This is the text version of a talk written for BBC Radio 4’s Four Thought programme, first broadcast on October 23rd, 2013. It was recorded at Somerset House in front of a live audience with David Baddiel hosting. Huge thanks to Giles Edwards for the invitation to speak, and for help and advice. Thanks also to Bill Wasik’s excellent And Then There’s This for the account of Jonah Peretti’s work.

 

Empires of Attention

Thank you for inviting me to come and talk today, and in particular, I want to thank you all for your attention. Your attention is a very valuable thing, and to decide to spend it listening to this talk here today, or at home on the radio, or later online, is not an insignificant act.

I’ve worked in digital media and broadcast for over 15 years, and I’ve become obsessed with attention. My story tonight will be about this obsession, and what I’ve learnt about the way attention defines our culture.

Because how we understand audience attention – how we ask for it, measure it, and build business empires by selling access to it – is fundamental to our culture. For the last few hundred years, the business of culture has essentially been the business of measuring audiences’ attention. We can trace a line of entrepreneurs of attention from today’s culture backwards through the last two centuries – from Jonah Peretti, who has used his intimate knowledge of the patterns of digital attention to build The Huffington Post and Buzzfeed, two of the biggest news and culture sites on the web; through Arthur Nielsen, who invented the ratings technology that the US TV giants ABC, NBC and CBS were built on; to Charles Morton, who took the raucous entertainment of supper-clubs and taverns and developed the more mainstream and wildly popular Music Halls of Victorian England, from which came the talent that would dominate the early years of cinema and radio.

These entrepreneurs were not leaders, but listeners – their particularly skill was in realising that audiences were consuming culture in new ways, finding new ways to measure these new patterns, and new ways to make money out of them. The story of these ‘empires of attention’ is the story of how we – the audience – have engaged with culture,  and how the interaction between artists and audiences has moved from visceral participation to abstract measurement and back again. This story starts amidst the raucous popular culture of Victorian England.

Back in the 18th and 19th century, public entertainment was found in pleasure gardens, inns and taverns, with acts brought in to perform as the audience worked their way through dinners and rounds of drinks. ‘Song and Supper Rooms’ such as the Coal Hole, in The Strand, The Cyder Cellars in Maiden Lane or Paddy Green’s of Covent Garden offered food and drink till the small hours of the morning. London Gentlemen would go there to find foaming tankards of stout, a dinner and a cigar, all set to the warbling strains of the comic or sentimental vocalists attached to each establishment. The proprietor would act as ‘chairman’, leading the entertainment and calling on regulars to perform their favourite songs and comedy acts.

One regular act was the ‘judge and jury’ show, a thinly-veiled skit in the form of mock-trials of society scandals, with the audience and performers acting as barristers, jury and witnesses – in many ways the fore-runner of TV satire like That Was The Week That Was or Spitting Image.

The entertainment in these venues was a collaboration between the audience and the acts, with the line between the two often blurred by bonhomie and alcohol. The audience was noisy, and the acts used this noise as a feedback loop – a connection between performers and audience that created culture through call and response.

But this connection would start to fray as the 19th century went on. First the Theatre Regulation Act of 1843 relaxed the rules on theatre ownership, but on the condition that no eating, drinking or smoking could happen during a performance. The law was vague and poorly enforced, but it led to entrepreneurs like Charles Morton starting a new kind of entertainment venue, one that appealed not only to the male-dominated taverns and supper clubs, but women and families too.

Dining tables were replaced by rows of seats, the acts became more visual and spectacular through introducing scenery and sets, the participatory judge and jury shows were replaced by comedians and novelty acts, and the shouting and carousing moved from the auditorium to the bar in the foyer. Audiences became quieter, and stage lights were introduced to focus attention on the acts instead of the audience. Here’s a description of Charles Morton’s Canterbury Hall from 1858:

“We make our way leisurely along the floor of the building, which is really a very handsome hall, well lighted, and capable of holding fifteen hundred persons; the balcony extends round the room in the form of a horseshoe. At the opposite end to which we enter is the platform, on which is placed a grand piano and a harmonium, on which the performers play in the intervals when the professional singers have left the stage.

