Felix Salmon

Matthew Ball has a long examination of the economics of Netflix’s original content, looking at it on a show-by-show level. He starts with the cost of producing something like Arrested Development, and then works out how many extra subscribers Netflix would need to attract in order to justify that cost. (Or, how many extra months existing subscribers would have to keep their subscriptions for, compared to when they would unsubscribe otherwise.) He writes:

I’d argue that it is unlikely that Arrested Development will convince millions of users to stay an extra month in 2014 and 2015. If this is the case, the show would need to achieve its return in the immediate future. Therefore, if we don’t see Netflix adding four to five million new subscribers during the quarter, one of two things are true. One, the show was a poor investment whose draw was a fraction of those anticipated, or two, the show is instead intended to convince many of the million subscribers currently churning away each month to defer their cancellation. This would be telling.

While Wall Street analysts are assessing the success of original content in terms of new customers, I believe Netflix’s primary goal is on imminent service cancellations.

Ball lists three reasons why Netflix is making original content. There’s the way in which that content keeps people subscribing for longer; the way in which original content will allow Netflix to raise its prices in the future; and then there’s this:

Hedging against rising content licensing costs, which are up 700% over the past two years. While per-show licenses will never surpass the cost of original producing a series, their increases will make ongoing investments in House of Cards less expensive on a differential basis.

The ever-increasing cost of licensing is a huge issue for Netflix, and it’s the reason why its business model is a very tough one: any time that Netflix builds up a profit margin, the studios will simply raise their prices until that margin disappears. Netflix had to pay a whopping $1.355 billion in licensing costs just in the first quarter of this year; that number is only going to increase, unless Netflix can find some other way of finding content. Like producing it in-house. At the margin, the more material that Netflix produces on its own, the less it needs from third parties, and the easier that Netflix finds it to say no to ridiculous demands.

But what Ball misses, I think, is that Netflix is playing a very, very long game here — not one measured in months or quarters, and certainly not one where original content pays for itself within a year. Netflix doesn’t particularly want or need the content it produces in-house to make a profit on a short-term basis. Instead, it wants “to become HBO faster than HBO can become Netflix,” in the words of its chief content officer Ted Sarandos.

Most importantly, the thing that Netflix aspires to, and which HBO already has, is an exclusive library of shows. If everything goes according to plan, then the Netflix of the future will be something people feel that they have to subscribe to, on the grounds that it’s the only place where they can find shows A, B, C, and D. That’s what it means to become HBO — and Netflix is fully cognizant that this is a process which takes many years and billions of dollars.

If Netflix gets there, then it becomes a license to print money, just as HBO is today. Shows like Arrested Development and House of Cards may or may not pay for themselves over the short term — in fact, they almost certainly won’t. But that doesn’t matter. In the long term, they will become part of a library which has massive value on two fronts: the shows can be licensed out in jurisdictions where Netflix doesn’t want to compete, and they will also help make Netflix a service that can guarantee you a great show that you want to watch, whenever you want to watch it.

Ball says that “Arrested Development is an established brand that’s intended to be a one-off event to convince its fanatical (and tech-savvy) followers to give Netflix’s broader streaming service a try.” That’s true — narrowly. But the series is much more than that: it’s also a way for Netflix to signal to all its current and potential subscribers that it is home to high-quality exclusive content, if and when they ever feel like giving it a try. In a weird way, Arrested Development is worth more as the number of people who haven’t seen it goes up.

No one today is likely to subscribe to Netflix just on the grounds that they think they might like to watch Arrested Development at some point. But when there are dozens such shows — none of which are available anywhere else — that begins to add up. At that point, not only does Netflix provide something for everybody; it also becomes the only place to watch certain shows with cultural-touchstone status. And presto, the decision is no longer whether Netflix is worth the subscription price; rather, the question is whether you can afford not to have it.

There’s no guarantee that Netflix is going to succeed at this strategy: many have sailed into the treacherous waters of Hollywood video production, and few have thrived there. And in the first instance the strategy just means that it’s no longer just the content companies managing to extract enormous rents from Netflix; it’s the production industry and the talent as well. The old argument still applies, mutatis mutandis, to the new strategy: as high-quality original content becomes increasingly important to Netflix, Hollywood will find ever more ingenious ways of forcing Netflix to pay through the nose for it.

Still, for viewers, this can only be good. The viewing audience doesn’t care whether Netflix makes money: they just want great shows to be produced. If they like House of Cards and Arrested Development, they should be very heartened: there’s going to be a lot more new shows where those ones came from.

Many thanks to Peter Rudegeair for pulling this data from Factiva: it shows how often the word “billionaire” has appeared in headlines from 1984 to date. We’re looking, here, only at three publications: the NYT, the WSJ, and Reuters News. (And we’re excluding Bloomberg, which has its own dedicated billionaires section.) For 2013 the dark-blue line is the year-to-date figure; the light-blue line is what happens if you extrapolate what we’ve seen so far to the year as a whole.

There’s a billionaire bubble here: we’re on track to see 158 billionaire headlines this year, or more than three a week. In most of those headlines the term is entirely gratuitous: “Cyprus rescinds citizenship of Assad billionaire cousin”, for instance. In many cases, it’s not even entirely clear that the people being referred to are billionaires.

Back in 2006, 59 appearances of the word “billionaire” in headlines was an all-time high — but that record was obliterated in 2007, when it popped up 103 times. Never again would we see anything near 2006′s relatively modest total, not even in the depths of the Great Recession. And now, it seems, there’s no limit to how high the totals can go. Can anything burst this bubble? Please?

Two years ago, when I wrote about the death of blogging, I contrasted the decline of old-fashioned reverse-chronological blogs with the huge success of Twitter and, especially, Tumblr. Since then, of course, Tumblr has been sold to Yahoo for more than $1 billion, while Twitter is reportedly valued at some ten times that amount. Clearly there’s a lot of value in becoming a successful publishing platform.

One big reason for the success of Twitter and Tumblr is that they have a very clear idea of what their product is, and they focus on making that product great rather than on being all things to all people. Blogging was historically thought of as a one-stop shop: the place where people would post a series of things, and where those things would appear in reverse chronological order. Every so often, a blogger would write something elsewhere, usually for money, and then link to that piece from their blog. But when Twitter came along, something changed: bloggers would write some things on their blogs, and other things on Twitter.

After I realized that Twitter had powers that no individual blog could ever have, I started thinking increasingly about the best platform for any given thing that I wanted to do. Photos of my friends and family of course are easy: they go on Facebook. Less personal photos go on Instagram. And if something could be distilled into 140 characters or less, there’s a good chance that it belonged on Twitter. But Tumblr was great too, for certain things: I’ve now posted to my Tumblr 1,159 times, and have 36,475 followers there. I haven’t spent much time making it pretty or coherent, but short-form content, especially if it’s visual and doesn’t work on Twitter, is often perfect for Tumblr.

There are lots of other platforms too: I’ve seen people like Jerry Saltz get really good at Facebook, although I’ve largely shunned it as a publishing tool. Bloomberg’s Tom Keene is very active on LinkedIn. I had a YouTube channel for a while; that was a lot of fun, and allowed me to do things which don’t necessarily work in print very well. For conversations, podcasts can be wonderful, as can live events. And if I need it, there’s always my own personal blog. Basically, different types of communication require different platforms, and it’s silly to expect everything to fit into one template.

This morning, I posted some thoughts about CitiBike on Medium — which is also the place I chose to publish my big piece on bitcoin. I like how clean Medium is, and it’s quite well suited to content which is longer than a Tumblr post but also maybe not the kind of thing which I’ve started to concentrate on here at my Reuters blog. Don’t ask me to define exactly what I put where: a lot of these decisions are spur-of-the-moment things, and sometimes I change my mind at the last minute. (This post, for instance, was drafted in Medium before I decided to move it over to Reuters.) But I really do love the ability to create and present the stuff I create in the format which puts it in the best possible light.

