The a16z Show - a16z Podcast: Advertising vs. Micropayments in the Age of Ad Blockers

Starting point is 00:00:00 The content here is for informational purposes only, should not be taken as legal business, tax, or investment advice, or be used to evaluate any investment or security and is not directed at any investors or potential investors in any A16Z fund. For more details, please see A16Z.com slash disclosures. Hi, this is Chris Dixon. This is the A16Cd podcast. I'm here with Benedict Evans. Let's talk about, so a lot of people are talking about ad blocking lately. iOS 9 allows you to download. add blockers or content blockers, which are mainly ad blockers. I guess the first question I have is, is this, there's a lot of talk about it. Is it actually a big deal? It's an interesting one. I think there's several ways you can look at that. One is obviously the majority of time spent online on a smartphone actually isn't inside

Starting point is 00:00:50 the web browser. It's inside Facebook and it's inside other apps. This would also, though, does it affect it when you're in the web view? So yes and no. if you implement the new, faster, cooler web view, then it's affected. But if you're actually just looking at stuff actually in your newsfeed, then obviously it's not. But if you tap on a link and they implement the web view, then it's affected.

Starting point is 00:01:09 A lot of content is viewed that way. Yeah. So there's that kind of wonky question, well, how much content is this affected by? Then there's how many people are going to install these things? And right now, you know, clearly you're going to have a search of people who really care about this stuff. It's always seemed to me that there's kind of several different, there's a portion of people who really get upset.

Starting point is 00:01:28 about tracking, and I'm not sure how a broader percentage of the overall population that is, and then there's another percentage of people that get really, really upset about the principle of advertising, and again, I'm not sure how many people that is. The interesting thing on mobile, I think, is that the actual just the amount of code and the amount of

Starting point is 00:01:44 JavaScript and ad tech stuff that's happening actually slows down your experience. So there's actually like a real user benefit as opposed to like a kind of a theoretical benefit to installing this stuff. I mean, I've always assumed it's just sort of the power of defaults here, which is if Apple were to make this the default, that it stripped out advertising, that would be a big deal.

Starting point is 00:02:01 If it requires people to go and download a special thing, and it's actually quite wonky right now, you have to go in or clujy, you have to go in in the settings and install it, yeah, like 10% of the, what, 5 to 10% typically do things like that that aren't default. And those 5 to 10% are probably in their basement screaming at the computer, and they're the types who don't buy stuff anyways. And, you know, I don't know. So, like, is it actually going to affect that?

Starting point is 00:02:27 So I think the kind of high watermark here would be Google Maps on iOS, which has something like 100 million active users out of maybe 400, 450 million iPhones. And I think that gives you a pretty good indicator. The power of defaults. Yeah, here is like an actual really, I mean, it's not like Apple. Google Maps is a vastly better product. And yet only, your point is that only 100 million people use what is clearly a vastly better free product. Yeah, only a quarter of the base have installed it.

Starting point is 00:02:50 I would say it's vast. I mean, it's like Apple Maps isn't terrible, and most people don't use it that much. It's not bad until it tells you to drive off of a bridge. Yeah, most people don't use maps that much. Most people don't actually care that much. And there you are, even where that's a chord of the base. So how many people will install this and then how much does it flow through? I mean, there's a kind of a broader kind of set of questions here around, well, as a publisher, is advertising really sustainable anyway?

Starting point is 00:03:13 Is this just like a catalyst for saying, you know, well, screw this stuff? This isn't working very well. How do you get read? Should you be thinking about micropayments? Should you be circling back on apps? should you be thinking about other ways to get distributed? And so at the same time as Apple launched this, obviously they launch their Apple News,

Starting point is 00:03:31 news aggregation app, and then you have Facebook doing instant stories inside the Facebook. One argument, which I think I believe is it's a diversion to worry about what's happening in Safari when the real ad blocker is called Facebook, right? And it's not that they actually block the ads, is that the stuff, if they want your stuff to get shown pretty soon, you're going to have to use instant articles. And instant articles is not going to let you show your ad.

