There’s an image that’s repeatedly appeared on my Tumblr dash in the past few days: a screengrab from It’s Always Sunny in Philadelphia originally captioned, “Everybody’s dying, bitch. Let’s get you some fruit,” in which a Tumblr user with the handle “toxiczombie” has replaced “everybody’s” with the words “Tumblr’s always.” As of this writing, it has a quarter-million likes and reblogs. The posting date: November 18, 2018.
Off the top of my head, I couldn’t remember exactly why my fellow Tumblr users were predicting the platform’s death in November 2018. The adult-content ban, a monumental event in the site’s history that undeniably altered its trajectory, wouldn’t be announced until a few weeks later and implemented a few weeks after that. A bit of digging revealed the post was likely related to the ban’s precursors: when the Tumblr iOS app suddenly vanished from the App Store with suggestions it was related to “child pornography issues” and when a rash of blogs, many of them explicitly NSFW, were deleted without warning overnight.
But the comments and tags on the post also serve as a rough sort of timeline for the tumultuousness of the platform’s past five years. The porn-ban user exodus and the ineffectiveness of hastily implemented content filters. The 2019 sale of the site from Verizon to Automattic. The deeply confusing mass tag ban in 2021. The semi-regular flare-ups over interpretations (or misinterpretations) of the language in staff-authored posts. In their own replies on the It’s Always Sunny meme, toxiczombie adds: “Every couple months the notes on this blow up and leave me wondering what happend [sic] this time.”
In November 2023, I know exactly why this post is making the rounds. Last week, the Verge’s Adi Roberston posted a screenshot of part of a leaked memo from Matt Mullenweg, the CEO of Automattic. Within a few hours, Robertson had the full story, confirmed by Tumblr itself: The memo, which dates from early October, laid out Tumblr’s plans to downscale after failing to grow its user base and revenue in the past few years. The majority of people currently working on Tumblr will be moved to other parts of Automattic—the parent company of WordPress, among many other software properties—leaving mostly support and security staff in place.
Shortly after the Verge published the initial leak, Mullenweg took to Tumblr, posting the memo in full and opening his ask box for questions. Some were apprehensive; others were effusive. But it was one from user “catgirlnipples” that spoke for me and much of my dash: “Does this mean no more asanine [sic] UI changes and features nobody wanted? because that sounds amazing!” Mullenweg answered gamely, saying in 2024 they wanted to “really hone the core of the site.” He continued: “I can’t promise we won’t make mistakes—we’re humans. But we will keep a tight feedback loop with the community and users to see how things are going.”
It was a question and answer that underscored a fundamental disconnect between Automattic and Tumblr’s user base over the past few years. Much of the aforementioned tumultuousness happened in Tumblr’s Yahoo/Verizon era—the porn ban and bungled content filtering chief among them—but there was also a persistent sense in those years that under the ownership of a massive faceless media conglomerate, Tumblr could be shut down on a whim. Yahoo initially acquired Tumblr for $1.1 billion in 2013, and in 2017 Verizon acquired Yahoo. When Automattic purchased the site in 2019—for just $3 million, leading Tumblr users to proudly celebrate “losing Verizon a billion dollars”—there was a particular sort of hopefulness in the contrast, that perhaps a smaller, friendlier organization might not just keep Tumblr online but actually want it to succeed.
It’s perhaps ironic (or maybe prophetic) that a site whose users cheered over the previous owners losing a ton of money would proceed to lose the next owners a ton of money. In official communications throughout the past year, Automattic has been explicit about the fact that the site is hemorrhaging cash at an unsustainable rate. Monetization experiments like being able to “blaze” posts into strangers’ feeds, a merch store in tune with the site’s sensibilities, or the beautifully implemented joke, in response to the launch of Twitter Blue, that let Tumblr users buy not just one but two useless checkmarks for $8, have apparently been modestly successful. But many of the big changes, from a widely reviled livestreaming feature and altering the dash to look like Twitter to breaking the core functionality of reblog chains, were all aimed at luring in mythical new users—and they’ve all angered Tumblr’s existing and deeply entrenched user base.