Let us look round us; evidently the majority present are respectable mechanics, or small tradesmen with their wives and daughters and sweethearts there. Every one is smoking, and every one has a glass before him; but the class that come here are economical, and chiefly confine themselves to pipes and porter. The presence of the ladies has also a beneficial effect; I see no indication of intoxication, and certainly none of the songs are obscene.”

By the late 19th Century, Music Hall was the dominate form of popular culture, and the audience experience had moved from the rowdy tavern to the theatre experience we would recognise now. The evening’s entertainment shifted with changes in popular taste, with opera and classical music giving way to variety and comic acts. The acts became increasingly professional – the comic George Laybourne was hired by Charles Morton for the princely sum of £20 per year – and they started to become well-known names, drawing a crowd on their own terms. But although these developments allowed the music hall entrepreneurs to build hugely valuable empires, they also sowed the seeds for Music Hall’s downfall.

At the beginning of the 20th century, cinema arrived with an even more spectacular form of entertainment than music hall, perfectly suited to the increasingly dark and quiet auditoriums. The transition took decades, with early films appeared as ‘acts’ within music hall itself, sharing the same comedians and singers popular at the time. But over time the longer feature film format became an evening’s entertainment by itself, and the music hall circuits and entrepreneurs – the ‘empires of attention’ of the 19th century – gave way to the new empires of cinema, and, by the mid-20th century, broadcasting.

This was a turbulent time for popular culture, with traditional business models struggling, and new technologies rapidly innovating formats and distribution models as audiences moved their attention to these new forms of culture. That might, perhaps, sound familiar to anyone here working in the media industry.

Throughout this transition, the one constant was the gradual breaking apart of the relationship between the audience and the artist. In 1850 popular entertainment happened in rowdy and participatory taverns and supper clubs. By 1950, the audience sat quietly in cinemas or at home, hearing and seeing entertainment that was recorded in a different time or space altogether. The feedback loop – the call and response of the music hall – had disappeared almost completely.

The rise of broadcasting brought new problems with understanding audience’s attention. Advertisers had no way of knowing how many people would actually hear their commercials, so without finding a way to measure attention, early radio broadcasters would struggle to compete for ad income with print and other forms of media.

In the early years of radio many techniques were used to measure audience attention, from sampled surveys and interviews to ‘radio diaries’ that asked listeners to record their own listening habits. For a brief period, fan mail was used as a way of indicating audience size and interest, often linked directly to the product sponsoring the show.

For example, Kate Smith was an early radio star with shows like ‘Kate Smith Sings’ and ‘The Kate Smith Matinee’ on CBS and NBC. Her show ‘Kate Smith and her Swanee Music’ was sponsored by La Palina Cigars, who ran a competition for listeners to send in cigar bands in return for a signed photograph. Over 50,000 cigar bands were submitted in response, demonstrating to La Palina the power of radio as an advertising medium.

But competitions like these were unpredictable and too reliant on the individual appeal of stars like Smith. Radio broadcasters understood that they needed to agree on a consistent measure of attention to convince advertisers to invest in radio, so they worked together to define a single measure across their industry.

This wasn’t just happening in radio- the period from the 1940s to the turn of the 21st century is known by academics studying the ratings industry as the era of one big number as the competition for advertising money between newspapers, magazines, radio, TV and outdoor media forced competitors in each sector to agree on how to measure their audiences. The result in every case was a single measure – the one big number that was used to represent the behaviour of millions of people.

Central to the developments of these measures was Arthur C Nielsen, who took his expertise in market research in the 1920s and 30s and developed new techniques to measure radio audiences in the 1940s. Nielsen ratings then became the main measurement of US TV from the 1950s, based around ‘Nielsen Sweeps Week’ – the periods during the year when Nielsen sampled the television audience, initially by sending out paper diaries for viewers to record their viewing, then later introducing boxes in selected viewers’ home that automatically measured the programmes they watched.

