Behavioural economics of paywalls #paywalls11

I couldn't make it to Paywall Strategies 2011 today, but it's a subject that's very interesting to me so I thought it was a good time to write about it. The whole question of paying for online news - and other information - combines two of my main research areas. First, it's all about pricing, and one of the most interesting and important new phenomena in pricing at that. And second, because the argument can only be resolved with a deep understanding of cognitive incentives (and disincentives).

In fact, the area of micropayments was one of the earliest case studies I considered when getting into the pricing world several years ago. It's one of the clearest departures from conventional supply and demand dynamics. Here's an example:

13% of people say they would be willing to pay for content online. And yet only 4% do. Why the gap? Yes, there's always a difference between what people say they'll do, and what they actually do. But from experience with other products we would expect this gap to be in the opposite direction, if publishers were using the right methods to win paying customers. With many products, people who say they would not pay, or would pay only a low price, can be persuaded to hand over more money if the structures are right. Why not with electronic content?

One of the conversations I've been watching on the #paywalls11 hashtag this morning has gone down the route of "What can we offer that consumers want to pay for?" This tweet is typical (and seems reasonable at first):
This is a trap. Of course there's some truth in it: quality will have an influence on the price point achieved. But in fact, it's not what you offer that determines whether people pay - but how you offer it.

Back to that demand curve. In economics, the cheaper a product gets, the more of it people are meant to buy. But with content, below a certain point, when price drops, quantity demanded does not rise. In fact, the actual price level is not really the force that the seller is working against. Instead, there are two distinct cognitive barriers to purchase, neither of which is directly related to the price level as such.

The first barrier is the inconvenience of making payment. It's a hassle to enter your credit card number, register an account with your email address, or whatever it is they make you do. In fact, people who are quite willing to enter their credit card details for a £1.50 purchase become unwilling to do so for a 5p payment. In conventional economics this makes no sense. But what if individuals are using the price of an article as a cue to its value? The cheaper the article gets, the less valuable it looks; and the less worthwhile it is to bother with entering their details. Consumers are quite logical, sometimes.

(A secondary consideration here is on the supply side - publishers may incur credit card processing costs which are greater than the revenue received from each transaction. But that kind of problem is easier to overcome, if a genuine revenue stream exists.)

The second barrier is an issue of principle: many people simply refuse to pay for online articles, having grown used to getting them for free. In this regard, the difference between free and 1p is far greater than the difference between 1p and 5p - or 50p. And this psychological and behavioural barrier is the most important thing for online publishers to overcome.

Many people are trying to lower the first of these barriers: a constant stream of new micropayments schemes have entered the market over the last fifteen years. Until recently I expected Facebook credits to have the best chance of success, but Google OnePass looks like it might also be a competitor. Each of these two has the benefit of universality: many people are logged into both sites all the time already, meaning there is no need to log in to yet another system in order to trigger a payment. I am sure that, sooner or later, this will stop being a problem.

The second is harder to bypass. Ingrained assumptions are notoriously hard to shift, especially when there is competition out there who are happy to pander to them. One way to do this is to reframe the product as something different. With the launch of new platforms - notably the iPhone and especially the iPad - publishers have had the opportunity to start afresh, with few existing expectations working against them. Since there is no history of free news content on iPad, there are lower barriers for many people to the principle that they might have to pay for it. The drawback of this approach is that by changing the nature and name of what you offer, you risk losing the momentum and loyalty of the 20-50 million unique users which several of the top news sites each have. The more seamless the migration from standard web use to any new access method, the more people will expect it to remain free.

But apart from "make an iPad app", what more can behavioural economics tell us about how to improve takeup and increase revenue from selling news?

I'm going to go a bit deeper to analyse this, to the level of decision theory. The goal is to understand the whole lifecycle of a person's decision to consume, and/or pay for, news and the influences on it. So consider the cognitive process of a person who is about to show up at a page on your online newspaper's website.

They come to this in one of three ways: through searching for something specific; browsing in an undirected way and happening upon it; or checking the page as a part of an ingrained newsreading habit. Each of these involves its own behaviours and motivations.

1. Directed Searchers

Those searching for something specific are the most purposeful and "rational" group - the most likely to be willing to accept a conventional offer to pay for something. They have a goal in mind, they are carrying out a specific strategy to achieve it, and a financial transaction - at least potentially - can form a legitimate part of that strategy. This model works well for many types of specialist content - clip art, legal documents, iTunes, or films. But the challenge with news is twofold: one, most people probably don't look for news in this way; two, there are lots of free competitors out there who will be found just as easily as your paid content.