The new Reuters site is based around the concept of streams, and the Felix Salmon stream on the site, which will appear at some point in the coming months, is by its nature going to be a different animal, on a different platform, than the current Felix Salmon blog on Reuters.com. I don’t know exactly how it’s going to work — there will be a lot of trial and error involved, for sure — but right now the way I’m thinking about it is as a place where the various things I do on various platforms can be brought together in a single place. If I write something on Medium, or if I write an article for Wired, or if I post something on Tumblr, or if I create a video on YouTube, that thing can live in its own natural habitat while still being aggregated on my own Reuters platform. And of course I’ll always be writing a lot of original content right here — content which in turn will end up being linked to from Twitter, or syndicated on Seeking Alpha, or otherwise shared around the social web.

To put it another way: the death of the blog was really just the death of a single template into which all of a certain person’s output had to be able to fit. Now we have a multiplicity of options, and it’s silly not to take advantage of them. Media organizations have generally embraced their journalists publishing ultra-short pieces on Twitter; the future, I think, is going to be ever further in that direction. Certain journalists will be wonderfully active on Pinterest; others will develop huge followings on LinkedIn. As they do, their employers will be gifted with a brand-new way to extend their brand out to people they never reached before, in what feels like a very personal manner.

I remember having drinks with one editor-in-chief, a couple of years ago, after a NYT op-ed I had written received wide attention. He said that he would be furious if one of his writers had done such a thing: he had a clear expectation that his own publication would be the only place that his writers published pretty much anything longer than 140 characters. That expectation didn’t make much sense to me in 2011, and it just seems silly now. The same NYT op-ed, published as a Reuters blog post, would have been a very different animal: the medium helps create the message, as well as the audience that it reaches. Media companies should be in the business of curating and publishing what works best on their own platforms, rather than becoming jealous of what appears elsewhere. Indeed, many of them publish too much, and would be well advised to publish much less than they do.

Everybody is a curator, these days: publishers design platforms for certain types of content, editors shape publications by deciding what to leave out; journalists try to make sure that the stuff they’re doing is expressed to its best possible effect on the best possible platform. The result is a more fluid media ecosystem than we’ve been used to, but also a more effective one. Let content live where it works best; that way, the publishers of that content will be able to present something with maximal coherence and a minimum of feeling that they’re trying to do something they’re not particularly good at. The publishers who win are going to be the ones with addictive, compelling, distinctive content. Rather than the ones who are constantly flailing around, trying to copy everything that’s good somewhere else.

Amidst all the positivity coming out of Yahoo and Tumblr, any self-respecting pundit is going to want to pour cold water on the whole deal. Especially since billion-dollar mergers almost never work out very well. But here’s the weird thing: the more I look at this tie-up, the more it makes sense to me.

Yahoo has more than enough money to pay for Tumblr in cash, which is exactly what it’s doing. Here’s one easy way of seeing why this is a good deal for Yahoo: profitable tech companies (think Google, or Apple) tend to have too much cash and tend not to know what to do with it — with the result that it just sits there, uselessly. For Yahoo, having $4.4 billion in cash plus Tumblr is clearly going to be better in terms of the future than having $5.5 billion in cash, waiting interminably for some kind of Godot to come along and be bought. $1.1 billion is a lot of money, but it’s not so much money that it’s going to change the way that investors look at Yahoo’s balance sheet.

More fundamentally, Yahoo is acquiring 300 million young, mobile users at a stroke — along with invaluable information about what they like to consume online, and how they like to consume it. It’s a four-fer, in fact.

First, there’s the immediate traffic boost.

Second, there’s the ability to use Tumblr’s data to help optimize the rest of Yahoo’s pages. If I’m logged in to Tumblr, as I normally am, then when I go to Yahoo, I should see the kind of material that Tumblr knows I’m interested in, rather than some one-size-fits-all generic home page.

Third, Tumblr is Marissa Mayer’s opportunity for a Flickr do-over. The big portals have been extremely bad at building out genuinely interactive properties in the age of self-expression, and Tumblr knows how to attract a new generation of users who want to create rather than just consume.

And finally, Tumblr is the perfect platform for Yahoo’s brand advertisers to use if they want to start building up relationships with consumers, rather than just bombarding them with banner ads. (My friends at Percolate, an official Tumblr partner which was designed to solve this stock vs flow problem, are incredibly well placed to be huge winners from this deal.)

From Tumblr’s point of view, founder David Karp has extracted many promises from Mayer that she will leave the company alone, in New York City, to do its thing. “We’re not turning purple,” he says. More importantly, Karp can now outsource to Sunnyvale a lot of the gnarly monetization problems which the NY team was only slowly beginning to solve. The plan right now — which might change — is to give Tumbloggers the option to start running ads on their sites, presumably with some kind of revenue-sharing deal. But from day one, Yahoo’s sales team could simply start insisting that any brand wanting to buy ad space on the Tumblr dashboard will also have to buy a bunch of space on the Yahoo network. That’s a great way of leveraging the amount of money that Tumblr brings in.

And then of course there are the itchy VCs: Tumblr raised its first money back in 2007 at a $3 million valuation, resulting in a glorious 365X return for early investors including Fred Wilson and Jacob Lodwick.

Lodwick is on the record saying that acquisitions like this one are always a failure for the company being bought: “Big companies aren’t just big versions of small companies,” he writes. “They’re another class of entity entirely, more concerned with sustaining their own rhythms and control structures than experimenting with strange ideas from acquired ex-founders.” But part of the deal you make, when you accept VC funding, is that there will almost certainly be an exit within 5-10 years, and it will almost certainly not be an IPO where the founder retains control. This kind of exit, where the company is big enough to retain a modicum of independence, is the least bad outcome that Karp could realistically achieve.

It won’t be easy: as Peter Lauria points out, Yahoo’s decision to ban Kara Swisher and Peter Kafka of All Things D from the press announcement is exactly the kind of heavy-handed corporate meddling that Lodwick is talking about. And Tumblr’s users are predictably unhappy about the whole thing. But Yahoo certainly has the tools to help boost Tumblr’s flattening traffic numbers, while Karp should be able to retain enough control of Tumblr that his users don’t revolt entirely. After all, it’s far from clear where else they could go.

Most mergers fail, and this one could fail as well. But on the spectrum from “obviously doomed” (NewsCorp/MySpace) to “obviously sensible” (Google/YouTube), I’d put Yahoo/Tumblr well within the “sensible” half. Which is rare enough to be noteworthy.

All social networks are based on a cognitive con. No one likes to give out valuable personal information to some huge corporate entity, so the trick is to make people feel as though they have some kind of control or ownership, even when they don’t. Most of Facebook’s privacy controversies boil down to much the same thing: people share personal information with their friends, using the convenient Facebook platform, and then are shocked when it turns out that Facebook has access to that information and is making money from it.

The more centralized and controlling a social network is, and the less that it’s run on a peer-to-peer basis, the more likely it is to run into this kind of trouble. So it probably should come as little surprise to learn that Bloomberg, the highly-centralized and highly-controlling social networking company, has now run headfirst into its very own privacy scandal. (Bloomberg is a competitor of Thomson Reuters, which is my employer and owns Reuters News.)

The Bloomberg terminal is a take-it-or-leave-it proposition, as far as its users are concerned: they either sign on the dotted line, pay their $20,000 per year, and get their terminal — or they don’t. Just as with Facebook, the Terms of Use are non-negotiable: unless you agree to them, you don’t get to use the service. And as everybody who’s ever tried to put a naughty word into a Bloomberg message knows, once you’re signed into the Bloomberg system, Bloomberg is watching you.

Zach Seward‘s sources say that Bloomberg logs every keystroke of every customer; Bloomberg declined to comment to Seward and declined to comment to me, so it’s hard to know what the truth is. But it’s well known that Bloomberg accumulates a truly enormous amount of information; right now, for instance, it says it is hiring big data architects in an attempt to manage it all. That dataset is one of Bloomberg’s great competitive advantages.

