Tim Hwang’s new book Subprime Attention Crisis makes the case digital advertising is a lot like America’s 2008 subprime mortgage crisis.

The gist being in the lead-up to the mortgage crisis, sellers bundled up lots of mortgages into good batches and bad batches. The good ones were likely to pay out and the bad ones to default.

The problem was the people buying these investment portfolios couldn’t tell and didn’t know whether they were buying a good batch or a bad batch. Once the bad ones started to default, buyers lost confidence in the whole market, making even the good ones look bad by association.

The attention crisis, Hwang argues, is a lot like 2008’s subprime mortgage crisis. But instead of big investment banks, Facebook and Google have bundled up attention and metrics like clicks and likes, but they do a very good job of hiding which ones are good and which ones are bad. The comparisons stop there, when you consider digital advertising has ad blockers and people who just plain ignore the ads.

About halfway through the book is a big meaty section I almost want to copy entirely. But because it’s somewhat long for a blog post, a few excerpts will have to do:

When they launched in 1994, the first banner ads generated a remarkable click-through rate of 44 percent. …Today, banner ads, command far less attention. One data set drawn from Google’s ad network suggests that the average click-through rate for a comparable display in 2018 was .46 percent. For some industries, that number is as low as .39 percent. That’s about one in every two hundred people. Recent attempts to measure click-through rates on Facebook ads reveal similar rates of less than 1 percent.

…

…Even the sub-1-percent click-through rates may overstate the effectiveness of ads on some platforms. On mobile devices, close to fifty percent of all click-throughs are…accidental “fat finger” clicks.

(I hear from people a lot, “But we do so much better on phones!” No, you don’t.)

…

In 2009, one study estimated that eight percent of internet users were responsible for 85 percent of all advertisement click-throughs online.

…

Another “large-scale experimental study of online search ads in 2014 concluded that, “brand-keyword ads have no measurable short-term benefits.” And, “Ironically, ads generated engagement mostly among ‘loyal customers otherwise already informed about the company’s product.’ The ads, in other words, were an expensive way of attracting users who would have purchased anyway, leading to ‘average returns that are negative.’.”

He goes on to cite how internet users age 20-40 experience “little or no effect from the advertising”. And despite being just 5% of all Internet users, people age 65+ are responsible for 40 percent of the total effects observed.

All this is to say most online advertising is “vaporous”. The sort of confusing, opaque, problems we had in the mortgage failures but now in the “attention advertising market”.

Much of this is driven by poor ad placement (looking at you, Google Ad Network), putting the wrong things in the wrong places (like advertising for bananas or whatever on Facebook. Who cares?), and ad blocking that is highly prevalent in the United States and Europe and expanding fast in Asia.

And we haven’t even mentioned all that Hwang says about fraud, click-farms, bots, and other junk. He estimates enough money is wasted to fraud in all online advertising each year to equal the entire current value of Facebook. It’s as much as 20% of all online ad clicks.

I have said this a zillion times:

• Online advertising is full of fraud.
• If you are paying for ads on Google or Facebook, it had better be the right kind. For Facebook, that’s “awareness campaigns” (which are impossible to calculate value and lead to this vapors nature of things) and digital goods that can be handled on-device, like downloading an app. For Google, that’s narrowly defined buyer-intent searches like hiring a handyman.
• Older people can’t discern the difference between an ad and something not an ad. On the upside for some advertisers, seniors hold just about all the world’s wealth, so it can be lucrative. If you have to advertise to teens, you’re better off with a billboard by the high school.

And now that I’ve read this book I think another thing is likely to happen: a complete collapse.

Hwang thinks we should do a “controlled demolition” so the bottom doesn’t fall out. But I think the bottom will fall out eventually as small businesses (which account for 80% of all of Facebook’s advertising revenue) wise up to the realization this doesn’t work. And it doesn’t work because most of the ads are dull, boring, and just plain bad with lousy copywriting and cheesy stock photos.

Already I tell clients they need ever-more money to have any chance. $300 a month doesn’t cut it anymore. To have any hope, you need $1500 a month for AdWords or Facebook, depending on your market. Godspeed if you’re advertising insurance, financial services, or attorneys. Or to anyone under the age of 35.

If advertising online completely fails, Hwang says (and I agree) that we’ll see a lot of publications completely vanish. Ad sales won’t be enough anymore because no one will trust any of the ads, despite static ads being no more or less more effective than programmatic, text-based, or targeted ones. Paywalls will abound. No more free Google Maps or Gmail, which rely entirely on Hoovering up data for advertisers. There could be pay-to-play placement in search results (as in, to be included at all, not just as the top in the ads).

Increasingly I think publications like The NY Times, Wall Street Journal, and others that have paywalls are playing it right. And companies like Apple and Amazon will be in a good position to continue delivering maps and other “free but not free” services to their customers who pay for physical things.

One obvious downside to this pay-to-play approach is low-income people are going to be left out of a lot. Right now, the web is largely equitable. Everyone can get at the same Google or Facebook or Maps or YouTube or Wikipedia or whatever. But not if everything requires a subscription first.

At best, this could result in a lot more creativity in advertising, something that’s desperately needed.

If you must advertise, anywhere, you need to inject some creativity in the ads that make people even remotely want to look at it. Which may be impossible given most people’s aversion to being anything but bland or corporate or boring.

Think more advertising akin to Cards Against Humanity digging a big hole and less of whatever your local hospital is putting out.

Same goes for the junk most people throw at their Facebook walls every day. An appalling amount of attention is paid by marketing people and social media managers about the importance of posting at least 3x a day and other wasteful nonsense. If one more organization asks me what my hashtag goals are, I’m going to poke my eyes with rusty spoons.

Regardless, people are already so allergic to ads and this boring junk even the interesting ones are getting thrown out. Hwang notes studies of Snapchat users that the average view-time of a video ad is 1 second. It’s no better on YouTube, where it’s “whatever the minimum time required is to watch an ad”.

In the meantime, buy a subscription directly to a journalist or publication. And expect the Internet to look a lot different in the next five years.