What I learned co-founding Vox

What I learned co-founding Vox

Matthew Yglesias — Read time: 12 minutes

What I learned co-founding Vox

Talent is scarce, training is hard, and the competition is smarter than you want them to be

I mentioned recently to former Book Club guest Brad DeLong that I needed some fresh ideas, and he suggested writing about what I learned from my time co-founding and working at Vox.com.

To revisit that period of my life is to revisit a time when I learned an enormous amount about myself, business, friendship, and everything else, so much so that I can hardly conceptualize it. I’ve approached the subject with trepidation that anything I try to say will spiral into a sprawling epic the length of Moby Dick. But it’s a subject that’s been rattling around my head for a long time, and I recently gained some clarity when I mentioned offhandedly to a younger journalist that I thought I’d failed with Vox and she was surprised because Vox is still there, running articles and putting out podcasts and employing lots of people. And while I don’t love every Vox article, they are producing a lot of great stuff like Rachel Cohen on abortion rights referenda and education in the midterms and Li Zhou breaking down the actual content of the congressional marriage equality bill.

Quitting a good job that I liked at Slate to launch a new website was a ton of hard work and a huge pain in the ass! I only did it because I had insanely high ambitions, and I think Ezra and Melissa felt the same way — the current state of the site is a failure only relative to these ambitions.

Ezra ended up as a New York Times columnist with a popular podcast, obviously at the top of the game. But before Vox he ran a popular section at The Washington Post, and achieving his current stature probably didn’t require the detour into running Vox. The point of launching Vox was to accomplish something bigger than adding one more website into the mix. We thought we had ideas around explainers and different ways of thinking about journalism that could genuinely disrupt and dominate, changing the way journalism works. That didn’t happen.

The folks who run Vox Media, Jim Bankoff and Pam Wasserstein, are smarter about business than I’ll ever be, and they have what I think is a reasonable plan to build a large-scale digital media portfolio, of which Vox is just one part. I assume most people who read Polygon don’t think of it as closely aligned with The Cut. And while I can see that The Ringer, while not part of Vox Media, uses Chorus to power its website and Concert to serve its web ads, this is so inside baseball that I’m not sure most people who work there even know.

So when I say that I think that we failed, I’m not dumping on the current Vox management team or editorial staff, who are doing a good job executing a reasonable strategy — it’s just not the strategy we set off with. There’s no innovative editorial project along the lines of our long-abandoned cardstacks. They don’t really do explainers anymore. You’d be hard-pressed to describe what differentiates a Vox article from an Atlantic article or a Slate article or an article on any other quality website. As I wrote in “The Regrettable Death of the Slatepitch,” there has been an incredible flattening of the content landscape, and Vox has become part of that flattening rather than an alternative to it.

This post is my effort to unpack how we fell short of those aspirations.

Hiring is hard, training is harder

One of the things that a lot of us old-time blogger types are good at is writing a lot of words across a broad range of subjects. And those are complementary skills: writing a lot is easier when you can veer across subjects, and veering across subjects is easier when you can execute stories relatively quickly. This was not an important aspect of the analog journalism skillset because editorial staffers had to fight to obtain precious column inches. In digital, the space is limitless, so it’s generally good to do more.

And our vision for Vox really revolved around embracing this aspect of digital more fully than existing publications — we wanted to cover a wide range of topics with more depth and more speed, taking advantage of digital tools but also the timelessness of the web. The founding group included a tight-knit group of very talented digital journalists who all knew each other, were familiar with each other’s work, and were basically on the same page in terms of what we wanted the site to be. And we thought this was scalable — that it would be relatively easy to hire a lot more people and train them based on this vision and our methods.

This was basically wrong.

We did launch right out the gate with some fantastic high-volume writers and video visionaries, and over time Vox has continued to add amazing people, including since I left. But I would say that at least on the text side (video may be different), there was never a point at which a change in investor sentiment that allowed us to double or triple the pace of hiring would have let us easily fill those slots with more great folks. There were people ready and willing to do high-quality, fast-paced analytical journalism across a range of policy areas at the same level as the original group that came over from the Post, but it was a modest number, and many of them already had good jobs at other publications and were hard to poach. Training people who didn’t have the knack for it right off the bat was just massively more difficult than I expected — it turned out we didn’t have any “secret sauce” methodology we could reliably impart to new journalists to make them do work that was Ezra-esque.