This data would be used to set local and national advertising rates, so broadcasters would fight for audience attention with cliff-hangers, ‘stunt-casting’ ( for example, bringing in a hollywood celebrity to guest on an episode) and special editions of popular shows, as in 1997, when the hit show ER broadcast an episode that was filmed and broadcast entirely live. In fact, the popular phrase for the moment when these desperate tactics mark a show’s decline is ‘jumping the shark’, based on a Happy Days special produced for Nielsen Sweeps Week in 1977 when, in a desperate bid for attention, the show writers wrote an episode where The Fonz jumped over a shark-pen on water-skis.

Most of the multi-billion dollar ad spend around the world today is bought and sold based on ideas of audience measurement that can be traced back to Arthur Nielsen and his early work in the fledgling radio industry. There is probably no single person who has been more influential in the creation of the global media industries – the ‘empires of attention’ – of the 20th century, yet he is nowhere near as well known as the writers, directors, performers and executives of the media empires built on his work.

But for artists and performers, these numbers were a thin and ineffectual connection to the audience. The rich engagement with the audience in live performance had now become a series of numbers on a report. During the late twentieth century – the ‘golden era’ of broadcasting – creators would sometimes not even see these audience reports, and would have to find other ways to get feedback.

In a documentary about the making of Blackadder, Director Richard Curtis said that in the early 80s, he never saw the ratings for Blackadder early series, and still to this day does not know how many people watched. He said he used to walk the streets of Shepherd’s Bush during the broadcast, looking in people’s windows to see if they were watching, and if so, whether they were laughing, as it was the only way he could know the audience’s response.

It’s a wonderful anecdote, but also a telling illustration of how far the gap between audiences and artists had grown in the 20th century – from the communal rowdiness of the supper clubs and taverns to the silent and invisible audiences of the broadcast era.

By the time I started working for the BBC in 2001, multiple generations of TV executives and creatives had only ever understood audiences as numbers. Ratings had become an obsession, but real connection with audiences was rare.

I remember being at one of Greg Dyke’s awaydays for senior BBC management during his time as Director General, where senior executives were asked for their opinions about the lives and tastes of sample members of the BBC audience. At the end of the session, there was a theatrical reveal as the real people we were discussing were brought into the room, invited to sit on our tables and talk to the BBC executives about the kind of programmes they watched. I remember the extraordinary sense of surprise and novelty amongst executives that we were meeting actual members of the public, and I felt a bit sorry for the people brought into the room as if they were some kind of exhibit. It made me realise how far TV industry executives were removed from the audience who enjoyed their content – they obsessed over abstract representations of audiences, but were almost speechless when they were brought into the same room.

This made me curious about how we’d ended up like this, how audiences and artists had ended up so far apart. I started researching the history of how we measure attention, and how this affects the way that we make and experience culture. I discovered that the last 50 years or so have been an unusually quiet and stable period – the dominate forms of culture, the business models behind them, and most importantly the way we measured audiences’ attention, hadn’t change that much at all.

If you grew up in this period, you experienced mainstream culture as a passive member of an audience, either sitting at home watching TV, or in a darkened cinema, with the exceptions being live music or comedy, which were mainly aimed at younger audiences. I remember seeing a chart at the BBC showing how the volume of TV watching changes as we age. The curve starts with a few hours a day as children, then drops sharply in our late teens and twenties as we go out to parties, gigs and pubs to get drunk and find mates. After we settle down as couples and families in our twenties and thirties, TV viewing goes up again – the most voracious TV viewers in the UK are not the advertisers‘ main targets of 16-34 yr olds, but the middle aged and retired.

A friend who works in digital media once asked a senior UK TV executive whether he was worried by younger audiences spending most of their time on social media and gaming consoles instead of TV. “Don’t worry”, the executive replied, “they’ll get older and more tired and then they’ll want to just flop out on the sofa and watch TV”.

Perhaps these younger audiences will return to TV as they get older, but with a new and very important twist. For the last 10 years has seen the return of the noisy crowd. The invisible voices of the audience are now being heard again over twitter, facebook and other social networks. When I moved from the BBC to Channel 4 in 2007, Twitter had only just launched and ratings were the only measure commissioners cared about. By 2011, when I left, nearly every commissioner knew how to search for the number of tweets alongside their shows, and the most followed people on Twitter were the actors, Presenters and Performers execs were commissioning to make TV. The feedback loop between audience and artists had, in a very new and odd way, been reconnected.