Indeed, Bloomberg’s clients in many cases may want it to implement Panopticon-style monitoring of everything their employees do. As Seward says, “at Bloomberg, omniscience is a feature not a bug”: even if the individual employees aren’t enthusiastic about losing all their privacy, their employers, institutionally, see a lot of upside, in terms of compliance and risk management, in keeping a record of everything that their workers do online. If Bloomberg will help do that work for them, included in a terminal subscription they’re going to pay anyway, then so much the better.

Besides, both Bloomberg and its clients have an aligned incentive when it comes to making the terminal as excellent as possible in terms of giving subscribers the information they want as quickly and comprehensively as possible. You can’t do that unless you know how clients are using the terminal — what they’re looking up, where they’re getting frustrated, where they spend most of their time. By sharing usage data and trusting Bloomberg to keep it confidential, subscribers can help make the product even better.

But journalists are a special case. As Bloomberg editor in chief Matthew Winkler says, Bloomberg’s “reporters should not have access to any data considered proprietary” — and it is “inexcusable” that they did. The problem derives from Bloomberg’s in-house news organization priding itself, at least at the outset, on being deeply embedded into the broader company, and making the maximum use of the information on the Bloomberg terminal.

Bloomberg News’s Kevin Reynolds, for instance, talking to Brill’s Content in 2001, boasted that “as a reporter here, you have knowledge going into the interview that your competitors don’t even have at the end,” thanks to the information in the terminal. And in case there was any doubt that, as Amy Chozick writes, “the news operation was assembled in the 1990s primarily as a way to sell more terminals,” it was laid to rest by Mike Bloomberg himself, in his autobiography. “Most news organizations never connect reporters and commerce,” he wrote in Bloomberg by Bloomberg. “At Bloomberg, they’re as close to seamless as it can get. That’s our system.”

In that context, it was absolutely natural for journalists to have access to the same data being seen by the sales staff: at Bloomberg, the reporters were deliberately tied as closely as possible into the commercial terminal-sales function. Too closely, it turns out. As times changed, and, in Winkler’s words, “as data privacy has become a central concern to our clients,” it became necessary for Bloomberg’s journalists to be completely removed from client data. They weren’t, and, as Bloomberg CEO Daniel Doctoroff writes, “although we have long made limited customer relationship data available to our journalists, we realize this was a mistake.”

So, Bloomberg says it made a mistake, it has apologized, and it is not going to happen again. End of story? Not entirely. For one thing, the Europeans have pretty strict privacy laws, and are talking to Bloomberg about what happened; no one knows how those talks could conclude. On top of that, for all that Winkler’s apology runs under the headline “Holding Ourselves Accountable,” so far there have been no reports that anyone at Bloomberg is being held accountable. A Bloomberg spokeswoman declined to comment.

Bloomberg’s reporters use the Bloomberg terminal for everything they do: they’re an inextricable and central part of the Bloomberg social network. And while the newsroom has now lost its access to certain functions, the company would not comment on the degree to which the changes are affecting the vast majority of Bloomberg employees who don’t work in the newsroom. For the time being, it seems, thousands of Bloomberg employees around the world have retained their access to key information about employees of Goldman Sachs, the Federal Reserve, the ECB, the US Treasury, and countless other organizations — information which, in many cases, is fiercely protected even within the organizations themselves. (If an employee has been quietly suspended and is no longer actively working for the organization in question, that’s not going to be common internal knowledge, but it’s easy to see if you can see when they last logged in to their Bloomberg.)

Many of Bloomberg’s clients, especially the Europeans, are likely to be unhappy about the fact that such sensitive information could continue to be widely available within the company. But, just like participants in other social networks, they don’t have a lot of choice in the matter. The more time you spend on your Bloomberg, the more value you get out of it — and the more that Bloomberg staffers are going to know about when and how you work. That’s been the bargain from the beginning, whether you liked it or not.

Back in December 2011, the Daily Mail had 45.3 million unique visitors, according to ComScore. By March 2013, 15 months later, that number had grown to 46.4 million, again according to ComScore. We learn the latter figure — but not the former — from Christine Haughney’s article about the website today:

For now, many analysts consider the Mail Online a growth source for a strait-laced media company. The parent company’s total annual revenue is about $2.7 billion and its net income is $466 million. It depends on newspapers for about 20 percent of its profits, according to Mr. Reynolds.

Alex DeGroote, a media analyst with Panmure, Gordon & Company, said that while Mail Online was still not profitable, its growth had helped its broader company’s stock price grow roughly 80 percent in the last year.

This is crazy. For one thing, as we’ve seen, the Mail Online isn’t really growing: on the internet, 2.4% growth in 15 months is decidedly weak. For another thing, as DeGroote points out, it isn’t making any money. But more to the point, DMGT, the parent company, is so enormous that Mail Online can’t possibly account for the rise in its share price.

DMGT’s market capitalization is $2.84 billion; it has risen some $1.26 billion in the past year. If DeGroote really thinks that Mail Online accounts for a significant chunk of that growth, he would have to think that the rise in the value of Mail Online, just in the past 12 months, has been the best part of a billion dollars. By that logic, given that the site was already extremely popular a year ago, the overall value of Mail Online would probably have to be more than the entire value of its parent company.

In reality, the value of DMGT has almost nothing to do with Mail Online. The site might be a traffic powerhouse, but the internet is full of high-traffic sites which are worth very little. Traffic, in and of itself, is worth very little, and there’s no indication that readers are willing to pay for Mail Online, or that advertisers are willing to pay much for those readers. (The site’s revenue of $7.2 million is about 0.25% of DMGT’s $2.7 billion total revenue.)

DeGroote, here, is falling victim to the visibility fallacy: Mail Online is by far the most visible part of DMGT’s business, and so he thinks that it must account for most of the change in its share price. In reality, however, if DMGT decided to shut down the entire site tomorrow, its value would probably barely be affected. If you want to find a reason for why DMGT’s share price has performed so well of late, you’re going to have to look elsewhere.

Brad DeLong has found a 1995 article by Niall Ferguson which pretty much puts the lie to Ferguson’s claim about his take on John Maynard Keynes. Here’s what Ferguson now says:

My disagreements with Keynes’s economic philosophy have never had anything to do with his sexual orientation. It is simply false to suggest, as I did, that his approach to economic policy was inspired by any aspect of his personal life.

And here’s what he wrote in 1995:

Research in German archives shows that Keynes’s critique of the Versailles Treaty was based on anything but dispassionate economic analysis. Few, if any, of its readers can have appreciated how far the ideas contained in The Economic Consequences of the Peace — the book which made him a celebrity — were actually inspired by members of the German peace delegation at the Versailles conference. Still fewer knew that their appeal to him owed as much to his homosexuality as to his Germanophilia.

Ferguson writes that “the attraction Keynes felt” for the German representative Carl Melchior “strongly influenced his judgment”, and adds for good measure that “those familiar with Bloomsbury will appreciate why Keynes fell so hard for the representative of an enemy power”. Here’s the whole thing:

This is a slightly different argument, of course, to the idiotic remarks Ferguson made at a conference in California last week, where he said that Keynes didn’t care about future generations because he was gay and didn’t have children. If those remarks were, in Ferguson’s own words, “doubly stupid”, then maybe his Spectator article is maybe only singly stupid. Except it was carefully written, edited, and committed to print: Ferguson can’t claim that his article was merely a regrettable “off-the-cuff” error.

At first blush, Ferguson’s apology is full and unqualified. But in light of this and other information, it seems that Ferguson has rather more to apologize for than a single verbal response to a question from Paul McCulley. Either Ferguson still believes today what he wrote in 1995, or else he has changed his mind and now believes that what he wrote back then is “simply false”. It’s about time he clears this up.

