When Yahoo CEO Marissa Mayer officially announced the company’s blockbuster Tumblr acquisition, she didn’t shy away from the numbers. In fact, a good bit of Mayer’s blog post on the acquisition was devoted to rattling off a laundry list of Tumblr’s statistical accomplishments: 105 million different blogs. Over 300 million monthly unique visitors. 120,000 daily signups. 900 posts per second. 24 billion minutes spent on site each month. But one number, arguably the most important, was missing.
How many actual users does Tumblr have?
Or, to rephrase the question for Yahoo’s shareholders: How many living, breathing, income-having humans can Tumblr actually reach with its ads?
The answer, it turns out, wasn’t just left out of any announcement — it’s hard to come by. Twitter boasts about how many people log into its site every month; so do Facebook and Google. The monthly active user count is, in many ways, the most important number in tech. But if you ask Tumblr how many people log into their dashboards on a regular basis, your question will be met with silence. (Previous requests have been denied; another has been sent for good measure.)
In situations where such a number might be relevant, Tumblr tends to either divert attention or obfuscate. Take Tumblr’s ad deck, which explains out Tumblr’s primary “Dashboard” native ad product. According to the deck, which cites Quantcast figures from March 2013, Tumblr delivers “20 billion pageviews” and “225 million uniques a month.” This is true of Tumblr, the platform.
This is not true of Tumblr’s dashboard, which is only part of the Tumblr platform but is its only ad product. Why these numbers are in the ad deck is clear: they’re impressive. But that doesn’t mean they’re relevant.
Unlike Facebook, where users are generally required to log in to use the service in any way, there are two ways to experience Tumblr: as a visitor to one of its millions of blogs, some of which are large, popular websites in their own right; and as a dashboard user. The dashboard, which avid users might refresh dozens of times a day, is very good at generating page views and “engagement” time. The sites, on the other hand, are very good at generating unique visitors who aren’t necessarily users.
Someone who stops by to gawk at Hot Dog Legs for three minutes, for example, is irrelevant to Tumblr’s advertising products. Much more relevant to advertisers than a global “uniques” number would be the active user number — the total potential audience of a Tumblr ad campaign.
Analytics firms, including Quantcast, which Tumblr allows to directly measure its traffic, are unable to separate the dashboard numbers from the unregistered blog views; they too conflate uniques and users, which seems to be how Tumblr wants it. Numerous sources BuzzFeed spoke with in the advertising industry confessed that not only did they not know the dashboard numbers, but that, in some cases, they weren’t aware of Tumblr’s big difference between a dashboard user and a unique visitor. One source pointed to a May article by Peter Kafka, which speculated Tumblr has between 30 and 50 million active users, as the best guess as to Tumblr’s monthly unique dashboard traffic. “It’s a back of the envelope way to back into it, but it’s the best best/only guess I have seen,” the source said.
Assuming that estimate is in the ballpark, which most sources agreed to be the case, Tumblr is a very different service than portrayed in its ad deck — a 30 to 50 million-user web service that is substantially smaller than either Twitter or Instagram. And while it isn’t exactly unheard of for an ad deck to fluster clients with a parade of confusing numbers, Tumblr’s feels especially misleading — it implies, unsubtly, that ads can reach a potential audience of 225 million people. In reality, that number is almost certainly much, much lower.
That’s not to say that Tumblr is lying to advertisers. Its promises are carefully worded and guarantee impressions and proportional reach. Tumblr’s dashboard ad unit, Radar, draws more than 120 million daily impressions, and the network’s user base is fanatical enough that the ads are put in front of enough eyeballs to meet clickthrough requirements. And it would be an extremely large campaign that needed to reach more than 30, or 50, million users.
Former Tumblr editor in chief Chris Mohney sees this as a non-issue. “Anything’s possible, but, in my opinion, Tumblr has little reason to mislead in an ad deck,” Mohney said in an email. “People fluff those numbers when trying to achieve a certain threshold or play in a certain league, and even a reduced figure puts Tumblr above such concerns really.”
But if Tumblr is truly “above such concerns,” the question remains: Why not share how many users it has?