The new entrepreneurs of attention in the 21st century understand this new connection- they understand that culture spreads not by distribution – as with cinema and broadcast – but by circulation – sharing between friends over digital networks.

One of the most influential cultural entrepreneurs of the early 21st century has been Jonah Peretti, the co-founder of the Huffington Post and Buzzfeed. Peretti’s understanding of how digital culture spreads started in 2001, when Nike launched a ground-breaking online campaign that let people stitch personalised messages onto their trainers. Peretti, at the time a student at MIT in Boston, chose the word ‘sweatshop’, and when Nike refused his order, he started an email exchange to find out why. He sent this story to a few friends, and they in turn circulated it to more people, sharing it on email lists, websites and forums. Within a few days it had been read by millions of people and Peretti was on the TODAY show on NBC, talking about both the story and the interesting new way it had spread across the internet.

The Nike experience made Peretti curious about how stories spread through digital circulation, so he went around MIT asking for help to understand how it worked. The first factor was the science of networks and the number of nodes – or people in this case – you need for stories to become ‘contagious’ and spread widely. The second factor was cognitive science – why do we decide to share stories with each other? What kind of emotions – laughter, shock, outrage – drive us to circulate stories around our friends, and what are we saying about ourselves and our communities when we share stories?

Peretti called this ‘the social imperative’ – the ‘hook’ that drove people to share something they’d found with their close friends. After more experiments with creating viral culture he was hired by Arianna Huffington in 2003 to launch The Huffington Post. Derided at its launch as a vanity news project, its mixture of traditional news content and ‘meme’ culture grew an audience of 13.2m unique visitors by 2011, when it was acquired by AOL for $315m. By this time, Peretti had already launched Buzzfeed, an even purer distillation of circulation friendly content, with endless streams of list articles (called ‘listicles’ by Buzzfeed staff) amusing cat pictures and celebrity gossip. In July 2013, Buzzfeed had an audience of 37.9 million unique visitors, not far from the Huffington Post and mainstream media brands like the New York Times and CNN. Peretti had succeeded in creating not one, but two of the biggest ‘empires of attention’ in the 21st century. In a memo to staff in September 2013, he emphasised how this success was different from the broadcast media empires of the 20th century:

“There are many exciting, tempting, glamorous, lucrative opportunities that we will NOT do in the coming year and as more of these opportunities present themselves it will take discipline to stay on track. We will NOT launch a BuzzFeed TV show, radio station, cable network, or movie franchise — we’ll leave that to the legacy media and Hollywood studios. We will NOT launch a print edition or a paywall or a paid conference business — we’ll leave that to other publications. We have a great business model that has a bright future as social and mobile continue to become the dominate form of media consumption. We will stay away from anything that requires adopting a legacy business model, even a lucrative one like cable syndication fees or prime time television ads. What seems like a lucrative opportunity today is often a distraction from building something much more exciting tomorrow.” 

That last line is a stark illustration of how the empires of attention are shifting as we move from an era of distribution to an era of circulation. Peretti’s memo lists a series of traditional media business models that BuzzFeed will not be following, and then closes by emphasising that they aren’t lucrative opportunities but are actually distractions from his core goal – inventing a new kind of media empire for the 21st century.

Peretti is a building a new Empire of Attention – one that synthesises the call and response of 19th century music hall and the incredible scale of 20th century broadcast distribution. The combination is a potent one – the sheer visceral impact of thousands or millions of people sharing and discussing your stories is a new experience for anyone used to traditional broadcast media, and we’re having to learn how to tell stories in an age of digital attention. We’re already hearing TV commissioners complaining that knee-jerk responses from audiences on Twitter are killing new TV shows before they have a chance to build an following. We are no longer a passive audience, but the judge and jury of what will survive and be recommissioned, deciding the fate of culture by how we spend our attention.

