January 31, 2011

A Little Free Press

How much content should a news organization give away?


Sometime early this year, newspaper industry standard-bearer New York Times will put up a paywall on its site, its second attempt at creating an additional revenue stream to supplement waning ad sales. An earlier effort, TimesSelect, ran from 2005–07 and limited unpaid subscribers to news content while blocking access to columns and editorials, restricting access to its most popular content to paid subscribers only. The new paywall will be more flexible, allowing all online readers to view a number of articles each month before paying a flat fee that grants unlimited access to the Times’ full site.

The Times is not the first to implement this kind of paywall, but its venerable position means that competitors and less prominent papers are likely to take note and mimic if it’s successful. Most often described as a meter, the new system can also be described as sampling.

The free sample is a tried-and-true marketing tactic: snacking on hors d’oeuvres in the supermarket or test driving a car reduces uncertainty — especially about the new or unfamiliar — by offering consumers an immediate experience of a product. For media, entertainment, or information businesses, sampling is an exceptionally attractive tactic because it costs little or nothing, particularly in the age of online distribution.

The one hitch is that, unlike other goods, once a consumer takes a paper off the rack, reads an article, and puts the publication back, she’s already consumed it. This tension is as at the center of what the economist George Akerlof famously labeled the information paradox. Once content has been consumed, its value to the reader has evaporated.

Media organizations have long employed sampling tactics to manage the information paradox, such as providing free trial-subscriptions or complimentary issues to convince consumers that future issues will contain the same kind of valuable information as current issues. But the rise of the web has heightened competition and depressed ad revenues, making the question of how to strike the right balance between free samples and paid products more salient — and perhaps more complicated — than ever; few if any major news outlets have not experimented with permutations of free and paid online content.

When Florian Stahl of the University of Zurich proposed creating a technique to determine the optimal amount of free information goods a firm should offer, and how it should price its paid content, Professors Oded Koenigsberg and Donald Lehmann joined Stahl and his Zurich colleague, Daniel Halbheer, to explore the issue. The researchers created a model that breaks consumer demand into two parts. The first part describes the impact of the size, quantity, and quality of the sample that consumers receive on initial purchases. The second part captures what the sample allows consumers to learn about the product — for example, whether the sample was interesting, useful, or informative — and how this influences whether they later purchase the product.

Overall the model captured the two components of purchases of the paid product in conjunction with estimated ad revenue from both free samples and paid versions. By exploring a range of sampling and pricing combinations, the researchers determined the optimal balance of sampling and pricing single-use, single-edition content, for example, a novel or an article. The decisions recommended by the model correlated closely to those used in the news industry, as measured by an analysis of a select group of German media firms.

Nevertheless, the researchers found evidence that media firms may not be offering large enough samples to convince consumers to purchase paid content. Increasing the number of samples given out leads to an increase in views or circulation that in turn generates increased ad revenues. More or bigger samples also provide consumers with greater opportunities to update what they know about content: a free sample—provided it’s big enough and of higher quality than expected—can entice a customer who might not have otherwise considered purchasing the product.

One innovation of the researchers’ model is that it allows for a scenario where the firm collects ad revenue through the paid version of its content as well as the free version. While the model considers single-use, single-edition media, it could easily be extended to other dimensions of information goods, such as newspapers (single-use, renewed), reference books or music (multiple-use, single edition), or magazines (multiple-use, renewed). The model can also be adapted for use in a wide variety of industries to help firms determine the optimal amount of free goods they must offer in order to stimulate sales.

Oded Koenigsberg is the Barbara and Meyer Feldberg Associate Professor of Business in the Marketing Division at Columbia Business School.

Donald Lehmann is the George E. Warren Professor of Business in the Marketing Division at Columbia Business School.

Read the Research

Florian Stahl, Oded Koenigsberg, Daniel Halbheer, Donald Lehmann

"Sampling Information Goods: How Much Should Be Free?"

View abstract/citation  Download PDF  

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>