Starting point is 00:03:53 I mean, it's going, if they're not native ads. And that's what, you know, in some ways, that's what you should, you know, if you're a publisher, you should keep your eye on that ball, not. Yeah. If half your traffic is on Facebook, then Facebook's already going to be blocking now to say, you know, this other stuff doesn't matter. Yeah, I mean, I've... In some ways, Facebook is the new browser. It's the new RSS reader, and it's the new ad blocker, all combined into one. And what's happening in Safari with plugins is a sideshow.

Starting point is 00:04:19 I think that's right. I mean, there's a kind of, I wrote a blog post the other week, kind of starting with, with the chart of the traffic to my blog, which now gets like 100 to 200 to 300,000 views a month, depending on what I've written about. But I started in 2010. For the first three years, I basically got 100 views of months or 200 views of months.

Starting point is 00:04:39 And the point of that is, it's like, yes, you can write stuff. You could write stuff in, you could code the stuff in notepad and put it on an FTP server, but actually you're going to get you, what you want is to get read. And if you want to get read, you've got to be in some kind of a platform that's giving your audience. and increasingly that means Facebook or it means medium or it means LinkedIn or it means some other intermediate platform that kind of comes with an implicit ad blocker, a revenue model, a filter, some other set of stuff that comes between you and your audience, which is always like the problem for big publishers looking at Facebook,

Starting point is 00:05:10 it's like it's kind of a deal with the devil issue. They can give us traffic, but it comes on their terms. So a lot of the ad blocking discussion has then led people to say, well, what, if it's not, you know, Does this mean we should be thinking by different advertising models for, you know, for example, for newspapers, et cetera? Or micropayments or some other thing. And at least the discussions, the micropamance, which we were discussing and actually what triggered this podcast was a discussion with Startup L Jackson, a fictional Twitter character who apparently dictates what we discussed. I can't believe Mark has time to tweet under two characters at the same time. No comment.

Starting point is 00:05:47 So Startup would like us to talk about micropamines. So you, I think, are skeptical of micropayments. Well, I'm not skeptical. It's just we've kind of, we've seen 15 years of people wishing that you could do this. And the frictional barriers to having one platform on every site, having people enter their credit card, having some sort of identity platform that would make it work. It's always been one of those things that's like, if we'd had the information superhighway instead of the internet, and it had been built by, you know, AT&T and people like that,

Starting point is 00:06:18 and it was like a system that happened on your TV, then yes, you'd have micro payments, but in fact we have this permissionless, you know, completely open, defragmented environment, and it's really hard to make something like that work. I mean, when Apple launched the iPad and things like Newsstand, people thought you were going to be able to do micro payments on the iPad

Starting point is 00:06:36 and you, everyone was going to make an app on the iPad, and everyone was going to read stuff in there, and there'd be micropayments and subscriptions, and it kind of didn't work for a whole bunch of reasons. And I think the question would be, It's easy to say you should be able to bill one cent for each article or 10 cents for each article. But getting that frictionless payment platform in there and having everybody use it has just been too hard so far. That's not to say it won't be.

Starting point is 00:07:02 So you walk around New York City and you have these little sort of small casual payments. And actually if you read, it's always worth reading like the original Bitcoin paper. There's a lot of discussion around Bitcoin. But the original paper is actually, first of all, extremely well written, very interesting from a tech. perspective, but also the first paragraph is very clear about what the point is. It's small, casual payments is the point, right? So you walk around New York City and I, you know, I am late to a meeting and I go to a street vendor and I buy a cup of coffee. I don't want to have a long-term relationship with that vendor. I don't want to fill out a form. It's very useful

Starting point is 00:07:33 to give the guy, whatever it costs now, $2 for a cup of coffee. Well, $6.50. $6.50. Maybe these days. But, you know, it's very useful. Cash is a very useful way to to to make small casual payments for example walking around a city right um so the argument is we never you know it would be nice to have it would be great to have that on online and so i don't give you a practical example um Spotify in many ways as a payment bundler um you could imagine if if we were to start over today with the entire music industry and everything else starting from scratch how would you architect um music service what you might architect it where um you know i give put ten dollars in into my browser and it charges me every month. And then my browser tells me, my music browser,

