PorNO! How age verification exposed the power of friction in the internet age

Introducing Scrappy theory, a convoluted bastardisation of nudge theory

Back in 2008, behavioural economists Richard Thaler and Cass Sunstein published a book called Nudge: Improving Decisions about Health, Wealth, and Happiness. The book – and its associated thesis, ‘nudge theory’ – posited a form of what the authors refer to as ‘libertarian paternalism’. Without banning things, the world could be organised to ‘nudge’ people towards better decisions. The oft-cited example from the book is the nudge of putting fruit at eye level in a school cafeteria. Banning junk food would be illiberal; privileging healthy food in displays is an acceptable compromise.

Nudge theory, it always seemed to me, was just a way of reframing classic marketing standards from a public policy perspective. For better or worse, the world of advertising has known exactly how to nudge people, ever since the first caveman laid out his berries and demanded furs in exchange (I’ll bet he put the best berries – the money berries – out in front). In a way, the more radical idea behind nudge theory was an ideological one. America – which originated the theory – is the country that prizes individualism above any other. That is the American dream; the right that all people have to rise, from nothing, to have it all. But this has also stymied public policy initiatives. People want lower taxation, so that they can afford a nicer house (or a car the size of a nicer house). They don’t want to legislate on fatty foods or cigarettes or gun ownership, because that’s a personal choice. So, how do you retain this maniacal focus on individualism, whilst also attempting to improve the social fabric of the nation?

Nudge theory, then, is an ideological compromise. After all, that was the promise of free market capitalism. The market would win out, and, ultimately, good products would prevail. If everyone who eats some greasy burger from a fast food chain ends up with heart disease at 50, that won’t be good for business. You don’t need to legislate, you need to let customers make their own, sensible choices. And yet, that hasn’t happened: consumers continue to make bad, selfish choices. And America’s foundational contract continues to defend those bad, selfish choices. And so, in this reality, nudge theory attempts to correct for this imbalance without undermining the core premise that free markets will solve all problems. All you need to do is put apples at eye level.

In the past few weeks, the UK has launched age verification on pornography. This has been an extremely controversial part of the Online Safety Act. There is plenty to dislike about this legislation, which is lumbering (where the internet is fast moving) and erodes that fundamental sense that the internet is a digital town square, unregulated and unmoderated by the social forces down here, in the IRL world. UK public services don’t have a good reputation when it comes to digitisation (the NHS’s adaptation to digital records has been a famous nightmare, a slow and public car crash) or privacy. Campaigners, rightly, don’t trust the verification services that have often been mooted, such as inputting credit card details or uploading passport/driving license. The idea of potentially being included in a leaked list of PornHub users is – understandably – unappealing.

My instinct was to find this illiberal. I have a fairly low opinion of the ability of centralised enforcement to understand the issues surrounding online content, let alone properly regulate it. But I’m also sympathetic to the ambition. It’s clear that pornography has grown metastatically in the past decade. The image of schoolboys gawping over a pair of tits in a magazine – handed down, through the generations, by horny teenagers – has been replaced by something demonstrably worse. Preteen children have easy, unfettered access to an endless, evolving library of the most extreme content. Parental controls, an innovation of the early-00s, have proven wholly insubstantial, like putting a baby-gate in front of a border collie. Generation Z – and the Alpha babies to come – are not going to be restrained by the sort of parental controls their Gen X and millennial parents can implement. And so, a form of age verification is probably a good idea: after all, even the most strident libertarian is unlikely to argue that school-age children ought, philosophically, to have access to Brazzers or Bang Bros.

The question was more: can we implement age verification in a way that doesn’t destroy the user experience or create a privacy insecurity? After all, while some people argue for a total ban on porn, most people are reasonably content with a small market by and for consenting adults. The global pornography industry has grown like topsy as a consequence of the ease of access. The inhibiting factor with porn is embarrassment – people don’t want to buy a dirty magazine from a newsagent or rent Debbie Does Dallas from some seedy video shop – so the privacy of the internet has created a total disinhibition. This has turned the market for pornography from sex-starved teenagers and trenchcoat-clad perverts into something almost universal. Commensurate to the expansion of the audience has been the growth of advertiser revenue, which has precipitated an influx of ‘free’ porn. The more ‘free’ porn you have, the bigger (and younger) your user base. And so the cycle starts spinning: the audience keeps growing, advertiser revenue keeps increasing, the pantheon of ‘free’ porn becomes ever larger and more exotic. (This is all putting aside, for a moment, the psychological impact of widespread porn usage, or the sociological impact of the debasement of normal sexual etiquette.)

So, in order to depress the porn industry to an acceptable standard, you have to reduce ease of access. That will shrink audiences, reduce ad revenue, and squeeze ‘free’ porn (thus putting the onus on paywalled porn, which is a much healthier model). The Online Safety Act has, I think, pretty much done this. The new system, rolled out on July 25th, requires users to pass an age verification test which utilises facial analysis to confirm that a visitor is over-18 (I assume that people failing the test have the opportunity to upload government ID instead). I did the test on Reddit and it took less than a minute to get approved and all set (Reddit, like various non-porn sites, meets the criteria for requiring approval due to widespread NSFW content). This reassured me that the process was sufficiently unobtrusive that most people would be willing to give it a go (I cannot vouch for the accuracy).

A couple of weeks after these new measures were introduced, data analysis from firm Similarweb suggested that age verification had initiated a staggering reduction in traffic to porn websites. Their report said that PornHub, the UK’s premier adult content platform, had seen a 47% decrease in traffic between 24th July, the last day of total pornographic freedom, and 8th August. XVideos, a competitor, saw the exact same 47% drop during that period.

