Rupert Murdoch's recent suggestion that News Corp might stop accepting links from Google has stirred things up. Cory Doctorow thinks Murdoch is bluffing. But as Silicon Alley Insider noted, this may not be such a big gambit for News Corp. According to SAI's estimates, the Wall Street Journal "depends on Google for about 10-15% of [its] site's revenue. This is a meaningful amount, but it's not an amount that couldn't be foregone for a while if Murdoch wanted to try boycotting Google."

This is plausible. A typical newspaper visitor who arrives via Google reads only a fraction of the pages that a loyal reader does. The WSJ also has significant subscription revenues, which would not be affected by a blockage. And for local papers, search-engine visitors are far likelier to be out-of-market, which means that the publisher gets less advertising money per page-view than from loyal, local readers.

So maybe Google's traffic is not worth all that much. But then that's not the point. Even 10% of a site's revenues is an important amount, and you don't throw that away without some thought. If Murdoch were to carry out his threat, few of Google's users would notice News Corp's absence in search results, and even fewer would be bothered enough to go to the missing sites directly. As I've argued in a recent post, this means that for every dollar that Google looses, News Corp would lose many more. And as a result, News Corp has little bargaining power here. In fact, a negotiation might backfire. If a recent report is to be believed, other content providers are actually paying aggregators to link to them.

But News Corp is a sophisticated operation, and you can assume they have gone over this thinking (and the same applies to Google). So what are they doing? It's easy to speculate – e.g. it could be a bid to get other publishers to follow – but I can't think of anything convincing. I'm puzzled.

From the Telegraph:

This argument, that it costs money to produce something, therefore it has value, was put forward by Karl Marx in Capital. The exchange value of goods is equivalent to the amount of labour required to turn base materials into products. [...] As a thoroughgoing Marxist, Murdoch believes that [his content] must now have financial value, calculated as the labour expended plus his surplus.

In an earlier post I argued that in the relationship between online aggregators and publishers, aggregators have the upper hand strategically. In a nutshell, the reason for this was that if an aggregator were to stop linking to a publisher, traffic to its site would drop only by a fraction of the corresponding drop in the publisher's traffic – if it drops at all. As I said then, this means that aggregators have all the bargaining power, and publishers have no hope of extracting payments from them (barring legal options, which may or may not exist).

As a corollary, I could have added that publishers should not push their luck trying to get payments from aggregators. Taking a leaf from the Godfather, aggregators could reply by not only refusing, but demanding that publishers pay them for their links instead.

Well, today I read this fascinating interview of Andrew Breitbart, second-in-command at the Drudge Report and the man behind Breitbart.com. According to the interviewer Greg Beato, Brietbart gets Reuters to pay him for linking to Reuters stories from Breitbart.com, and possibly from the Drudge Report too. The key bit (my emphasis):

In the early days of Breitbart.com, Breitbart licensed content from the Associated Press and Reuters, as this archived page shows. But according to documents generated in a 2005 legal dispute between Breitbart and two other parties, Reuters terminated its contract with Breitbart.com in late September.

In October 2005, however, Reuters approached Breitbart with the kind of offer that generally occurs only in the less believable tales in Penthouse Forum or when a Nigerian vicar is planning to rip you off. To wit, Reuters wanted to pay Breitbart "a fee for traffic to driven to Reuters [sic] own website."

Typically, newspaper sites pay newswires to license their content, and that's what Breitbart was doing until Reuters cancelled its original contract with him. Now, it wanted to switch things up.

Compare and contrast this with the ongoing saga in the UK around aggregator newsnow.co.uk which is being sued by several papers for linking to their websites. Discuss.

In an article published today, the Guardian wonders if Rupert Murdoch's announcement (a few months back) that News Corp will introduce pay-walls across its online portfolio amounts to collusion. The key bit:

When Murdoch announced that he intended to introduce charges for access to all his news websites, he said that he believed other publishers would follow suit. [Alan Davis of legal firm Pinsent Masons] said that a pattern of such statements, in effect a signal to rivals to do the same, can be interpreted as a "tacit cartel", even if no meeting or explicit deal has taken place.

I am not a lawyer, so I won't comment on the legality of this. But I'll make two related comments.