Starting point is 00:08:17 whatever it might be, you know, it looks at what music I listened to that month. And then it divvies up my $10 and pays it directly to the artist. And there's no label and there's no Spotify. And it's a relationship between me and the artist. And, you know, this kind of clay shirky argument that no one will ever adopt micropayments because of the mental friction, that that, that I think, is based on what I think is a flawed assumption that micropayments presumes that a human is deciding every single time you pay something. Well, they have to look like micropayments. If they look like micro payments, it might's one thing, but if you can package it in some

Starting point is 00:08:48 other way. Well, I think that's, I mean, I always assumed, I guess, for me, micropayments always meant the infrastructure supported small casual payments. So two entities could make a transaction of a small amount without incurring transaction fees that make it prohibitive and without having to have an ongoing relationship the way you currently do with a credit card and without having a 20% charge back rate and like all these other and like the charge. and aisles. So you had a way that you had sort of internet cash, right? And you could make small

Starting point is 00:09:15 cash payments. There's no reason that can't be software assisted that you can't have, you know, bots and other things assisting how you divvy out those micro payments. There's no reason a human has to be saying, let me decide who to give 10 cents to and 10 cents to. Those are two, that's just, that would just be an old-fashioned architecture to me to have a human making those decisions. So to me, the argument that sort of a human has, will never want to, you know, the mental cost of making the decision. That's just a very old-fashioned assumption that there's a human deciding how everything's done. Architecturally, it seems to me that having the ability to have sort of something

Starting point is 00:09:46 in the equivalent of cash, low transaction costs, no ability to have fraud, not having to have an ongoing relationship between the two entities would be a very useful architectural thing. Bitcoin is one very promising example of that. There could be others. The existing credit card system seems extremely flawed. PayPal is extremely, I mean, the way PayPal deals with many issues is they simply block about half the world from transacting, right? I mean, on eBay, 20% of all international transactions are blocked because the way they deal with the fact that their system is not built for the Internet is they tune the algorithms to just block almost everything that looks remotely suspicious, right? It's, it's, if you submitted PayPal to your CS 101, you know, class as your, as your ideal payment system, you'd fail out.

Starting point is 00:10:39 You'd fail out of the class. You must have been complete idiots, you know. Well, no, I mean, the PayPal is a great, you know, there's obviously smart people. But the point is it was, it was an extension of the credit card system, which was built for a different era. It was not built for this. It simply wasn't built. These things were not built for the kind of the way the modern internet works. I would argue.

Starting point is 00:10:56 Yeah, I mean, I think I, there's a sort of observation like Mark likes to point out that it's the, what is it, 403? 402. Yeah, well, this is, yeah, this is the payment. There's also no identity. And I think there's a other. So this is the error code for, four, is. no, there's no web page, there's also an error code 4-2, built an HTTP spec, which is payment required.

Starting point is 00:11:14 And you're saying, yeah, except that we never want in building that. And I believe, so like, I believe that we, so you go back to you rewind in 1993, and you had different possible futures of the internet, right? And one possible future, you had built identity and you had built payments into it. And to me, that would have been, I mean, I love the internet, and it's awesome. And thanks to Mark and all these other great people who did it. And I'm not complaining and looking and gift towards in the mouth. That said, rewinding it, it would have been really nice to have built identity and

Starting point is 00:11:39 payments in the first place. So my first job, like I think in my, in 99, I joined an investment bank. And I think in my first week, I was sent along to sit at the back of the room in the IPO drafting session for a company called Trintech, which makes chip and pin payment terminals. And they were one of the companies behind something called SET, which, and the idea of was it was basically an escrow system for your credit card. So your browser had your credit card. but the e-commerce site never got your credit card

Starting point is 00:12:10 it was kind of passed through like a kind of tokenized It's actually kind of how Apple Pay works Exactly, it's exactly how Appleby works Yeah And of course Which is by the way a far better system You don't pan, like the credit card system

Starting point is 00:12:21 You hand the password over Exactly which is just You literally fail out of class If you submitted that in the project Exactly This was the worst architecture This was like 9899 And this was like the idea