Update: It seems we can’t take Ferguson’s apology at face value after all. In an ill-tempered letter written to the Harvard Crimson, Ferguson says that he can’t be prejudiced because he has a Somalian wife and a gay friend; says that “the strong attraction Keynes felt for the German banker Carl Melchior undoubtedly played a part in shaping Keynes’ views on the Treaty of Versailles and its aftermath”; and adds that Keynes and the other members of the Bloomsbury Group “had no doubt at all that sexual orientation had a significance beyond the narrow confines of the bedroom, and that intellectual life and emotional life were intertwined”. I guess it’s not “simply false” after all to suggest that Keynes’s approach to economic policy was inspired by any aspect of his personal life.

One of the funnier subplots in the media universe these days is the one about Aereo. Aereo is the kind of company which sounds like a thought experiment, but it’s very real: it takes free broadcast signals, uploads them to the cloud, and rents them out — at a fee — to people who want to watch broadcast TV on their computers. It’s a way of showing the broadcast networks how silly it is that they don’t put their programming online, and it’s also an argument for why cable companies shouldn’t have to pay through the nose for the right to retransmit content which has always been free-to-air.

Real-world companies are largely immune to thought experiments, however, and so it was only when Aereo started operating in the real world that the court cases and ultimatums started being thrown around. If Aereo isn’t shut down, say the broadcasters, they might have no choice but to take their networks off the air entirely. This of course would effectively kill Aereo, whose CEO is rather desperately drawing an analogy between the right to receive broadcast TV and the right to vote.

“The real question is a consumer question: Can you rightfully disenfranchise 50 million consumers?” he asked. “Is that what the preferred policy is?”

In the event that the networks did go through with it, he speculated that other programmers would be quick to replace them in the role of public broadcasters. “That spectrum is incredibly valuable. Somebody’s going to take advantage of that,” he said.

The 50 million number, by the way, should not be considered particularly reliable: it’s Aereo’s guess as to the number of people who ever watch free-to-air TV, even if they mainly watch cable or satellite. (Maybe they have a hut somewhere with an old rabbit-ear TV in it.)

But Aereo is absolutely right that America’s broadcast spectrum is incredibly valuable. The problem is that it’s much more valuable to cellphone companies than it is to broadcasters. The government has a plan to start a series of cleverly-designed auctions, whereby broadcast spectrum would end up being bought from broadcasters and consolidated in the hands of wireless-data companies who value it more highly. That plan can’t be put in place too quickly: the fact is that we’re living in a world where TV broadcasts create much less value than wireless companies could realize with a fraction of the bandwidth.

At the same time, broadcasters are realizing that their retransmission revenues are significantly more valuable than the marginal advertising revenues they get from households which are still reliant on rabbit ears. That trend is only going to strengthen going forwards, especially given that most new TV sets can’t even receive broadcast signals in the first place. What’s more, broadcasters can give themselves a little extra leverage if they shut down their free-to-air service (and Aereo). Once that happens, then if they refuse to provide retransmission rights during negotiations over retransmission rights, the cable companies’ customers will be cut off from their content entirely.

None of this is going to happen quickly, or cleanly. But broadcast TV is rapidly becoming an obsolete technology, and the distinction between cable channels and broadcast channels is a distinction which has outlived its usefulness. Aereo’s very existence is testimony to the silliness of the status quo, and the logical end point is for all the current broadcast spectrum to end up in the hands of institutions which can use it much more effectively as digital bandwidth.

The losers in this process will be Aereo, of course, and also the households which still rely on broadcast TV — somewhere between 10% and 15% of the total. I suspect, however, that those households are precisely the ones with the least amount of political clout. Which means that sooner or later, they’re going to lose their access to free-to-air broadcast TV. They won’t like it, but there’s pretty much nothing they can do to prevent it.

*Update: I’m informed that it’s actually illegal to sell a TV which can’t receive over-the-air broadcast signals. That said, it’s legal to sell a “monitor” which only has HDMI inputs, and which is designed to be used mainly as a TV.

The Boston bombing and subsequent manhunt was in many ways the first big interactive news story. It wasn’t the first big event to be covered obsessively on social media, but it was the first big event where millions of people became part of the story themselves. Some did so through choice, combing through photographs on Reddit or 4chan; others simply happened to be in Boston and saw their public lives, as broadcast to the world on social media, become part of the story just by dint of where they were.

The result was a veritable deluge of streams, in a world where the news has become a hard-to-navigate rapids at the best of times. For anybody who wanted to stay on top of what was happening without drowning in noise, experience and level-headedness were invaluable, and were displayed most prominently by Pete Williams of NBC.

But while Williams was the most visible of the people who got it right, there were many others, mostly unsung, working at places like the NYT. The paper’s public editor, Margaret Sullivan, rightly praises its editors for “staying away from unconfirmed reports” and treating with suspicion anything coming from “unnamed law enforcement sources”.

In the tradition of journalistic oxpeckers everywhere, Sullivan concludes that the NYT’s “reporting from Boston all week was fast, deep and accurate”. Which is the truth, but it’s not the whole truth: I’m quite sure that a very large part of the credit should go to the editors in New York, rather than the reporters in Boston. In a story this sprawling, no one reporter, and no one law-enforcement source, can possibly see anything approaching the big picture; it falls to the editors to take the various streams of information coming into the newsroom, many of which outright contradicted each other, and to weave them into a coherent and accurate whole.

Anybody who was on Twitter over the past week knows how hard that job was. It’s an exercise in massively multivariate real-time Bayesian analysis: as the news streams in from multiple sources over the course of the day and night, every new piece of information has to be analyzed in light of everything else that’s already known, or thought to be known. A clear on-the-record statement from the governor can be assumed to be perfectly reliable, but just about nothing else can be — not even the reporting of your own employees, who can easily make good-faith errors during such an extended and chaotic story. Sleep deprivation alone can account for that.

An experienced editor will use her hard-earned judgment to weigh the relative reliability of all the different sources of information. Some people added enormous value on Twitter — Seth Mnookin, for instance, had fewer than 7,000 followers on Sunday, and more than 40,000 by the end of the week, for good reason. Others, like Williams, proved their reliability on television, even as their rivals at other channels were reporting things which turned out to be false.

There’s an art to working out where to find fast and reliable information, and to judging new information in light of old information, and to judging old information in light of new information. And there’s an art to synthesizing everything you know, from hundreds of different sources, into a single coherent narrative. It’s not easy, it’s not a skill that most people have, and it’s precisely where news organizations add value.

But in this particular case, as Noah Brier points out in a post headlined “Being Part of the Story”, it’s something that millions of people ended up attempting to do, on the fly, anyway:

Everyone wanted to be involved in “the hunt,” whether it was on Twitter and Google for information about the suspected bomber, on the TV where reporters were literally chasing these guys around, or the police who were battling these two young men on a suburban street. Watching the new tweets pop up I got a sense that the content didn’t matter as much as the feeling of being involved, the thrill of the hunt if you will. As Wasik notes, we’ve entered an age where how things spread through culture is more interesting than the content itself.

The crowdsourced hunt was, in the end, unambiguously counterproductive: it hurt much more than it helped. But it wasn’t just Redditors and hive minds which got caught up in this particular mindset. If you look at the missteps of outlets like the New York Post and CNN, it’s easy to see them in this light — breathlessly passing on every new tidbit of information, rather than taking their function as editors and filters as seriously as they should have done.

Which brings me to an important and quite wrong essay from Ben Smith, the editor of BuzzFeed, and his colleague John Herrman.

Under the old rules, a responsible citizen passed any potential bit of news he could find on to the professionals. The professionals collected tips, corroborated them, published the ones that panned out. Reporters could protect their readers from bad information — indeed, for reporters, the story was defined largely by what was kept from the public…

Now we should assume our readers and viewers see virtually everything that we see. We can no longer decide which rumors and scraps of information should be dignified with publication — a sufficiently compelling scrap of information, be it a picture of a man with a black backpack or an anonymous, single-sentence Reddit post from the scene of the crime, will become news on that merit alone…

The media’s new and unfamiliar job is to provide a framework for understanding the wild, unvetted, and incredibly intoxicating information that its audience will inevitably see — not to ignore it. A Reddit post seen by millions without context is worse for the story, and the public, and to the mission of reporting than the same post in a helpful and informed context seen by many more. Reporting is no longer a question of whether or not to dignify new and questionable information with attention — it’s about predicting which of it will influence the story, and explaining, debunking, or contextualizing it the best we can. That is, incidentally, what our readers want.

