig interessant problemstilling (er digital annonsering en potensiell boble?), men boka hadde nok hatt godt av en runde til med redigering. Den er litt repetetiv og baserer seg i for stor grad på å peke på mekanismer i finanskrisen i 2008 og så anstrenge seg for å si at det er «akkurat det samme» med digitale annonser. Den lykkes med å overbevise om at det er verdt å spørre om vi har en boble og at konsekvensene av det ville vært enorme, og lykkes i mindre grad å vise AT det er en boble. Men det er kanskje typisk for bobler?
January 26, 2022
The first one in the FSGO x Logic series, as well the best one. While advertising is a complicated industry (and online advertising is arguably more so), the author did a great job introducing all the terminologies, their histories, their significance in advertising, why we should care about them, and why they matter to the larger society. The argument is simple: Online programmatic advertising is a market with bubbles, bound to collapse. The author being a lawyer and someone who studies technology policy enhances the credibility of the whole discourse. Highly recommend.
November 8, 2020
Inspired to read this book after the dopamine rush that I experienced when misleading Trump campaign advertising dollars with my fellow BTS stans. and boy am I glad that I did! Tim Hwang convinced me that advertising infra today is a big bubble and it's gonna pop soon just like the 2008 crash. Tim taught me how much digital advertising suxxxxxxx at doing its job
February 26, 2022
I don't know that the catastrophizing of the book really matches the 2008 mortgage crisis, so that analogy fell flat for me. Yet, the problems with the advertising and lack of attention was intriguing, at the very least to show the inner workings of the internet ad markets.
May 8, 2023
I think this just ended up being more economics focused than I was expecting — would also have loved some nuance around how google/FB’s ad platforms & bidding systems might differ from the general programmatic ad marketplace since many businesses focus on those 2 platforms and google/FB fully control the market on their platforms.
March 24, 2024
As someone who is not from the world of advertising or marketing I found this book to be very interesting and a fairly easy read. Great points were brought up correlating the targeted advertising bubble to the 2008 housing bubble. Good book!
August 20, 2022
A fascinating thesis and bundle of ideas to explore. Great little book.
March 15, 2023
It’s definitely a niche topic, but if you’ve ever spent thousands of dollars on a digital ad with basically no result, I think this is a fascinating and alarming read.
August 16, 2023
SO INTERESTING!! Made me proud to be an econ major >:)
September 4, 2023
Full review on StoryGraph.
September 7, 2023
This took big brain energy. 💀
November 3, 2020
Very well articulated on the opacity of programmatic buying leading to ad frauds. Also the growing skepticism on digital advertising.
Displaying 1 - 30 of 180 reviews