Update: Trying to write about Gannett and accurate numbers simply isn’t possible. One reader notes that USA Today didn’t start offering digital subscriptions until 2021 — and yet Gannett was reporting paid (or unpaid?) digital for USA Today to the Alliance for Audited Media starting at least in 2012. So how is that possible? Another reader hints at an answer — if you subscribe to any Gannett paper, or maybe just any Gannett daily, you get a subscription to USA Today included. Or you used to. Maybe that changed after USA Today’s paywall went up.
So it could be that USA Today’s paid circulation was far lower in 2018 than what it reported to AAN — not the 2,632,392 that Joshua Benton used, and not the 1,584,462 that I used. Instead, maybe what we ought to look at is the 631,076 print figure. And since USA Today seemed to be selling an e-paper option as well, that would bring total paid circulation in 2018 to 654,743.
Now let’s go for an apples-to-apples comparison. The 156,453 that Benton reported for USA Today’s current paid circulation is the total of print and replica. That’s a nausea-inducing decline of 76% over the four-year period, but that’s still not nearly as much as the 93% Benton’s numbers showed. It’s also a lot worse than the 33% estimate that I offered.
But wait! USA Today has been selling paid nonreplica digital subscriptions for nearly two years now. How many? As I explained, Gannett stopped reporting that figure a while back, so we don’t know. Surely it’s not the “zero” that Gannett claims on its most recent report to AAN. (It should at least be one; I mean, I bought one.) We simply can’t know how by how much USA Today’s paid circulation has declined without knowing that important figure, or whether subscriptions to other Gannett papers are included. Without access to Gannett’s internal numbers and insight into exactly what they mean, it’s an unsolveable mess.
Earlier: Did USA Today’s paid circulation drop by 93% between 2018 and 2022? The near-certain answer to that is no — yet that’s the astonishing claim that Joshua Benton makes at Nieman Lab. I knew there was a problem with his numbers as soon as I saw them, mainly because I recently put some effort into figuring out how USA Today’s corporate owner, Gannett, compiles its circulation figures. So let’s dive in.
Benton reports that USA Today’s paid circulation in the third quarter of 2018 was 2,632,392 and then fell in the third quarter of 2022 to just 180,381. That’s a staggering loss of 2,452,011, or 93%. But as I’ll show, much of that apparent loss is the result of a change in the way Gannett reports its paid digital circulation to the Alliance for Audited Media.
What I was able to dig up at AAN uses slightly different time periods compared to what Benton found. I’m going to use all of 2018 rather than the third quarter because the latter wasn’t available when I looked. But it should tell the same tale. It shows that the average weekday circulation that year was 2,708,983, which is in the same ballpark as what Benton reported. A lot of that, though, consists of “affiliated publications” such as Local/Life and Sports Weekly. The circulation of the paper alone was 1,584,462. Now, pay attention to the following breakdown, because it will prove important:
- Print: 631,076
- Digital replica: 23,667
- Digital nonreplica: 929,719
“Digital nonreplica” is the term for digital subscribers who access the website but don’t bother with the e-paper. As you can see, it comprises the vast majority of digital subscriptions — and, at some point, Gannett simply stopped reporting that number.
Now let’s look at the third quarter of 2022. Paid weekday circulation is reported as 180,381 at the top level at ANN (the figure Benton used) or 156,453, which is the number that pops up at AAN if you click through. That latter number comprises 132,176 for print and 24,277 for digital replica (the 156,453 figure, which I didn’t immediately grasp) — and zero for digital nonreplica. So, yes, print circulation is down by a stunning 79%, which may have more than a little to do with the COVID-19 pandemic. USA Today, after all, was a staple of hotels for many years. But digital replica is up slightly. And digital nonreplica simply isn’t being reported.
I encountered this recently when I was analyzing some numbers for Gannett’s Burlington Free Press in northern Vermont. I discovered that, not only had Gannett stopped reporting digital nonreplica, but that — according to confidential internal reports I had obtained — it was underreporting its total paid digital circulation by about half.
Gannett is trying very hard to sell digital subscriptions for its incredible shrinking news outlets. Keep in mind, too, that people don’t buy subscriptions to the replica edition — they buy digital subscriptions, period, and the papers themselves report how many readers are accessing the e-paper so they can tout that number to advertisers. (AAN recently explained all of this to me. As you’ll see, it’s pretty complicated.) In other word, Gannett is telling AAN how many subscribers are accessing the e-paper, but they’re keeping total digital circulation to themselves.
Now, I’m going to take a leap here and assume that USA Today’s total digital circulation was the same in 2022 as it was in 2018, or maybe even a little higher. I base that on several factors: digital circulation was up at all of Gannett’s New England properties, according to the confidential report I mentioned; USA Today’s digital replica circulation was up slightly; and Gannett has been pushing digital subscriptions hard. I even signed up for one, and it was a great deal — with a little fiddling, I can use it to access every Gannett paper in the country. Of course, there’s little in them.
With all that in mind, I came up with a guesstimate that USA Today’s paid circulation in the third quarter of 2022 was about 1,056,000. I’m building in a nonreplica figure of 900,000, a decline (as I said, unlikely) compared to 2018. Put all that together, and using a 2018 circulation figure of 1,584,462 (that is, not counting “affiliated publications”), and I come up with a drop of 33% between 2018 and 2022. Now, that’s still a lot — but it’s also in line with a lot of non-Gannett papers that Benton used for comparison.
Everything else Benton says about Gannett is right on target. The company has decimated its papers, is closing them and selling them off, and generally appears to be squeezing out the last few drops of revenue they can muster before people like top executive Mike Reed, the $7.7 million man, walk away. It’s an outrage, and we really can’t call attention to it often enough.
But the crazy circulation drop at USA Today and other Gannett dailies is more a function of Gannett’s decision to stop reporting paid digital nonreplica subscriptions than it is an actual measurement of readers fleeing for the exits.
Outing an anonymous commenter leads to a libel suit against Nieman Lab
By Dan Kennedy
On July 20, 2021
In Media
Is it acceptable for a website operator to make use of registration data not known to anyone else in order to expose the identity of an offensive commenter? That’s one of the main issues in a libel suit against Nieman Journalism Lab founder Joshua Benton. Bill Grueskin explains the case in detail at the Columbia Journalism Review. (Disclosure: I know and like Benton, and wrote for him from time to time when he was the Lab’s editor; he is still a staff writer. I continue to contribute to the Lab occasionally.)
Way back in 2008, when the internet was still powered by coal, The Eagle-Tribune of North Andover did something similar. I wrote about it at the time. A Haverhill city councilor was caught posting to the newspaper’s website under 38 different screen names. The Eagle-Tribune outed him using information no one else could have known, arguing:
The average citizen does not take an oath to serve the public. An elected official does. An attempt to deceive the public is clearly not serving it, and a public official who does so is not only undeserving of the protection of confidentiality, but deserves public criticism.
Two differences between the cases. First, the person suing Benton, former Temple University journalism professor Francesca Viola, is not a public official. Second, Viola claims that in addition to exposing her for comments she made at Nieman Lab, Benton also attributed to her anti-Muslim comments made on another site — and she contends she did not make those comments.
As Grueskin notes, these problems can easily be avoided by requiring commenters to register and post under their real names. But, he adds, “an administrator can’t have it both ways, promising anonymity and then using special access to expose someone’s identity.” I agree — and I remain troubled by the choice that The Eagle-Tribune made nearly 13 years ago as well.
Become a member of Media Nation for just $5 a month.