It’s possible that Smith and Herrman are right that their readers are clamoring for BuzzFeed to explain, debunk, and contextualize the constant stream of noise and misinformation coming from Reddit and Twitter. But I suspect that if there is such a clamor, it’s coming from a vocal minority. For one thing, only a minority of BuzzFeed’s visitors come for hard news at all. And of those who do, only a minority of them care very much what BuzzFeed’s interpretation is of the material they’re reading on Reddit and Twitter. Finally, by their nature, Reddit and Twitter are going to be presenting a different narrative to each of their millions of users: what BuzzFeed’s editors are seeing on those platforms is not going to be the same thing that BuzzFeed’s readers are seeing.

It’s undoubtedly true that in the age of social media, it’s become very easy for anybody to peer behind the news curtain and see the chaotic raw material from which it is produced. But that in no way weakens the onus on responsible and experienced news editors to filter that material and form it into a fast, deep and accurate report. Indeed, the value added by those editors has never been more obvious than it is in situations like this one.

Smith and Herrman are absolutely wrong that a compelling yet false factoid, being shared willy-nilly across various social-media platforms, “will become news on that merit alone”. News is something true and important and relevant; it is not, and should never be, misinformation. Neither is it “whatever our readers happen to be finding on the internet”. Smith and Herrman are essentially taking a hugely important story, here, and reducing it to the status of covering a viral meme: the Gangnamization of terror. I have no problem with news stories covering viral sensations, but they’re what you do after you cover the important stuff. They’re not the important stuff themselves.

Which is not to say that BuzzFeed did a bad job last week. Debunking corrosive memes is a genuine public service, and it’s great that outlets like BuzzFeed and Gawker are doing it. Where I part with Smith and Herrman, however, is in their implication that everybody else — the NYT, the WSJ, the Boston Globe, Reuters, Bloomberg, CNN — should be doing it as well. That’s silly, and I can’t believe that many people would want to live in a world where a relatively small number of Redditors could effectively set the news agenda for the entire country.

On Monday, I received an email from someone calling herself Sarah Hanson, in which she claimed that she had successfully auctioned off 10% of her post-tax future income for the next ten years, raising $125,000 in the process. I wasn’t the only journalist to hear from Hanson: she had already, at that point, managed to score an interview with VentureBeat, which in turn begat lots of other coverage around the internet. But various aspects of the story didn’t smell right, to me, so I sent an email to VentureBeat, asking if they were sure this girl was for real. It turns out that she almost certainly isn’t. I was perfectly happy for VentureBeat to write the debunking; in fact, that was the perfect place for it to happen: there was very little point in me writing a story saying “some person you probably haven’t heard of is very unlikely to actually exist”.

Given the amount of information pouring onto the internet every minute, it’s statistically inevitable that a substantial amount of that information is going to be erroneous — especially when the source is something as unedited as Reddit or Twitter. No mainstream journalism outlet should allow its coverage of a major story to be hijacked by backchannel noise — especially when a large part of the value such outlets provide is that they filter out the noise and transmit only a reliable signal. Just because your readers can peer behind the curtain, doesn’t mean you have any responsibility to yank it open yourself.

When Matter launched, last year, co-founder Bobbie Johnson told Christopher Mims that “done right, we can offer something valuable and remain sustainable in the medium term.” Little did he know how close he was to hitting the nail on the head: it turns out that it’s not “the medium term” which is going to provide Matter’s sustainability, so much as a term sheet from Medium.

This morning, Matter announced that it’s being acquired by Medium, the Ev Williams company which makes it incredibly easy to put up really good-looking articles online. (That’s where my own most recent longform article appeared; it recently generated its 150,000th pageview.) This is fantastic news for both companies.

Medium, which already has first-rate editorial talent in the form of Evan Hansen and Kate Lee, now owns a company which produces fantastic articles on a regular basis, and which makes those articles look great. With the Matter team in-house, Medium will learn a lot about exactly what writers and editors want and need. What’s more, it now will have real revenues — and the ability to play around with different ways of generating those revenues. From the Matter FAQ:

We still think that selling individual articles and subscriptions is a great way to fund long-form journalism, and we’ve got no immediate plans to change that model. But we also want MATTER to evolve. Experimenting with tweaks to the model and the way we distribute our content will be a vital way of making MATTER robust in the long term. Joining Medium means we can get the help we need to run those experiments.

I’ve been saying for a while that Matter should look for ways that readers can pay them after they read a story, rather than before. It’s the distinction between forcing people to pay and letting people pay; my feeling is that it would massively increase the number of people interacting with Matter’s content, while also — quite possibly — increasing its revenues as well. (Especially since some small number of people will give significantly more than the 99 cents they’re limited to at the moment.)

I hope that Matter doesn’t feel too constrained by obligations to its existing subscribers: I’m quite sure that very few of them would mind very much if Matter allowed a bunch of its stories to come out from behind the existing paywall. But beyond its new-found ability to experiment with its business model, Matter more importantly now has access to some of the best designers and technologists in Silicon Valley, and as a result should be able to produce ever more gorgeous, immersive, and interactive journalism.

Both Matter and Medium have created spaces where longform articles can breathe, free of banner-ad intrusions. Both of them are looking forward to building a sustainable business around such articles. Now that they’ve merged, I’m optimistic that they’ll continue to innovate and help build the future of how people — amateur writers and professional journalists both — are going to want to express themselves online.

Jay Rosen asks, reasonably, that people start drawing useful distinctions between buzzy terms like content marketing, sponsored content, native advertising, and even brand journalism. Here’s my stab at it:

None of these distinctions is hard and fast, of course, but at least it’s a start; basically, it all comes down to who writes the content in question.

Was the material written by a professional journalist, writing a piece for an editorial outlet? In that case, any advertising message embedded within it falls pretty squarely into the realm of public relations. But what happens when the publication in question syndicates that content for use on some brand’s website? In that event, it becomes content marketing: independently-produced material, repurposed by the brand in question.

On the other hand, was the material commissioned by the brand itself, rather than any editor? In that case, it’s sponsored content. It might be written by a group on the ad-sales side of the publisher; such groups have existed for as long as there have been advertorials. Or it might be written by some group within the brand’s ad agency. The distinction between sponsored content and native advertising is a bit squishy, but it you do need to make a distinction, then I’d say that sponsored content is material designed simply to convey information to the readership of the publication in question, while native content tends to aspire more to going viral, and being actively shared by that readership.

Of course, if a brand takes that sponsored content and simply puts it up on its own website, then it’s just marketing. But it doesn’t necessarily make sense to think of all brand-produced content, on brand-produced sites, as marketing. Look at Sun, for instance, which as far back as 2006 was encouraging all of its employees, up to and including the CEO, to blog. We’ve moved on from there: instead of blogging at their own websites, executives more commonly express themselves on Twitter or Facebook or LinkedIn. But however it’s done, if you’re not paying a publisher for the privilege of expressing your own opinion, I’ve put you in the bottom right-hand corner under the broad rubric of “blogging”.

Which leaves the top-right corner, probably the fuzziest part of the matrix. Sometimes, sponsored content is written by real executives, rather than by people in the marketing or PR departments, and when that happens it feels a little bloggier. And at other times, of course, executives manage to get op-eds published and don’t need to buy any kind of sponsorship product at all. If “brand journalism” means anything, it’s probably this: brand executives doing something which feels a lot like opinion journalism, whether they’re paying for the privilege or not.