Starting point is 00:12:30 for how you were going to get people To be comfortable putting their credit cards in And it turned out that the answer was You just put your credit card in And the credit card companies eat the fraud costs and PayPal blocks after transactions and so on so you just deal with it that way. But

Starting point is 00:12:43 you know to your point if you could there's always this sort of I mean it's what you say about going back to 1992 we kind of look at what happened and we presume it's inevitable and we look at all the micropayment things that failed and presume well that's just because micro payments aren't going to work just like we know we look at the failure of the portals in

Starting point is 00:12:59 America and think well portals didn't work you look at China portals did work and a lot of the things that look inevitable as failures is just that's just kind of how it turned out and the technology didn't quite work out like that. And it's certainly possible that we could have, you know, whether it's Bitcoin or something else, a micropayment system of some kind or subscription system of some kind that did work. The challenge is always like, well, how do you get everyone to adopt the thing and all the publishers to adopt the thing? It's a massive, massive.

Starting point is 00:13:25 It's the mother of all chicken neck problems, right? It's like, it's the big chicken neck problem. It's the biggest one of all. And the thing that comes in parallel is you have Facebook or BuzzFeed or Medium creating, solving it from the side. So Facebook has both the distribution and the business model for third parties. Well, the business model for themselves, the distribution for third parties. And BuzzFeed has the distribution and the business model for themselves. And Medium right now has a distribution for third parties, but no revenue model, although presumably that will come. And I thought it was interesting just to look at, you know, how BuzzFeed in particular comes to this,

Starting point is 00:13:58 going right back to advertising and ad blocking, in that what they haven't done is kind of slap a banner ad as invented by Wired in about. at 1996 or 1997 and put 10 meg of JavaScript on the back of it to try and work out who the hell you are, which is my point about identity is the reason the advertising is 10 meg of JavaScript is because they don't know who you are and they won't have some vague sense of who the hell you are to show you so they know what kind of ad to show you. Whereas if your BuzzFeed or your Facebook, if your Facebook, you have that identity because it's in the Facebook profile. If your BuzzFeed, you build native advertising and you build advertising.

Starting point is 00:14:31 Well, and the native advertising is inherently doesn't, the identity is built in by who shares it. Exactly. Right. So the person who likes cats shares with another person who likes cats, and that targeting, quote-unquote, is happening by humans, not by... Yeah, so you're not building 10 mega kind of crappy JavaScript to try and guess as to who this person is. It's actually inherent in the product. And that's actually also a much better revenue model. And so it may be that, you know, you come at this from that other angle, which is you build it as a single site. You solve it as medium. You solve it as Facebook. You solve identity as Facebook rather than open ID. Well, to that point, like, you know, one way to look at this is you go to the New York Times, and it's, you know, New York Times is a, you know, it's a beautifully, you know, from a journalism perspective, it's an amazing product, right? It's great writing. They spend tons of money on research. And so you're looking at, you know, if you go to just a desktop web and you go to the New York Times, you're looking at the page of the text, and you've got an A-plus product. And then you look around the text. And you've got literally, like, I have screenshot. I always screenshot whenever I see these kinds of ads. It's like the Tahiti diet or teeth whitening, right?

Starting point is 00:15:37 It's the lowest of the low, crappy ads, like sitting around on these banners, which I guarantee you on mobile. If there are any clicks on those ads, I bet you 95% of them are fat thumb, fat finger accidental clicks. Like nobody is willing. Like the odds of somebody reading an article about, you know, whatever, something, you know, foreign policy in the New York Times, and then clicking on a Tahiti diet. I'm sorry, that is not, that it's an, if it's happening, it's actually. Banner ads are like Yahoo. It's like you know that hundreds of millions of people must be clicking on them, but you've never met anyone who does it?

Starting point is 00:16:09 But they don't. I don't know. So I would just say this that you have, if you just look at it through the lens of product development, you've got on the one hand at a great product, you know, the economy, not the economists, they don't do it. Well, they have banner ads, I think. They have ads, yeah.