It’s a great irony, too, that this past summer did see an influx of new users. When Reddit began imposing fees on third-party developers, some subreddits went dark or began posting humorous protest content, but a fair number of Redditors migrated to Tumblr. For weeks, the site was awash with welcome posts and guides to Tumblr functionality and etiquette, and former Redditors regularly shared their observations on learning and embracing the site. Rather than expecting it to look and function like another place, they largely seemed invested in learning how Tumblr worked—a clear counter to the trendy idea that the way to appeal to new social media users is to make a platform look exactly like all the others (or, let’s be real, like TikTok).
Because a decade and a half into its existence, Tumblr has a unique way of operating that can seem opaque to the outside world, even if an extraordinary amount of content generated on the site winds up screenshotted elsewhere. That culture is partly due to Tumblr’s core functionality: With its chains of iterative reblogs, it’s long been a sort of “yes, and” improv space, with each addition building on what came before it. Someone tells a story; someone else appends it with art and the caption “I just had to draw this.” Someone writes a paragraph with advice on a situation, and half a dozen others chime in with different perspectives in reblogs. Someone invents a fake movie and the entire site goes absolutely wild with it. On Twitter, I repeatedly saw the complaint that “Goncharov isn’t funny”—but the joke doesn’t work out of context. Tumblr isn’t about a screenshotted post that goes viral on other sites, but about the interchange within the space: It’s my mutuals, trying their hand at a Goncharov post, even if the results weren’t particularly noteworthy, just to be a part of the creative experiment.
Some internet culture commentators have kept pace with Tumblr’s continually evolving ecosystem, but the mainstream narrative in recent years is that Tumblr is dead. Journalists will cite keeping a blog there in the early years—when it felt like a cooler, more indie option than WordPress or Blogger—and rarely mention any social-network elements in their decade-plus-old memories. Similarly, those who cut their teeth in the “Tumblrina” era—the rise of Your Fave is Problematic and social-justice warrior-hood that made Tumblr a punch line on the rest of the internet—often reference dynamics (and users) that migrated off the platform and onto the rest of social media years ago. (That’s not to say that stuff has left Tumblr, but it’s everywhere now.)
The porn ban had an undeniable and detrimental effect, especially for the sex workers and NSFW artists whose livelihoods and safety were threatened when they were forced off the site, and I don’t intend to downplay that in any way. But the suggestion that Tumblr was only about adult content prior to 2018 is false—and the idea that there’s no adult content there now is also false. (Trust me: I’m seeing female-presenting nipples every day.)
Every time I read a “What Was Tumblr” headline I grumble a little, and then I open up Tumblr, where I encounter a vibrant dash full of beautiful art, genuinely hilarious stories, and some of the most unhinged things I’ve seen in my life (compliment). Every time I encounter a screenshot of a Tumblr post on another site with the caption “hey, remember Tumblr?” I prepare myself for the moment I stumble upon the post on Tumblr itself and learn it was written two weeks ago, not in 2014.
This strange gap between the broader internet’s perception of the site and the reality—despite clear evidence to the contrary—isn’t necessarily a problem. I’m a firm believer that our social media spaces would be better if they were less broad. I don’t want everyone to use Tumblr; I want it to work for the people who want to be there.
The answer to Tumblr’s monetization problems might lie in actually focusing on those people who want to be there rather than striving for some untapped base of new users. During his ask-answering session, Mullenweg revealed the site has 11.5 million monthly active users, but only 27,000—just 0.2 percent—have become paid supporters at $29.99 a year. “If that were 10 or 20 percent,” he added. “We could run the site forever.” Tumblr isn’t the only platform that’s aggressively changed its functionality in the past few years chasing some other site’s users, but it is one of the only ones that seems interested in course correcting. The solution might be creating a stable stage for all that yes, and-ing, and in exchange, its users pay to keep the lights on.
It’s been a few days since the big announcement, but when I open my Tumblr dash now, I see zero mention of it. There’s beautiful art and hilarious stories and GIF sets of shows I feel like I’ve now seen by sheer osmosis. There’s a confusingly large amount of content about House, M.D. (Should I also rewatch it?) The rise and fall of the communal creative experiment continues—at least for now. Tumblr’s always dying, bitch. Let’s get you some fruit.