And really, trying to draw these distinctions is always going to be a bit silly and futile. Ultimately, they’re all different flavors of the same thing: attempts by companies to get consumers to read things which the company in question, or its executives, wants those consumers to read. There are lots of different ways of trying to skin that particular cat, and none of them is easy. In fact, trying to get consumers to read anything at all, in a world where those consumers are faced with almost infinite choices, has never been harder.

That’s why Twitter and Facebook have multi-billion-dollar valuations: they’ve created an experience where people consume a stream of content, rather than looking for something in particular. And it’s much easier to drop your message into a stream which people are reading anyway than it is to try to persuade those people to stop what they’re doing elsewhere and read your message instead.

The term mainstream media, then, if it wants to compete in a world where ads are going increasingly native, is going to have to become the mainly streams media. (Sorry. I couldn’t resist.) Look at the best native ads out there, from 30-second TV narratives to long-copy print ads: they are native to the form, yes, but they also live within linearly-consumed streams. We sit back and watch the TV, watching ads along the way; we leaf through the newspaper from front to back, grazing as we go, and reading whatever catches our eye.

If native is really going to take off online, it’s going to do so by appearing in the streams we want to immerse ourselves into anyway, whether they’re the Facebook News Feed or the magazine-style format of Flipboard. It might even work for individual publishers, if they can build attractive enough streams of their own. But I’m not holding my breath.

Andrew Rice delivers 6,000 words on BuzzFeed in the latest NY Mag, which means he has the space to tell a number of different stories. The one I’m interested in is the way that BuzzFeed CEO Jonah Peretti wants native advertising to disrupt banner advertising. I apologize for the long blockquote, but it’s a lot shorter than the article:

Peretti has talked of building “the agency of the future for a social world.” …

Watts and Peretti first set forth their theory in a co-authored 2007 Harvard Business Review article, “Viral Marketing for the Real World,” partly basing it on data from an experimental ad campaign at the Huffington Post. Watts has since continued to refine his research. His standard is that for every ten views an advertiser pays for when it buys a viral ad, it should get two shares. (“There is no free lunch,” Watts likes to say, “but maybe you can have a cheap snack.”) Peretti is convinced he can engineer a higher reproduction rate. “You can make money with that,” Watts says. “If they are predicting 20 percent of the variance and the competition is predicting 10 percent of the variance, they’re kicking ass.”

Peretti’s formula for virality really adds up to a more mundane sales pitch: Buy lots of ad impressions and realize a modest, if unpredictable, viral bonus…

BuzzFeed has released some selective data about the fractional proportion of sharing it achieves—its so-called “lift”—and claims that for the median advertising post, ten paid views yield around three shares. Peretti adds that the brands that have embraced the format most enthusiastically have better results. Virgin Mobile’s ratio of shares to paid views is better than one to one…

Virgin Mobile’s posts received around 1.1 million views for the last week in March. Other campaigns running on the site during that period, however, showed smaller results: Geico, 140,000 views; GE, 65,000 views; Pepsi Next, 44,000 views. These numbers don’t quite match the hype around native advertising, which might be why ad agencies sound much less enthusiastic about the medium’s transformative potential than publishers do.

Peretti complains about “obstructionist agencies,” and when he looks at advertising—with its four dominant holding companies, rococo bureaucracies, and reliance on a lucrative television medium now threatened by ad-skipping technologies—he sees an industry ripe for disruption.

I think that Rice is missing a couple of very important points here. For one thing, he’s wrong that that native advertising is fundamentally “mundane”, and provides just a “modest” uplift to whatever you can achieve through more traditional channels. Native pageviews might hard to come by — but any smart brand would absolutely prefer a single native pageview to a dozen banner-ad impressions. The difference between the two isn’t something marginal, on the order of 20% or 30%: it’s huge — a good order of magnitude, at least.

That’s because a native ad is something that consumers read, interact with, even share — it fills up their attention space, for a certain period of time, in a way that banner ads never do. Rice does mention that the advertising industry is dominated by the television-ad market, but he doesn’t seem to understand why. Yes, TV ads have the kind of reach that no other medium can match. But they also have duration, and a storytelling arc: if you’re not ignoring them, they command attention, in the way that, well, TV shows do.

In that sense, TV ads are truly native; the way you consume a TV ad is the same as the way you consume a TV show. Similarly, long copy print ads are native, for the same reason. And the ultimate native ads are the glossy fashion ads in Vogue: in most cases, they’re better than the editorial, and as a result, readers spend as much time with the ads — if not more — as they do with the edit.

On the web, by contrast, the vast majority of ads are not native. Instead, they’re intrusive, annoying, unpleasant, and — in most cases — completely ignored. We’ve now been consuming content on the internet for 15 years; we all know how to do it, and we know what we like, and publishers, including BuzzFeed, have become very good at delivering exactly what we want.

In stark contrast to the increasing sophistication of web publishing, however, the overwhelming majority of web advertising is still based on standard IAB ad units which were introduced in 1996 and haven’t changed much since. We’ve all learned how to tune such things out, either mentally or technologically, with ad-blocker software. Banner ads are never engrossing, they’re never shareable, and insofar as they attract your attention they do so in an evil way, by animating or blinking or otherwise distracting you from whatever it is you are trying to read.

When someone reads a BuzzFeed ad from Virgin Mobile or Geico or GE, they might “only” have a 20% or 30% chance of sharing it. But that’s not really the point. The point is that they read it, and they liked reading it. The “social uplift” is an indication that the ad is connecting with consumers — it’s like clickthrough rates, but real. Native advertising (as well as content marketing, insofar as there’s a distinction) is a way of communicating with web readers in a language they’re receptive to. And it turns out that when you do that, they actually listen.

In terms of disruptive force, then, native has a huge advantage over banners in that it is much more effective in connecting with consumers. And there’s another way that it’s disruptive, too: it utterly upends the standard ad-agency business model. This is the real reason that ad agencies are less than enthusiastic about native — they can’t make money at it. Banner ads are a lovely income stream for agencies, and ad-sales networks, and the whole crazy ecosystem of display-advertising companies. Every time there’s an impression, lots of intermediaries are sure to take their cut.

Native, by contrast, works on a very different model: you spend a certain amount of money putting it together, and then it lives online forever, generating marginal views at zero marginal cost. The agencies can still charge for their creative work, but they can’t charge for media buying any more — which is where the real money is.

As a result, most native campaigns tend to be worked out between publishers and brands directly, with ad agencies helping out but not driving the decision-making. It’s the beginning of the disintermediation of the agencies, and so it’s hardly surprising that they’re unenthusiastic about the trend. This is real digital disruption: native shops like BuzzFeed or Barbarian Group will never be as profitable as the huge ad agencies, but they can still cause those agencies to suffer very large drops in their digital revenues.

The big unanswered question, then, is not whether native has disruptive potential — it clearly does. Rather, it’s whether native will ever be able to truly scale. Native is growth-constrained on two fronts, and that means that if you’re betting on industry-changing disruption, you’re making a risky bet. The first constraint is creative. Native is hard work. Rice talks about how Virgin Mobile has to come up with “several posts a week” when its running a BuzzFeed campaign, and his article is illustrated with a photo of a “creative strategy meeting” where I count 19 people in frame, plus untold others out of it. The amount of human time and effort that goes into a native campaign is enormous, continuous, and it doesn’t decrease much once the campaign is up and running. You can’t just run the same banner a billion times: the marginal daily cost of native campaigns is vastly greater than the marginal daily cost of buying banners.

And then there’s the second constraint, which Rice mentions: all of that effort is going into reaching a relatively small number of people. This is another way in which native ads are like long copy print ads: they reach a small audience, rather than a mass audience. As a result, any brand wanting to reach a mass market is going to have to use native as just one part of a much bigger strategy, and that in turn is going to keep the native-averse ad agencies in the driver’s seat.