The humourless Americans in my mentions seem to have missed the point…

If your goal is to reduce traffic to porn websites, then this sort of age verification works. Obviously, the current regime is a huge boon to VPN providers, and I don’t doubt that some of that 47% drop is accounted for by users migrating to use of VPNs (which wouldn’t show up in the UK data). But I don’t think that number is especially meaningful, and once similar legislation is rolling out more globally (the US should enforce a similar system imminently) then the impact of VPNs will reduce. But the figure is telling. A considerable amount of people, wanting to indulge an onanistic afternoon, have been deterred. My suspicion is that most of these have not been put off by the actual task of age verification (which is very simple) or even, particularly, by privacy concerns. Instead, they have been discouraged by a collateral requirement: to create an account.

In 2019, PornHub announced that it had 22m registered users. At the same time, it was receiving 115m visits per day (42bn visits per year). From these numbers (PornHub does not seem to have announced any more recent data on total registrations, though it is likely to have seen quite a big increase, from Blighty, this summer) it is clear that the majority of PornHub’s users do so without registration. Which is natural, because their whole model – which changed the world, if you’ve listened to Jon Ronson’s superb Butterfly Effect series – wanted to encourage open access to pornography. (In recent years, they’ve tried to compete more with OnlyFans, which will likely have led to an increase in registrations). Put simply: most PornHub visitors are impulsive rather than organised. The advantages of registration are limited, the disadvantages clear (who wants to run the risk of opening their email and seeing a new message, subject line: “Nick, we’ve got new BUSTY MILF content just for you!”). But age verification has to be linked to an account, and so users, naturally, steer away from PornHub.

The success of this approach reminded me of nudge theory. Without banning porn, the policy initiative has significantly reduced access. Over time, this ought to constrain the industry. But rather than nudging users towards preferred behaviour, this legislation has created a friction that disinclines them from exhibiting non-preferred behaviours. It reminds me of Scrappy-Doo, perpetually imploring the gang to “lemme at ‘em!”, yet constantly being held back. Like Scrappy, feeling the finger on his collar as he raises his fists, British pornophiles have been restricted from destructive behaviours because the incentive to do action A (visit PornHub) is offset by the deterrent of action B (create PornHub account).

It got me thinking about what else we can – and should – use Scrappy theory on. After all, we’ve tried nudging people for quite a while now, and society is still going down the crapper. The problem, I think, is that most centralised public-policy nudges are being offset by conflicting private sector nudges. You can put fruit at eye level in cafeterias, but McDonald’s will simply put an advert for the Big Arch burger at eye level at a bus stop. It reminds me of the famous ArsenalFanTV moment, where after seeing his team lose, the subject announces that “we mustn’t forget that it’s been raining”. “Are you being serious?” his interviewer responds, incredulously. “It’s raining for both teams!” It’s been raining for both teams – and everyone has the nudge tools in their, no pun intended, arsenal.

But there’s no incentive on the private market to use Scrappy theory, which creates a restraining friction. It is, in a way, the opposite force of a nudge, exerted on a public who have been nudged so much they feel like a lump of clay or Patrick Reed’s golf ball. One idea I had, to apply Scrappy theory, would be to mandate that social media websites had to enforce a log-out (and re-log-in) every day. You could make an argument that the policy would make it more secure for users, preventing them from leaving accounts logged in with the resulting vulnerability from that. But if Instagram and Facebook and Twitter logged you out every day, are you confident you’d log back in? At the very least, it would be tedious – a small erosion of that tipple of dopamine.

You could also mandate that social media sites must require a monthly password change. That, in combination with the mandatory log-out rule, would guarantee that, after a couple of months, I never went on Instagram again. If we believe that social media is a net negative for young people, a Scrappy approach like this strikes me as more intuitive than a blanket ban for under-16s, which would be agony to enforce and likely a significant expenditure of political capital in return for very little. Instead, make Snapchat keep logging children out of their accounts and then having to solve a maths captcha puzzle to get back in (two birds, one stone). Instead of banning disposable vapes (illiberal) why not introduce a vaper’s license which you can get, free, in-person at the Vape Licensing Office on the Island of Unst (liberal)?

The above paragraphs are free ideas for the architects of our public policy future. Friction, I think, is a better tool than a nudge. But within these recent policy changes, there are learnings too for digital media businesses. During the same timeframe when PornHub and XVideos saw a 47% drop-off in traffic, OnlyFans witnessed only a 10% reduction. This supports my theory that the real inhibition – for PornHub and XVideos – is the creation of account. OnlyFans – a service that allows adult creators to build a storefront to paywall their content – primarily runs on a membership model. There is a small amount of teaser content, but, essentially, its user-base need to sign up to secure the goods. For them, the new rules are probably a good thing: in order to view the free, tease content, users now need to create an account. That’s half the battle when it comes to converting lurkers to subscribers. (Plus, a site like OnlyFans never had a corporate incentive to receive underage visitors, which many ‘free’ porn sites do, whether they like it or not).

Most digital media business are faced with this challenge. Whether you’re a podcaster or a YouTuber, the second task – after getting someone to listen to or watch your content – is to convert them from a one-off engagement to a repeat one. So an ecosystem that pushes people more towards registration and account building is one that will, down the line, be more profitable. It is another correction for the original sin of the internet – that everything should be free – and one that publishers should embrace, rather than fight. It is a far more sustainable model than the creation of centralised digital passports, which would, in a very real way, erode the culture of the internet.

For now, the authors of the Online Safety Act should be pleased with what they’ve seen. The figures won’t look as dramatic forever. Children (and perverts) are ingenious and will figure out a workaround. But this is evidence that adding a small amount of friction to the UX can spark dramatic changes in user habits, and that’s a lesson worth knowing when it comes to an internet that is so often detrimental and so totally unruly.

Follow me on Bluesky and tell me if you agree or disagree.