Starting point is 00:16:22 Whatever. You have the journal, the F.T., the economists, the New York Times, all of these first-rate journalistic institutions with top-rate journalistic products. bundled with very third rate monetization products. And then people say, oh my gosh, digital monetization doesn't work. We're trading analog dollars for digital pennies. Now, there's two ways to look at it.

Starting point is 00:16:45 One is there's something magical about a computer screen that makes it less monetizable. The other is perhaps you've aligned yourself with a business model that is just a very low-quality product. And that would be, by the way, the BuzzFeed perspective. And people have this, I think, this very, they misunderstand what Native advertising. Native advertising are ads that people actually like, right? This is the goal of Native advertising. This is Native advertising done right. It's ads like, it's sort of like Super Bowl ads or something.

Starting point is 00:17:10 Well, the TV advertising industry has always known this. Of course it is. It always makes ads people like. You make ads people like, and in the social web, people will share them as opposed to, you know, thinking of what the publishing industry has been, I think, kind of misled or down the wrong path, of thinking of it is like you have the hamburger and you have the spinach and you have to take the spinach with the hamburger or whatever. I don't know. I think there's an interesting thing implicit all this is what happened with paywalls.

Starting point is 00:17:36 As soon as with the newspapers it went to paywalls and particularly like hard paywalls, like the Times, as opposed to the New York Times, or the Times you have to pay, that's it, you can't read anything. Do you mean that the London Times? Yeah, the Times that the New York Times are named after. The London Times. That's actually the Times. That's the country in the European history.

Starting point is 00:17:53 So the point is that when those newspapers did that, the page, the site experience got way, way better. because all the crap advertising disappeared all the link bait headlines disappeared all the tabula and the click on 500 other stuff disappears you get white space on the site because all of a sudden they don't have to do that anymore and it feels like you could kind of do like a two by two like a kind of a gardener two by two matrix

Starting point is 00:18:16 of you know traffic versus value or whatever the answers would be and like in one corner you would have the Guardian which is trying for massive global scale with no monetization worth speaking of and in the other corner you would have the Daily Mail which has got the same traffic, a tenth of the cost base, and a tenth of the content quality, and the rail of shame and everything else.

Starting point is 00:18:37 And in another corner, you would have, like, the Times, the Financial Times, the Economist, where it's like pay or go away in some sense. And the New York Times, to some extent, as well. You've kind of got to pay them if you want to read the New York Times every day. And then in the other corner, maybe you have BuzzFeed, which is, no, no, no, let's just actually take, instead of having a vast audience

Starting point is 00:18:58 and just throwing, we don't know what at it and hoping that we'll make it up, let's actually think natively about what money, revenue on the free internet looks like. You see what I mean? You've got these kind of

Starting point is 00:19:10 these different axes of quality versus traffic. I mean, for me at least, the big, the great modernization epiphany on the internet was Google, right? Like the first time you saw, just the assumption

Starting point is 00:19:22 had always been that you had to, you know, You had the carrot and the stick. The carrot was the product and the stick was the monetization. And Google said no. I mean, the ads are actually quite useful. Like, it turns out that in many cases, those are as relevant or more relevant than the organic listings. Well, it's the best story is, you know, I can't remember who it was that they were trying to license their product tube in the very early days.

Starting point is 00:19:46 And he said, well, I don't want to use the search engine to be too good because people will leave. This is a site. They were trying to sell the company. Yeah. And they tried to sell it for a million and they got off for $600,000. Because people will leave the site, so we won't make any money. Well, you remember at the time, it was all around stickiness. Yes.

Starting point is 00:20:02 So you had to keep them. Search was well known to be a bad business. And Portals was a good business. So you had to search was a way to get them into your portal where you then show like celebrity gossip and then banner ads. And that's why Yahoo actually outsourced their search to Ink Tomi and others. So that's a good cautionary tale. Well, so it struck me when you talk about finding new models.