When someone is really new and raw, there’s plenty a more experienced person can do to help improve their work. And everyone benefits from good editing and the feedback of other smart people. But just as after 20 years in this industry and plenty of opportunity to work with great editors, I’m still no good at writing narrative features, there was no simple trick to help people develop into the type of journalists we needed if they just didn’t have the instinct for it.

After an initial spurt of enthusiasm for investment and expansion, things pretty quickly pivoted to caution. And while that was disappointing on some level, I think it was clearly the right call. There just wasn’t any amount of money that would have made it possible to rapidly achieve the kind of scale we’d hoped for.

Programming the platforms

The other gaping conceptual problem is that while I believe there was and is an audience for the kind of broad explanatory journalism we wanted to do, there weren’t customers for it. Journalists like to think of our audience as the customers, and one thing I love about Slow Boring is that here that really is true.

But Vox’s readers weren’t customers; Vox was an ad sales business.

And Vox didn’t have the kind of tightly wound sales-audience relationship that you see with a niche product like Playbook or Axios, in part because our aspiration was to create a disruptive mass-market product. There’s another world in which Vox never launches, and Wonkblog at The Washington Post is built out into a somewhat more robust version of itself, a policy-oriented alternative to those kinds of politics-first D.C. publications. But that wasn’t what Ezra wanted to do — the whole point of leaving the Post was to not just do incremental expansion of Wonkblog.

In practice though, it turns out that to get a really big web audience, you need broad distribution from Facebook and Google. In Vox’s early years, Facebook was a dominant source of traffic. Thanks to changes at both companies, the balance shifted heavily in favor of Google by my final years there.

In both cases, to an extent that meant bending to the editorial vision of a different company. To serve the needs of advertisers, you needed to serve the needs of the platforms, and that meant we just weren’t masters of our destiny. Whatever the merits or flaws of the cardstack idea, it never had a chance because it didn’t work with Facebook on a technical level. And people to this day underrate the influence of these companies on the structural properties of the content you consume. Everyone complains about having to read through a few hundred words before getting to the recipe you wanted. But that was because Google gave priority to recipe pages that were structured like real articles. Then there was a time when the big news organizations were structuring lots of stuff as live blogs because Google was giving priority to live blogs.

There’s no sense in blaming the platforms for the problems this engendered. The underlying issue is that the web advertising audience has so little economic value that the customers demand a gargantuan scale. On video and audio, the audience is more valuable on a per-person basis (or at least has been so far), and I don’t think it’s a coincidence that the best manifestations of Vox were often in video and podcast form. You can serve the audience well when the audience is valuable.

But this platform dependence made product innovation essentially impossible. It was also editorially constraining. I think it’s very under-appreciated by people in the technology industry that most of the media trends they deplore are direct consequences of Facebook’s influence over journalism in the mid-2010s. The huge shift in media sensibility that happened during this time was absolutely not what editors and senior figures in the industry wanted to see. Even in retrospect I don’t really understand why this was the case, but objectively speaking, hard-core identity politics and simplistic socialism performed incredibly well on Facebook during this period. That doesn’t mean journalists started pretending to be left-wing to get clicks. But people who had some authentic left-wing opinions found that writing on the subjects where they were the most left tended to generate the most traffic, and early career journalists with authentic leftist views outperformed their colleagues. So you ended up with this whole cohort of discourse structured around “Is Bernie Sanders perfect in every way or is it problematic to vote for a white man?” as the only possible lens for examining American politics and society.

The triumph of the New York Times

The dream of executing a low-end disruption on the incumbent news media was also undermined by the fact that the New York Times just refused to be disrupted.