First, Murdoch's move can be seen as a textbook case of strategic commitment. Suppose that you and your competitors would all like to do something that nobody wants to do on his own. So you announce that you intend to make the risky move even if nobody follows you, and you do it in such a way that your announcement is public, credible and irreversible (you are 'committed'). If you succeed in being credible, now a competitor – without ever talking to you – would feel reassured that if it also makes its move, it won't stand to lose as much as it might if it had moved on its own. If you are big enough, the change in your competitor's strategic calculation may be big enough to compel it to move as well. And since you know this from the beginning, your initial move was not as rash as it seemed. Granted, Murdoch's move was hardly irreversible – all he did was make an announcement in an earnings call. But he would lose a lot of credibility if he were to back down, and in this sense he is committed. Presumably this is what the Guardian has in mind. But note that – as I've hypothesised it – this move doesn't involve any talking behind closed doors.

Second, all of this may be a moot point. Talk of anti-competitive coordination presupposes that erecting a pay-wall is a profitable move only if your competitors do the same. But that is far from clear. To be sure, if you just slam a big all-or-nothing wall that refuses to serve any free content to anyone then you might lose most of your readers to your competitors and end up worse-off – unless your competitors erect walls too, and hence the coordination point.

But such 'hard' walls are out of fashion. What is in fashion is 'smart' pay-walls that only affect a small percentage of your readers (see my previous post on this). The thinking is that – if done right – this should result in a net increase in profits regardless of what your competitors do, essentially because you are only charging those readers who are loyal enough not to go to the competition.

This is not to say that with 'smart' pay-walls competitors' actions are irrelevant, or that there would be no point in price-fixing. If your 'smart' wall can make you some money even if your competitors follow suit, it can make you even more if they do (because you could then cast your net wider and charge more people, or charge a higher price). But that's another discussion – it's about refinements to the walls and not about whether to erect them in the first place.

Much of what you read about charging for online content is ill informed. All too often the question is put as if the only choices were to charge everybody or nobody at all, for all content or for none of it. And although some coverage is more nuanced, it rarely sheds any light on how the people who are working on this problem are going about it. If you are trying to get to grips with the real issues, this post might help.

A recent survey claimed that somewhere between 5% and 10% of an average paper's online audience may be willing to pay for the right to access the newspaper's site. This is not an insignificant amount of people – for many papers, 10% of monthly online reach is comparable to the entire print circulation. So clearly there is some money to be made. The question is how.

Dumb models need not apply. If you just drop a big pay wall and force everybody to pay you will lose up to 95% of your audience, and a similar (although somewhat smaller) percentage of your ad revenues. For most papers the new subscription revenue would not be enough to make up for this. So, clearly, a big pay-wall that charges everybody for everything is not a good idea. Everybody knows this, and almost nobody is considering this approach.

A better idea is to come up with a clever pay-wall that only charges those readers who are willing to pay and leaves the rest undisturbed. If you can pull this off, you have a win-win: you keep all your readers, page-views and ad revenues, and on top of that you get 5%-10% of your readers to pay.

The trick is how to do this, and in this there are two difficult challenges: targeting the right people, and deciding what to charge for (and how much).

About targeting: Try to charge too many people and you lose readers and ad revenue. And if the people you lose are among your frequent readers, you could lose quite a bit of revenue: each of them may account for ten times the number of page-views that other readers do (or more). So, play it safe: err on the side of caution and show the pay-wall only to a very small group that you are confident will pay. You may not get it quite right, but that's OK. You may end up not charging some people who might have paid, but at least you keep their ad revenues. And if you end up charging some people who are not willing to pay, at least you were not charging that many people to begin with so the loss of ad revenue is small.

About what to charge for: the typical way of dealing with this is to identify (or develop) some content within your site that is particularly valued by your targeted segment, call it "premium" and put it behind a pay-wall – while keeping the rest of your site free. The problem with this is that your targeted readers will only pay according to how much they value the premium bits, not according to how much they value your entire site – and this may not amount to much. A different approach is the one used by the Financial Times, which charges anyone who reads more than ten articles per month. The idea there is that people who value the FT enough are also frequent readers of it; the downside is that not all (potential) frequent readers are willing to pay, and you lose the ad revenues you could get from them. There are many other ways to target people who value your site. For example, you could charge people who visit before 9am, or who access it via their iPhones.

And then there's the PR. You don't want your pricing policy to come across as too arbitrary. You can't just say "we think you fall in a segment that values our content and therefore you should pay us $50 per annum". You need to be upfront about which ways of using your site (e.g. frequency, device, etc) are free and which ones are not, so nobody feels singled out.

All of this calls for a lot of research, analytics and technology. The devil is in the details and there are many of them. But it can be done. And as you can see, the idea is simple.

Online publishers have a problem. They can see that aggregators are capturing value from just linking to publishers' content, and want a piece of the action. Some have resorted to accusations and some to legal threats. But if aggregation is so hot, why don't publishers just get in on the game and start linking to all and sundry?