Starting point is 00:20:26 It's like, in a sense, it's not that it's easy for BuzzFeed, but, you know, BuzzFeed you can do that. And if you're a top 10 site, you can think about doing that. If you're the next 10,000 sites, you know, once upon a time, you'd say, well, we're going to have to sell our banner ads, and that was a nightmare. And then you have ad networks, which kind of act as a natural intermediary. So is there an ad network for a native advertising in some sense?

Starting point is 00:20:44 You know, how do you reach a point where everyone can be doing good advertising as opposed to only the people who've got a big team of geniuses to work it out? It's a good question. I think it's a, it's a, it's a, it's a, it's a, it's a, it's, it's a, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's, it's in, in, in the news world, let's say, is they interacted through physical adjacency, right? In other words, you put the article on the page and you put the ad next to it, and that happened on paper, and it happened on the web. And TV. It's temporal adjacency, right? It's like, you know, you have this slice and that slice. The idea with BuzzFeed is you have that the adjacency actually comes in the layer of the creation and the data.

Starting point is 00:21:42 So you have lots of data and learning, and that is the sort of the center of the system, and that informs both the publishing side and the advertising side. And that's where the quote bundling happens. It doesn't happen in the articles appearing next to ads. And you can't because the assumption from the start was that the internet will unbundle those things. They will unbundle the ads from the content. Yeah, you can't understand BuzzFeed if you go to BuzzFeed.com because no one that's not. No, so, yeah, that's not. That's a very small portion of the experience and not really how it works.

Starting point is 00:22:13 You understand it when you go to it through a social network and you see, because basically ads have now become unbundled from, like the only place that ads are being bundled with, with content is in the platforms, right? It's in Facebook, it's in Twitter, and it's in native ads, and they will be the new ad networks in many ways, right? Ad networks, like this ad tech industry, unfortunately, I think will mostly go away. Banner ads, et cetera. It will be Google. It will be things in streams.

Starting point is 00:22:40 They will, Facebook knows who the person is. So is that the new search? Or the search industry before Google? You're saying, you know, people are mistakenly outsourcing it, and in fact, it will come back and be a good thing? in a more general sense that there were all these people doing not very good search, and then Google came along with a completely different way of doing search. And there will be like a similar kind of a nuclear bomb that hits the ad tech industry.

Starting point is 00:23:03 I think it's happening now, isn't it? I believe. I mean, also, if you just look at the public ad tech company stock prices. No, that 50% of traffic is bots and fraud. And so I think it's happening now. You know, for me at least, that said, I think the publishing industry, once it's going through this this kind of

Starting point is 00:23:21 turmoil now will end up being stronger than ever because you've got what is the number now, 3 billion people with smartphone internet access going to... Two and a half with smartphones going to probably four, maybe four and a half. I mean the scale, so even if you assume the worst case that you

Starting point is 00:23:36 reduce monetization by what a factor of 10 or even 100 the scale of reach here will go up by at least that much. You know, if you consider in the past a regional newspaper or something can now reach soon what the entire probably what six billion people by 2020 what I don't know what you're projecting um there's five billion adults on us probably there's probably two two and a half billion smartphones now that will go up to four billion but you know

Starting point is 00:24:00 a lot of those are farmers in rural africa and it's a multiplier effect right it's it's not only is it is it number people it's the amount of time spent yeah it's a time spent it used to be at best right like you would like someone who a very engaged news reader might spend 45 minutes in the morning reading the news and that was it now it's a constant thing throughout the day. It's deeply integrated. If you look at what's happening with Facebook, I mean, you know, they, it's now news not only is something you consume. It's the thing you talk about with your friend, you know, it's part of your social experience on the web, right? It's like, that's a lot of what BuzzFeed's insights are as well, is that they're not just things to

Starting point is 00:24:33 read passively, there are things to then share and discuss. And so the amount of time spent on news and on media and the number of people with access to it is, you know, I think like if you add up sort of you multiply those two times each other it's like a thousand X um it's very very promising and um i think just this this we just need to figure out in this transition period what the best models are and there will be some shakeout um but the good journalists and writers and content media creators will do very well in the new world yep okay thanks thank you

The a16z Show - a16z Podcast: Advertising vs. Micropayments in the Age of Ad Blockers

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.