My guess is that BuzzFeed’s investors will do OK for themselves, in the end. But a healthy exit for BuzzFeed is not the same as a genuine disruption of the digital advertising space. Although native ads have the potential to be incredibly disruptive, I’m far from convinced that their larger potential is going to be realized any time soon.

The news of the day in the media world is that Jeff Bezos has led a $5 million Series E funding round for Business Insider. Here’s the story, according to CEO Henry Blodget:

Jeff’s investment grew out of a dinner he and I had about a year ago. We talked about the business, and he was excited about it. (He sees some parallels with Amazon). A few months later, he expressed an interest in investing. My reaction was basically “Hell, yeah!”

Blodget has now articulated a simple public goal: “to become the best digital business publication on the planet”. It’s a conscious echo of Bloomberg’s stated aim to be “the world’s most influential news organization”. If he needs to invest millions of dollars of other people’s money to get there, that’s fine.

Blodget goes on to say that he’s obsessed with his customers — both readers and advertisers — and that his customer focus is the main thing he shares with Bezos. (Well, that and his famous Amazon call, of course.) He also says that Bezos’s money “will allow us to continue to invest in our editorial, technology, and client teams” — which almost certainly means that there’s no chance, now, of Business Insider being profitable in 2013. Six years after it was launched, the site is still in growth mode.

And frankly, there are quite a lot of things that Blodget could use the money for, if he is really focused on the reader experience — indeed, there are so many things that he could probably spend all that money quite a few times over, if he wanted. The site could use a redesign, for starters, to make stories pop more for readers and to provide more attractive opportunities for advertisers. On top of that, the architecture of the site should reflect the way that stories are covered. Here’s how BI’s editorial chiefs see the way that they work:

“We don’t really think of things we put up as ‘an article,’” said Carlson. “It’s a bit of information conveyed to people. One of my old colleagues used to say that the last sentence of your last post is the first sentence of your next post. Because by the time you reach the end you sort of come to a cliff, ‘Oh I have another thought on this and I’m just going to put it in the next post.’ In a way, it does sort of become a narrative. For sure, I think [that's] the attraction of reading something at Business Insider … It’s a live medium where the narrative is always coming out with the next thing.”

Weisenthal is often reminded how differently digital outlets such as BI work when it comes time to submit content for awards.

“They have the journalism competitions where they invite people to apply and they always say, ‘Submit your top three posts for consideration that you’re most proud of’ or something like that,” he said. “And I can never come up with the stuff. I don’t think I have a single great post last year that I’m really proud of. Everything I write is part of this bigger stream.”

He pointed to his real-time blanket coverage of the monthly U.S. jobs report as an example. “If you follow me on Jobs Day, within like 20 minutes of the report coming out, I have a summary posted,” he said. “Then I have another post singling out one detail I thought was interesting. I have another post saying what it might mean for interest rates and fed policy. I have another post talking about the political dimensions and so forth. I’m proud of the fact that it’s this whole suite of stories.”

I’m an admirer of this form of journalism, and I think that many media organizations, including Reuters, are going to move in this direction. But right now, if you go to one of Joe’s payrolls posts, it’s not easy to find all the other ones — to have them all in one place, together giving the bigger picture. In order to be able to allow that, Blodget will need to make some serious technology investments.

What’s more, a re-engineered website might well result in a website with significantly fewer pageviews. If you can see all of Joe’s payrolls posts on one page, then that means fewer pageviews for BI than if you call up all ten of them individually. For most of its existence, BI has been in an uncomfortable race, trying to increase the number of pageviews it serves up faster than its CPMs are falling. Investors are generally OK with losses, which reportedly reached $3 million last year, only so long as revenues are growing. And they are growing: Blodget tells me they were more than $10 million in 2012, up from about $7.5 million in 2011 and $4.7 million in 2010.

The problem is that in the chase for revenue growth, Blodget is sacrificing a pleasant user experience. He installs ugly automatic links under certain phrases, for instance, which when you mouse over them start playing video ads. Or he sells a lot of interstitial ads which force you to click another time before reaching the story you want to read. Quartz points out that there’s a good chance Business Insider is worth less than the much younger BuzzFeed, where CEO Jonah Peretti is adamant that he’ll never run a BI-style slideshow, or even “crappy display ads”, just because readers clearly prefer everything on one page and don’t get value from those ads.

The problem is that if Blodget decides to pare back on artificial revenue juicers which readers dislike, that hurts revenue growth as well as profits — even as BI is saying that it intends to accelerate revenues this year to something in the $15 million range. In order to keep revenues growing even as he re-engineers his site to make it sleeker and less optimized towards pageview maximization, Blodget would have to invest not only in technology, but also in sales — paying big money for expensive staffers to build relationships with brands. BI gets too much of its revenue from banner ads right now: it needs to diversify its ad revenue, and start finding more ways for brands to reach BI’s coveted readership. One of those new channels is conference sponsorship, and I expect that BI will use a bunch of its new money to invest aggressively in conferences. But one of the big hidden costs behind building a new kind of website is the fact that you need to build a new kind of sales team, too, selling the kind of products which are often referred to as “native”, whatever that’s supposed to mean.

Business Insider has always been run on something of a shoestring; it made the entirely understandable decision, for instance, to hold onto a large chunk of the capital it raised in the past, rather than blowing through it and then suddenly being forced to cut back for the sake of profitability. This new round allows BI to increase the amount it’s investing while still retaining a reassuring cushion. But $5 million is not remotely enough money to allow Blodget to pivot to a very different business model, even if he wanted to do so, which he probably doesn’t. For better or for worse, he’s stuck in a world of banner ads and CPMs, and although he’s done well in that world to date, the future of that world looks pretty bleak.

There are many sites, Gawker Media’s foremost among them, which have gone to great lengths to wean themselves off their addiction to banner ads. And in general it seems to me self-evident that “the best digital business publication on the planet” is not going to be one which aggressively chases pageviews and ad revenues at the expense of the user experience. By thinking of stories as streams, Joe Weisenthal found a great way of juicing pageviews, since every element of that stream, under the current architecture, is a new story and a new page. But he’s also stumbled upon a powerful and addictive new form of journalism, which is Blodget’s best hope for achieving his ambition. The question is: will Blodget be willing to give up his current business model, in order to let Weisenthal follow his editorial vision to its logical conclusion?

Last week, I hypothesized that the publishing industry was going to informally settle on a single management-consultancy company to ask for paywall advice from. That consultancy, having seen everybody’s internal figures, could then tell everybody else what “industry best practice” was. It’s the time-honored management-consultancy m.o., reselling other clients’ confidential information, suitably anonymized, of course, so that everybody learns from everybody else’s successes and failures.

This is a winner-takes-all business: it works best if everybody hires the same consultancy. And now it’s pretty clear which consultancy is going to win: Mather Economics. They say they’ve worked for pretty much everybody, at some point, and that they directly manage some $2 billion of subscription revenues for their clients. And today, fresh off a $1.75 million funding round, the paywall provider MediaPass has announced that it’s going to bake all that Mather knowledge into its own product. Given all the data being generated and analyzed by Mather and MediaPass, it looks like they have a pretty unassailable position in this particular niche.*

So, what do Mather and MediaPass see as the future of paywalls? What is best practice in the industry? Interestingly, as Anthony Ha reports, they’re not particularly enamored with the meter system, despite its high-profile successes at the FT and the NYT.

Although MediaPass supports “metered” systems, [MediaPass president Matt] Mitchell says he sees more potential in creating a specific mix of free and paywalled content, although that mix will differ from site-to-site.

Publishers should think of their free readers as leads who might eventually become paying subscriptions, he says. For example, for a long time Mitchell read ESPN.com for free, but a year ago, he stumbled on a paywalled article that he really wanted to see, and since then he’s been a subscriber.

“What a meter does is give you 10 views free, and on the eleventh you’re asked to subscribe,” Mitchell says. “That’s rolling the dice and gambling that the article I see on the eleventh view is the one I’m willing to pay for.”

