Do Newspaper Paywalls Solve the Online Revenue Issue for Shareholders?

Given the New York Times earnings call this week and all the chatter about their paywall, my curiosity was piqued about how shareholders should assess the paywall ability to offset low online advertising revenue.  The following analysis explores that topic.  The assumptions and conclusions are based on both public information from the 2009 Annual Report for The Washington Post Company and Scout Analytics experience.  The scenario is hypothetical but representative.

As print circulation and revenues decline, digital audiences and revenue have to take their place.  Because online advertising has not yielded the same revenue as print advertising, many publishers are looking to paywalls to solve the digital revenue problem.  The conclusion below is that the yield from online advertising has to improve or many shareholders will lose on their investments.

In 2009, total revenue from the Washington Post newspaper operations was approximately $529.4M.  Print advertising revenue was $317M or 68% of the total; subscription revenue was 29% ($133.2M); and 3% ($13.8M) came from other sources.  Overall digital revenue was $99M across all newspaper operations, but washingtonpost.com was the majority ($70M is assumed for this analysis).  Although not the case for the Washington Post newspaper operations in 2009, this analysis assumes a steady-state business with a pretax profit of 15%.

To maintain valuation on the move from print to digital, the newspaper operations need to continue to deliver the same profits (i.e., P/E ratio).  From a shareholder perspective, the 15% profit from the $459.4M of print revenue needs to become digital profits.  Because digital does not have the print production and distribution expense structure, a lower amount of revenue should be able to generate the same profits.  Assuming circulation revenue roughly covered print and distribution costs, digital would need to produce $325.2M ($459.4M-$133.8M) of incremental revenue to maintain the earnings and shareholder value.  The total digital revenues would need to be $395.2M ($325.2M + $70M).

The 2009 average revenue per user for digital was the $70M we assumed divided by 23.9M monthly uniques stated in the annual report or $2.93.  Generating the $395.2M requires either an increased ARPU or a larger audience or of course both.

To produce the $395.2M on current digital ARPU, the Washington Post would need an audience of 134.9M or 5.64 times the current audience.  Because changing the size of the digital audience to substantially impact revenue is not likely, the logic is to increase ARPU and do so with a paywall.

Let’s assume the Washington Post decides to put up a paywall and assume it has a meter such that once a reader reaches a predefined limit on articles per month the reader needs to subscribe.   As a result, only fans of the newspaper will purchase an online subscription.  Our studies have shown fans make up about 5% of the audience.  For the Washington Post, they would have had 1.19M fans in 2009.  Assuming the paywall has the same rate as the print product $280/year and a 50% of fans convert to a subscription, the revenue from the paywall would be $167M which would require $229M to come from advertising which was only yielding $70M.  The shortfall would be $159M.

At this point, a newspaper operation would have one of two options: cut expenses to achieve profitability or increase yield of advertising revenue.  Cutting the expense would result in lower earnings per share and lower valuation – the shareholders would lose money.  To generate the necessary earnings, the yield in advertising revenue has to be addressed.  Yes, a paywall can soften the blow of lower online advertising revenue, but it is not a silver bullet.  Advertising yield has to improve 3X or more to keep existing shareholders whole.

A final note.  My revenue projections for the paywall are likely to be generous.  A 50% conversion of fans is high as is the $280/year rate.  The actual paywall revenue for this type of digital operation is likely to be lower.

About Matt Shanahan

Matthew Shanahan brings Scout Analytics nearly 25 years of experience in the execution of business transformation. His specialties include business model innovation and new market development.
Share this Article
facebooktwittergoogle_plusredditpinterestlinkedinmail
About| Privacy| Terms of Use| Contact Us
© 2014 ServiceSource International, Inc. All rights reserved. Various trademarks held by their respective owners.
beste kredittkort . buy term paper . hcg injections . I found my new prescription for Zitromax as one of Pfizer's azithromycin