As Matthew points out, under rules adopted by the AAM last year, publishers can count a subscriber multiple times if, for example, that subscriber is reading the paper in print, on the web, and on a tablet, which has the effect of inflating reported readership. Take out that double and triple counting and the circulation picture painted by AAM looks a lot less rosy.
And he has little patience for the argument that there has always been a fair amount of water in readership numbers, with publishers including things like “pass-along” readership and deeply discounted bulk sales:
“This defence boils down to: ‘Newspapers have always done this, and no one believes these numbers anyway, so what difference does it make?’ A pretty weak defense, you might argue — and you would be right.”
Caveat emptor, I suppose. But as Matthew alludes to, there’s an air of unreality to the whole exercise. Focus on raw circ numbers is a holdover from the era of print, when a reader was a reader was a reader. Publishers focused on the total number because that’s how readership was sold to advertisers. Thus, each reader had essentially the same economic value to the enterprise.
While everyone understood that different readers read different parts of the paper and that some were more avid than others, there was actually a disincentive for publishers to know much about the behavior of individual readers, because they only mattered in the aggregate.
While that’s still true to some extent today, the sooner publishers get past viewing all readers as having equal economic value to the enterprise, the better off they’ll be. What matters in the digital realm, both for advertisers and in terms of how paid content is packaged and sold, is how different readers actually behave, toward the publisher’s content and elsewhere on the web.
Thanks to social media and other tools, readers are now distributors as much as consumers of content. They use content; they don’t just read it. And by using it they help create additional value around that content. The goal for publishers should be to try to understand those ad hoc value chains as particularly as possible. How many is not nearly as important as who, what, when, and where.