If publishers simply started adding links to other people's content on their front pages this could lead to a decrease in click-throughs to their own content – and, as a result, a decrease in profits too. And while it might seem this could be avoided by publishing less content – "cover what you do best and link to the rest", as Jeff Jarvis recommends – this is easier said than done.

There are two reasons for this. First, for many publishers content costs can only be cut by firing journalists – a slow and painful process. Second, online publishers who also have a print operation (i.e., all the mainstream publishers) have no option but to pay for all the content they publish in print. And once this cost is incurred, the extra cost of making content available online is negligible. So print publishers are stuck with hosting their own expensive content online, and then they have no choice but to try to monetize it by driving traffic to it. Because of this, they can't follow Jarvis' dictum.

Aaron Mishkin has recently proposed a new alternative: stop worrying about the web, and embrace other internet platforms where the dreaded separation between aggregation and content – i.e. the hyperlink– doesn't exist and each publisher can have its own self-contained environment. E-readers like the Kindle and app stores like Apple's could thus be media's saviours. There may or may not be money in this, but betting your company against the web seems an awfully risky strategy.

The strategic picture seems clear. In the web there are few if any synergies from simultaneously owning content and aggregation assets, while in print the synergies are at the core of business models. The billion-dollar question is how to live with both models at the same time.

(This post's title is a reference to a similar problem studied in detail by Clayton Christensen in his book "The Innovator's Dilemma". There as here, incumbent firms could see that new, low-cost competitors using new business models were eating their lunch, and yet were unable to react without killing their profits. More on that soon.)

If it makes it into policy, today's FCC proposal on net neutrality will be remembered by historians as a defining event for 21st-century media - for better or worse. The fact that the subject is so obscure, even among media people, is food for thought. I'll write more about this as time allows; in the meantime you may want to read my primer to the subject.

In a good post on the challenges facing online newspapers, Judy Sims notes:

News execs still aren't looking to understand their users or to seek gaps in the marketplace, but instead they are trying to do the same old thing only now get paid for what has become free. It's all about existing products (i.e. news and features), nothing new. What about data? What about helping users to make or save time or money? What about niche content that is rare or specialized enough to command payment? [...]

There are 4 P's of marketing. Product, Promotion, Place and Price. [Newspaper execs] just want to talk about price. That's because they're not thinking of their websites as products.

I share her diagnosis but not her prescription. Yes, it is undoubtedly true that newspaper people are too fixed on their content and need to think more broadly about their product and their customers. I wrote about this at length in an earlier post.

But a (much needed) renewed focus on customers need not mean a push into product development. It should start, first and foremost, with an effort to understand your current product – the website you already have and the people who use it.

There is an old school of thought in industrial design that says that you should design products "phenomenologically" – that is, by "bracketing" the thing itself out of your mind and focusing on what people do around it. Only then you can fill in the blanks and design the product itself. Or, to put it more succinctly, the product is the activity. (This fascinating book by Brenda Laurel touches on some of this).

Now back to online newspapers. If your product is the activities that people perform around it, do you know what your product is? For papers, the answer is generally "no".

Sure, anyone can think of some typical activities – e.g. the loyal reader who visits your front page every day, skims the headlines and chooses three articles to read, or the clueless reader who arrives via Google by accident. But what about the reader who visits the Drudge Report every day and once in a while follows a link to one of your articles? Or a reader who checks her Twitter feed throughout the day and every now and then follows a link, when she can steal a minute? Or the busy employee who suddenly receives an email from his boss telling him to read a certain article?

These activities are all very different: competitors' websites are relevant for some but not for others; they address different needs (some are about distraction while others are about self-imposed routine); and for some branding is important for while for others it isn't. In other words, each has a different value for your readers.

If you think of these activities as your products, you can ask how much pricing power you have for each. And if you segment your audience in terms of the activities they perform, you can begin to get an idea of how much money you could make. Yes, most of these "products" will have to be free, but you may be able to charge for some of them.

Of course you could apply the same kind of thinking to print publications as well. You could commission ethnographers to study how people use your print paper. But at the end of the day, with print you have no way to enforce a premium price on key activities, because you have no control over what people do with your paper once they've bought it. So you are forced to set a uniform price for the thing itself – the content. And if you transport that thinking to the web you are forced to either charge all your readers for some or all of your content – a risky proposition.

But the web is different. Here you know, in real time, exactly how your customers are using your content. And at any time you can step in and say "excuse me, but for that kind of activity you have to pay".

This is the inspiration for much of what we are doing at The Publishing Project. Of course the devil is in the detail.