Part of my vision for disrupting the New York Times was a belief that they’d find it beneath their dignity to write webby headlines or to copy gimmicky explainer article formats, but their digital transformation project was very smart and forceful, and they just did it even though some people at the Times didn’t want to.

We also kind of hoped that by not respecting the line between news and editorial, we’d create a situation where our staff was considered unhireable by prestige news outlets. After Max Fisher wrote a smoking hot take like 2015’s “Israel's dark future: Democracy in the Jewish state is doomed,” surely he’d be appropriate for op-ed work only. But they not only hired him and Amanda Taub away from Vox, but they also had the vision to create a whole new editorial product — the Interpreter — for the two of them to write, blurring the editorial/opinion line in just the way that Vox articles did. And once Max and Amanda left, I think it was clear to everyone that “get a good job offer from the New York Times” was a great capstone to a successful career at Vox. They hired Sarah Kliff and Brad Plumer. They hired our wonderful behind-the-scene editor Eleanor Barkhorn. They hired Jim Tankersley and Jane Coaston. And ultimately, they hired Ezra.

Because the New York Times is imperfect and because the New York Times is also the most important journalism institution in the world, people are constantly mad at them about one thing or another (me, too!) which tends to distract attention from how damn smart they’ve been.

But the Times has an editorial strategy that works at a large scale, even though there aren’t 100 Ezra Kleins in the world. They bundle and integrate different forms of excellence from across the journalistic landscape. Choire Sicha, who pioneered a very different kind of digital journalism, was an influential editor at the Times and Ben Smith who ran Buzzfeed had a brief but spectacular run there as a media reporter. The best conservative columnist in America — not just one who writes things liberal NYT readers agree with, but one who succeeds in writing things that are genuinely hard to dismiss and make them squirm — works for the New York Times. They have enough scale to have a substantial brand advertising revenue stream, but it’s a subscription-based business that generates a lot of revenue from readers. Indeed, the fact that their readers are paid subscribers only serves to increase the audience’s value to advertisers — it’s proof that New York Times readers actually care about and value the New York Times.

It seems silly to say that combining paid subscriptions with advertising is a genius business idea since it’s how basically every daily newspaper worked. But pulling it off in the digital space wasn’t easy, it wasn’t obvious there was a path to success there, and making it work has involved a lot of editorial flexibility and smarts.

What did we learn here?

The biggest lesson I took away from all of this is that launching and running a business is hard.

It’s hard primarily because it stacks a bunch of different medium-difficulty problems on top of each other — you need a product that people want but also a revenue strategy that supports it, and you also need to be good at recruiting and retaining good people. You need some good luck, but part of that is having a lot of situational awareness about how the world is changing so you can “make your own luck” rather than just hoping things will break a particular way. You can have a good idea and a great team (which is what we had), and it still doesn’t really work because there’s so much more to it than that.

One of my most neoliberal shill beliefs is that journalists and academics and intellectuals tend to underestimate businesspeople because they underestimate how difficult this stuff all is. And I think that leads to errors of political judgment.

Now to be clear, an appreciation for the intelligence and savvy of businesspeople doesn’t mean you need to genuflect before them or do everything they want in politics. A business executive being extremely smart and savvy is consistent with him pushing for tax and regulatory favors that are good for him personally but bad for society, and the right response is to say “no.” But I think you see every time a successful business person gets involved in a public controversy, the instinct on the left often isn’t a firm, respectful, “no, that idea doesn’t serve the public interest,” it’s to denigrate the person’s competence or intelligence and suggest that we ought to take the rich guys down a peg. I don’t think that’s right analytically. And I also don’t think it’s reflective of how more economically egalitarian societies function. The distribution of material resources in Denmark is much more equal than in the United States, but the executives of leading Danish business enterprises are still a really big deal in Danish society and their opinions carry weight.

At the same time, I wish everyone who’s critical of “the media” (i.e., everyone) would take more seriously the fact that media enterprises are businesses, and they are subject to commercial forces and pressures.

That’s not to excuse bad articles, which absolutely run, but it is to say that if your model of decision making doesn’t foreground the basic business model issues, you’re misjudging the situation.