It’s worth noting, here, that even the FT and the NYT don’t have “pure” metered systems, where every pageview counts towards the meter. In the early days of paywalls, some content was free, while other content you needed to pay for; the meter, in theory, replaced that system with one where the determination as to whether an article was free or not was a function of how many other articles the reader had read, rather than being a function of the content of the article itself.

There’s always a trade-off, however, and there are certain areas of the FT and NYT websites which are always free and don’t count towards the meter. Finance, interestingly, is one: you can read as much Dealbook and Alphaville as you like without a subscription. And Mather’s Matt Lindsay said that the NYT quietly does the same thing for its entertainment section, during peak season in the fall: there’s a huge amount of advertising demand, and it doesn’t want to put any obstacles in the way of tourists looking to the NYT to work out what shows they want to see.

Talking to Mather and MediaPass, it’s clear that their idea of “best practice” doesn’t rely much on meters at all. They have the numbers, remember: they know what kind of walls are best at maximizing revenues, and what kind of walls just end up turning readers away. And crucially, one of the biggest lessons they’ve learned is that it’s a mistake — at least from a purely financial perspective — to treat all readers equally. Some readers have a much greater propensity to pay than others; ideally, you want to extract a lot of money from those readers, while also allowing the vast majority of your visitors — the ones who will never pay you anything — to still consume your content and view the associated ads.

For instance, it’s often easier to persuade people to subscribe to sports content than to entertainment content, even as it’s easier to sell ads against entertainment content than it is against sports content. So it does make sense to keep entertainment free, and put some kind of paywall around sports.

And although readers hate the kind of extreme opacity practiced by the FT, where there’s basically no rack rate and nobody knows what anybody else is paying, from a revenue prospective it makes a lot of sense. The FT knows quite a lot about its registered readers, so it can be quite effective at charging the highest prices to people with the greatest willingness to pay, while charging much lower rates to readers in, say, India.

That kind of thing can be dangerous, from a PR perspective. Amazon, for instance, got into trouble when it was caught selling the same products at different prices to different customers. But there are other ways of achieving much the same end: you can set a relatively high official price, for instance, and then start showing various special offers to people whom you think might be willing to subscribe if you offer them a discount. No one really minds that.

And certainly it seems to be a good idea to offer a range of subscription lengths, priced so that there’s a strong incentive to go for the longer-dated annual subscription, even if again that means a substantially lower rate on a per-month basis.

I’s not all that hard to tell who’s likely to be willing to subscribe, and who isn’t. Print subscribers, for instance, are much more likely to be willing to pay for a digital subscription than a reader who doesn’t already pay for the print version. And people who visit frequently, and who read a lot of local news, or sports news, are also more likely to subscribe.

In general, the trick is to get as many subscribers as you can — because once a person subscribes, they generally turn out to be surprisingly loyal and price-inelastic. You can keep on charging their credit card, even at steadily-rising rates, and they’re not going to unsubscribe. And then, for the 90% of readers who don’t subscribe, it’s a good idea to find content for them, too. The paywall shouldn’t just be a “pay here or get nothing” option: the “no thanks” button should take you to valuable free content.

That’s why, as NYT spokeswoman Eileen Murphy confirmed to me, the NYT is looking at rolling out a new digital subscription product, priced below the current cheapest option of $3.75 per week. Most NYT readers are understandably reluctant to spend $195 a year on access to a single site, so the NYT might well offer something cheaper, without the full unlimited range of content that subscribers get with the current digital package.

What’s impossible to calculate, of course, is the long-term opportunity cost of driving away people who want to read your content but aren’t willing to pay. MediaPass’s Mitchell told me that in most cases, the act of putting up a paywall is the act of “essentially harvesting revenue from a loyal long-term audience” — people who have been reading the publication for years, and have turned it into a habit they don’t want to give up. That’s fine, as a short-term means of maximizing revenues. But it’s dangerous in terms of getting new loyal readers. Which is one reason why online media startups almost never have paywalls: they want as many people as possible to discover them.

My expectation, then, is that newspaper paywalls will become both increasingly sophisticated and increasingly expensive over time — but that paywalls are not going to migrate very quickly out of the newspaper world and onto the rest of the internet. In a dying industry, the sensible thing to do is to maximize your revenues before you die. Paywalls might well make money for newspapers. But that doesn’t mean that newspapers aren’t dying. Quite the opposite.

*Update: So this is embarrassing. The public press release notwithstanding, it seems that Mather got cold feet about the deal with MediaPass, and is not going to go ahead with it after all. I think Mather still has its longstanding relationship with Press+, the newspaper paywall company, but I’ll look into it and find out.

Update 2: This seems at heart to be a spat between Press+ and MediaPass, with Mather being enjoined from working with both.

It’s paywall season right now: the Washington Post, the San Francisco Chronicle, the Telegraph, the Sun — all have recently announced plans to erect paywalls in an attempt to extract subscription revenues from their most loyal online readers. And other paywalls are being tweaked: the NYT paywall is getting less porous, while Andrew Sullivan’s is being tightened up, with a new $2/month option to complement the existing $20/year price point.

The trend here is clear. There is now only one major US newspaper without a paywall of some description, although others have free spin-off sites, like Boston.com or SFGate.com, which act a bit like the outside-the-paywall content on other sites.

There are three big drivers of these decisions. The first is that there’s no hope that online ad revenues will ever grow to replace print ad revenues. They’re barely growing any more, even as they’re still only a small fraction of total ad revenues. The second is that for various reasons, newspapers need to “cling to the mantle of quality at near insane costs”, as Sarah Lacy puts it. If costs are stubbornly high while revenues are shrinking, then the only possible solution is to try to raise new revenues by any means necessary — or go bust.

Finally, there’s the behavioral aspect: newspapers in general, and the NYT in particular, are quite deliberately habituating readers to the idea of paying for content. This was an obvious strategy even before most of the paywalls launched, back in 2010: first get people used to the idea of paying at all, and then, slowly, raise the amount that you ask them to pay over time.

There are an infinite number of points on the spectrum between tip jar and paywall, but there does seem to be a clear move to the right over time, towards less porous and more expensive paywalls. Some paywalls, like the FT’s, are what you might call Metropolitan Museum paywalls, porous in name only. While in theory the FT works on a meter system, giving people a certain number of free articles before asking them to pay, in practice if you want to read an FT article you’re going to be asked to pay — even, annoyingly, if you’re already a subscriber. (I would dearly love a subscription which authenticates based on device rather than on an easy-to-forget and hard-to-enter username/password combo: can’t the FT just see that it’s my phone accessing the site, and let me read anything I want if I’m a subscriber?)

And in general, the more you’re asking for, the more coercive you need to be. At a buck or two a month, loyal readers are happy to support you. At $15 or $20 per month, you need to break out the sticks as well as the carrots.

One of the problems with paywalls is that everybody wants their paywall to be simple and transparent and easy for everybody to understand. But if you do that, you can’t A/B test; you can’t work out empirically what the optimum price is or what the best place to set the meter is. Which is where the raft of different paywalls out there comes in handy.

Here’s my prediction: At some point, the industry is going to informally settle on a single management-consultancy company to ask for paywall advice from. Everybody’s going to use the same company, with the result that the consultancy in question is going to see real internal figures from lots of different newspaper publishers, with lots of different models. The consultancy will then — for a price — tell its clients what “best practice” is in the industry, which is code for “this is the way that the most successful newspapers are doing it”. No one site can easily do A/B testing on its own. But put them all together in the head of a well-connected management consultant, and it becomes much easier to see what’s working and what isn’t.

But all of the paywalls and consultants in the world won’t change the fact that the amount of information freely available on the internet continues to grow very fast, and that the number of people willing to pay for any kind of news online is always going to be a small fraction of the total online news-reading population. As Lacy says, there’s an exciting future for online news — even if the prospects for legacy-burdened newspapers are dim. The paywalls might help with newspapers’ finances. But they’re certainly not going to help make them any more relevant.