To convert directed searchers, you have to make salient why your offer is different to all the others they are seeing next to it on Google. Here's the challenge. Almost any form of words you could use to do this; any format of news page you could design, or any keyword strategy, can be copied by the free people too. There's one exception. They will never copy the phrase "You must pay for this article".

My suggestion would be, in this case, to make a virtue of the paywall. Emphasise that this is a paid-for article - put it in the title tag so it shows up right in the search results - and is correspondingly higher quality than the free alternatives. Higher quality could be on a number of dimensions. It might mean "better checked". It might mean "more personalised". It might mean: news, analysis and opinion, balanced and all in one page. That hardly matters. The point is that price is a signal of quality, and at least some people are going to be willing to pay for it.

What NOT to do is get people into the page in the belief that it's free and THEN ask them to pay. If you let people know up front, conversion will be much higher. Yes, many people will be put off clicking on the link. But those people are unlikely to pay anyway. The people who might pay, will be _more_ likely to click; and (subject to testing, as always) I'd predict this approach would significantly increase revenue.

2. Undirected Browsers

A different approach altogether is required for people who browse semi-randomly. This behaviour occurs on twitter, on Facebook, and when people are looking around the Web for one thing, but get distracted by another. There's no explicit goal here, so you are relying on mini-goals created by the individual in the moment of browsing. These provide short-term motivation when there is no conscious goal.

The most common mini-goal is satisfaction of curiosity. So you must pique curiosity, and then tease them, holding back fulfilment until the reader coughs up. This picture from @ilana highlights a clever way to do this: show the article comments but not the article itself. An excellent way to build curiosity is to read what other people have been saying about something.

Other mini-goals are revisiting a forgotten goal, affiliating with a friend and various time-management goals such as one more click or getting back to work. Each of these calls for its own pricing mechanisms, but this short article isn't the place to go into them. I'll be publishing something longer on this if you're interested, so leave a comment below or email me ( and I'll send you a copy.


3. Habitual Readers

For many people, reading the news is part of a regular habit. They may do it once or twice a day at certain times, or on a Sunday morning (that particular tradition is already well monetised by printed Sunday newspapers, of course) or when there's a break, a boring moment or a tricky task at work they want to distract themselves from.

For these people, satisfying the habit triggers reminders of certain previous happy events (or deflection of negative feelings such as hard work) and that is the value being provided. Anyone who regularly satisfies a specific need is likely to develop specific strategies for doing so. The publisher's job is to become the habitual starting point of the reader; and to achieve this you need to:
  • achieve trial
  • make payment acceptable
  • make payment automatic

The FT's strategy for this looks effective: offer a little for free, persuade people to sign up very cheaply (4 weeks for just £1), then once they are in the habit of reading, bill them regularly. The habit-forming behaviour is embedded further by regular email reminders of articles that might be of interest.

Some people think the FT can only do this because it has a distinctive business-oriented product; if its news were the same as the Guardian's, or Telegraph's, or Times, it would be very difficult to get people past the first point of payment. But many news outlets have sufficient personality to get people into regular reading habits, and that's an important part of consumption too. Guardian readers certainly see themselves as Guardian readers - it's a badge of identity. (Indeed, Guardian readers probably have a stronger sense of the identity of Daily Mail readers than do the Mail readers themselves). The Times is experimenting to see whether the same principle will work for them.

Mixing it all together

Trying to capture all three of these different newsreading behaviours in one business model is not so easy. This is a different version of the marketer's traditional problem of price discrimination. Can you separate audiences into different levels of demand and charge each of them the appropriate price? In this case, it's about segmenting by behaviour instead of by willingness-to-pay.

It's certainly possible to present different versions of an article according to whether the user approaches from Google (directed searcher), Twitter (undirected browser) or direct from a bookmark or as a registered user (habitual reader). And as long as the price structures are not directly comparable, the segmentation should work. Charging a Guardian reader a sub of £7.99/month does not stop occasional visitors from paying 50p for an authoritative update on what's happening in Libya today; nor vice versa. And the type of articles of interest to directed searchers are probably (mostly) different to those which stimulate the curiosity of browsers.

So identifying key user behaviours should give you the ability to sell to all of them, with a business model that works for each group.
There's much more to talk about on this subject: retaining subscribers once you have them, upselling them to new services, the role of advertising. All of that will be covered in the extended version of this article, so once again: email me ( or post a comment and I'll send it to you when it's ready. 


Mike Young said…
Perhaps we could try alternative "currencies" such as Linden Dollars (2nd life) or Google credits. If people perceive these as "not real money" this may get them over the entry barrier.

The other problem is price anchoring. The price of online news is well and truly anchored at 0

Popular posts from this blog

Is bad news for the Treasury good for the private sector?

What is the difference between cognitive economics and behavioural finance?

Dead rats and dopamine - a new publication