We’ve seen lots of consternation over numbers recently. Take Groupon’s foggy (more opaque than fuzzy) math now being dissected by the SEC. Dissatisfied with all the usual metrics-for-investors the business world has produced, it decided to create its own: “adjusted consolidated segment operating income,” or adjusted CSOI.
The excellent WSJ piece, by Shayndi Raice and Nick Wingfield, lays it out well, though it’s a pinch-yourself, “Is-this-The-Onion?” kind of story. As one portfolio manager said, “In essence, Groupon is asking investors to look at their profit before any expenses.”
In June, we read the Triumph of Charts over commonsense, with the Huffington Post surpassing The New York Times in unique visitors and the UK’s Mail Online now the second most-read site in the galaxy. I remember the good old days when it was hard to put up a chart on the web. Now, it’s so easy, anyone and everyone is doing it. Too often, it’s form before facts of consequence. Too often, charts feed delicious, SEO’able headlines that drive incredible traffic…that put some sites high up on the charts. It’s quite a circle, though lacking virtue.
Others (check out Steve Myers’ smart Poynter piece) have pointed out fallacies in the NYT/HuffPo comparisons. Let’s add to them with a look at the newsonomics of ARPU, or average revenue per unique visitor. It’s a great benchmarking metric, long used by telcos and in the cable TV industry, and one being increasingly used, though not publicly, in the digital news industry. In addition, it’s a revealing number when we look at such players as HuffPo, the Daily Mail, and NYTimes.com.
ARPU basically says: Don’t tell me how many customers you have; tell me how much money you are making on each of them. While it’s not the only number anyone wants to use to run a digital business, it’s a big piece of the puzzle — especially as we compare like companies to each other.
Let’s look at the ARPU of traffic. I’ve used full-year 2010 data, the cleanest available, and extrapolated where necessary.
For NYTimes.com, I’m estimating digital ad revenues of $170M in 2010. That’s 80 percent of total reported digital revenues for the Times News Media group overall. The Times represents just under 70 percent of total company revenues, and, clearly, the Times itself is driving more digital revenue, proportionally, than the smaller papers in the company. So 80 percent should be close.
In December 2010, comScore reported 48 million global uniques for the Times. So each unique would be worth $3.54 for the Times for the year. (Of course, uniques vary by month, and domestic uniques — 32 million or so — are worth more than non-domestic.)
For the same month, comScore reported 31 million global uniques for the Huffington Post. Most 2010 revenue estimates for HuffPo come in at about $30 million. So, in 2010, each HuffPo unique would be worth 96 cents.
Let’s take two supposed competitors in the UK, both in the news business, both selling advertising but attracting quite different audiences.
The Guardian took in £37.5 million in digital revenue in 2010. Using the December ABCe number of 39 million uniques, each unique was worth about £.96, or $1.53 at today’s exchange rates.
For the Mail, I extrapolate, from its reports, about £16 million in digital revenue for last year. Using the March (aligning with its reporting period) ABCe unique number of 66 million, I figure each unique visitor is worth about £.24, or 38 American cents to the Mail.
If close to right, the value of a unique visitor is 3.5x greater for the Times than for HuffPo, in advertising. It’s 4x greater for the Guardian than Mail Online.
Why the differential? Reasons run a wide course. Take your pick from:
(As another comparison to our news calculations, it’s intriguing to run the numbers for Groupon and Twitter. Twitter has about 139 million uniques (May, 2011) and maybe revenues of $150 million this year, or $1.07 ARPU. Groupon, with $460 million in U.S. revenue in 2010, estimated by Techcrunch, and about 10.7 million unique visitors in December, would have an ARPU of $42.90. That’s off the charts — and why it has attracted its valuation, despite its “profits” accounting.)
Whatever deeper analysis will show, the ad revenue numbers are real. Would you rather have the Times’ $170 million in digital revenues or HuffPo’s $30 million? (I know Tim Armstrong’s answer, and you AOL shareholders can sit down now.) Of course, it’s true, HuffPo/AOL traffic may continue to ramp up (or not), on a much-smaller cost base. It’s also true that the Times, still profitable, owns a huge brand equity — now being leveraged in digital circulation money as well — and has lots of upside, as it is challenged by its own legacy cost burdens.
Whatever kind of battle this is, it’s not a battle of equals, and it’s not a battle that can be understood by charting unique visitors.
Unique visitors are a great dumb count. As I’ve noted, it’s as if in the print world we counted the everyday subscriber — consuming 5 hours a month of a news publication — the same as someone who, standing on a Midtown corner on a windy day, happened to catch a sheet of flying newsprint as she held up her hand to hail a cab. Hardly equal, yet that’s what unique counts level.
Unique counts play to the wonder of Google search and, now, by Facebook and Twitter touts, but they are increasingly meaningless in a world that still seems to operate on a single currency: currency. Expect the bounce rates (hit one page and then leave the site) of the fly-bys only to increase in our new age of ubiquity, with mobile devices providing everywhere-and-anywhere access. It is hard not to run into big brands: Add to the Times, the HuffPo, the Guardian, and Mail Online such top-of-Google sites as Examiner.com and eHow.
Counting unique visitors — increasingly — is like counting air.
ARPU itself is just a beginning at better counting. Some will say it’s too general, the “A” as in average, just too broad to be useful. So companies can segment it, especially as they value core customers — say, the RPU of readers who read 50 pages a month compared to those who don’t.
Consider, in addition, how ARPU can be stretched and fine-tuned: mobile ARPU, smartphone ARPU, iPad ARPU, video-consuming visitor ARPU. Into the future, as each digital reader is offered an array of niche (sports, travel, health) products, increasing the ARPU of core readers becomes even more important. Much easier to upsell a customer, in any trade, than get a new one.
In addition, increasing ARPU is a better investment, says Scout Analytics’ Matt Shanahan, than either increasing sales volume or decreasing sales expense.
Some execs told me that ARPU is getting more important in the age of paid reader digital access, as, this week, Time Inc. ratifies that new age with its all-access provisioning for all 21 of its consumer magazine titles. While eCPM (the effective cost-per-thousand rate publishers get for their advertising, taking into account sponsorships and several sales types) is the preferred metric for ad efficiency, the emerging ARPU number can combine both how much a unique visitor provides in subscription (or pay per view, day, week) revenue and how much advertising revenue, on average, that unique enables.
Bonus question of 2011: What does the cross-platform (weekend print/daily digital, for instance) ARPU look like?
This isn’t an idea that’s alien to the legacy newspaper and magazine business — we’d have to combine a few legacy numbers to get an Average Revenue Per Print Subscriber number — but the twists and turns, and added data, of the digital business give ARPU and its offspring increasing relevance.
It is all coming down to the same two revenue sources — circulation and advertising — just moved to the new digital business, gradually. (Lee Enterprises this week accepted Journalism Online’s advice and is up-charging print subscribers as it rolls out six tests.) As everyone toys with reader pricing, bundling, and add-ons, add up circulation and advertising, and we’ve got those increasingly familiar economics of transition.
Image by Thiophene_Guy used under a Creative Commons license.