This new feedback loop can be incredibly empowering, but it is also destructive – the anonymity of social media can encourage trolling and other kinds of abuse. Crowds amplify the good and the bad in human behaviour, and the internet amplifies this even further. But I don’t think it’s possible to have one without the other – the noise is also the signal, and we will have to develop new ways to tell stories that take this into account.

The culture of the 21st century will be defined by how we synthesise these contradictions – scale and intimacy, spectacle and conversation, signal and noise. We have seen the relationship between audiences and artists move from intimacy to distance, and now back to a strange kind of intimate distance. What will culture look like in an age of digital attention, and what new empires will emerge around it? How we will we measure attention, and how will this change the relationship between artist and audience? Who will be the Charles Morton, the Arthur C Nielsen, or the Jonah Peretti of the next 50 years?

So, to finish, thank you again for your attention. Thank you to those of you here in the room, for your attention is the feedback loop that made it possible to tell my story. For those of you listening at home or online, we can now use the feedback loop of social media – I’m @matlock or @storythings on twitter. Thank you again for your attention, and I look forward to giving you some of my attention in return.

A couple of years ago, when I started working at Channel 4, I came up with a model for thinking about social spaces online that focused on how users felt about being online, rather than the technical capacities of the platforms themselves. I did it so that people pitching to us would think about users rather than tech, but the model has stuck in my thinking, and seems more and more relevant today.

This is partly down to something that I thought about at the time, but didn’t write about. In the original post, I tried to describe a crude taxonomy of spaces in which users have relatively consistent expectations about what happened to the information they share, and relatively consistent behaviours that they expected from other actors (both real and technological) when they occupied that space. For example, if a space feels like its shared only by a group, users will share their information accordingly, and expect others to share their assumptions.

What I didn’t really talk about at the time was transgression – when a platform doesn’t meet the expectations of its users, or when other users move information from one context to another without permission. For example, Matt Mckeown’s infographic showing Facebook’s shifting privacy policy is a great example of platform transgression – a technology shifting information from one register to another without clearly signposting this transgression to the user.

Likewise, user transgression is when someone shifts someone elses information from one register to in a way that wasn’t expected. A common illustration of this is newspapers taking photographs from Flickr without respecting the copyright limitations that users had put in place when uploading the photo. Loaded magazine was recently cleared of breach of privacy by the PCC following a complaint from a woman who uploaded a picture of herself to Bebo in 2006. Over the next few years her picture was circulated widely on forums, and she became an internet meme as the ‘Epic Boobs’ girl. When Loaded magazine called for their readers to help track her down, she claimed the article had caused her considerable upset. But the PCC claimed that as the picture was so widely distributed online already (appearing in the top 3 Google searches for ‘boobs’) the Loaded article could not be considered to infringe her privacy, although it would have been a different case if they had taken it directly from her Bebo profile in 2006. It was the gradual disemmination of her image between groups of users online that made it ‘public’ – not her original act, which she probably imagined to be for a group that she controlled, but groups who could access and share her image without her knowledge or control.

What is remarkable about the Epic Boobs and Facebook transgressions is that they are gradual and hard for the person involved to track. In an analogue media world, the transgression between registers is sharp and obvious – a newspaper would have had to contact you to get a copy of a photo for them to use, and your personal photographs couldn’t become a global property without you knowing about it. We now live in an age where transgression is insidious and invisible, where users can’t understand the potential risks of sharing until it’s caused them significant pain.

Understanding trangression is going to be *the* most important thing for business and users working online in the next few years. Users will need to interrogate the services they use for potential transgressions of their information across contexts (as with Facebook’s gradual publicising of user data); platform creators will have to be more explicit to users about how information transgresses different contexts, and make these transgreses more tangible to the user (simply ticking check boxes is not enough – these transgressions need to have grain and weight built into the interaction); and large organisations will need to understand the implicit and assumed contexts of the spaces they are using to connect to their users, and how to ask permission when they take contributions or data from one context to another.

We’ve been through nearly a decade of excitement about creating and scaling these new social spaces online. We now need to focus very clearly on how information moves between them, as these transgressions are not simply about data and networks. The boundaries that users understand implicitly are defined by emotions, not software, and we need to bear this in mind when we cross them.