Pivoting to the Expert Economy

Danny Crichton:
The media world has been rocked by artificial intelligence, labor strife, the creator economy, the decimation of business models and so much more. But sometimes it's not collapse and crisis that's the most interesting story, but rather just another day of a assiduously growing a platform. That's the story I want to talk about today on risk gaming, and we're going to zoom in on Medium. It's a venerable media business founded by Ev Williams all the way back in 2012. And one that has become notorious for its pivoting dance to a brighter media future. But under Tony Stubblebine who became CEO two years ago. The company has reached cash flow break even, and he believes Medium has found a balanced business and media model for the decade ahead. I wanted to learn more, so let's dive in. Yeah, so maybe you should just dive into tell me your story. I knew you had joined, but obviously Medium is sort of locked into my head as someone who's followed it since its launch 12 years ago I think at this point, with Ev Williams and different business models over the years.
So how did you sort of stumble upon... And that's sort of a social media joke, how did you sort of stumble upon Medium and get involved in the program there?

‍

Tony Stubblebine:
There's actually a bigger story than just stumbling upon it? I think I've probably been in Ev Williams orbit since 2005. I was his head of engineering back when he was doing a podcasting company in Odeo. And I was Jack Dorsey's boss when we built the first version of Twitter. So I was publishing on day one and I just fell in love with the product. The idea of how beautiful it was to use, this inspired me to write more. And then I loved it as a writer and I loved it as... The sort of famously Medium has pivoted a number of times, but the main pivot in my mind is this business model pivot where Ev just he woke up one day, whatever, he is probably been thinking about it for years and I was like, "You can never build anything great on the internet with an ad-driven model. It'll pervert everything."
Essentially was his hypothesis, let's switch to a subscription. I ended up being the first external partner into that subscription, and I ended up growing what I was doing for them. I think because I actually understood the economics of what would work on Medium faster than anyone else, and so I grew what was going on there to the point where I was just a giant massive partner. I think something like 2% of medium traffic was going through something I was doing. It's like on a platform, no Power user should have 2% of the platform. That's almost too big. I could tell that Ev was floundering a little bit, I think it's fair to say. So I kept in touch with him and eventually made the case that he should step down and I should step in. And two years ago took over to essentially do what I would call two turnarounds. The first turnaround is just standard business. We were losing subscribers, we were not differentiated and we were burning cash for no reason.

‍

Danny Crichton:
Everything that a venture capitalist wants to hear.

‍

Tony Stubblebine:
Yeah. And that comes with all the problems, we hadn't been able to attract investment in forever. And so basically I had two mandates. One, get to profitability. And we just last month had our first cash flow positive month ever. So that's the first turnaround, but that's not that fun. There's some pride in being an operator, but for me the reason to do all of that is can I do a second turnaround? Can I prove that Medium can be a true gem on the internet again and forever? Now that we have a foundation to go from, this is like we know there's a lot that's broken on the internet. People don't want to admit this, but one of the obvious solutions is if you put a paywall in front of it instead of ads, you get the right incentives and you actually could build something delightful.

‍

Danny Crichton:
So obviously I've worked with paywalls for years, so TechCrunch build out a paywall from zero. TechCrunch from its founding was an ad driven site and as you've already pointed out, huge disincentives in the media business when you are ad driven. You're trying to shock, you're trying to write headlines that get clicks. You're not necessarily getting the most information and we don't get credit for saving you time. And I always describe this as an editor, things that should just be a headline that was in a summary article become a full article because we need that click. We need that ad space to be sold. That's how we make our money. So we have to, even if it was only a paragraph, we've got to stretch it out and make it work. This description makes sense. Of course the big challenge is no one wants to pay for media, from the beginning of the internet from '94, the model has always been free and there's some other subsidy that sort of magically covers the cost.
We were not built to pay up in front except for the ISP. And so it's a real struggle for any media business as a former investor in Patreon through a former firm, seeing this through other media companies through my own work, you just have this struggle of every single person that's beginning this every single time. Card costs are very high. It's very hard to sustain a long-term value given that people unsubscribe. And so I'm curious when you think about delivering that basket of content and saying, "Look, here's this relationship we're trying to build. Here's what you get. Here's what you pay for." How did you find a balance to make that successful, particularly given how late of a stage you were in your tenure sort of into the game as opposed in the first year or two?

‍

Tony Stubblebine:
I felt like Medium had this Goldilocks problem. There's a lot of parts to the problem, but there's this really specific one which is how do you deliver quality worth paying for? We tried it one way, which is this massive editorial undertaking. And I think at one point we had 70 editors working for us. And they were out paying authors to write, on a given month there might be 300 people that we're paying to generate that. It worked to solve the quality problem, but financially it's funky. The first thing that I thought when I was reading your summary of where the TechCrunch subscription went was like, oh, you were paying at one point $10,000 in article but what's the per article cost? It's actually probably for you 10,000 plus editorial.

‍

Danny Crichton:
Yeah, I think all configured could probably argue was 20 to 25K with illustrations and editing and copy editing and the whole shebang.

‍

Tony Stubblebine:
So we backed away from having in-house editorial because it wasn't working. And went back to a pure attention driven UGC and what we got was essentially the world's biggest content mill. A lot of listicles, a lot of punditry, very little of it informed by any deep experience. That was wrong in the other direction. I just happened to have had an early experience in my career with how powerful it was to hear from the middle of that. The amateur writer who is a professional or has personal experience in something else. So I had gotten my start at O'Reilly Media, which is publishing programming books. Before there was programming information on the web and the whole game at O'Reilly was let's take a programmer and teach them how to write rather than let's teach a writer how to program. They were very clear on that, that's how they had the highest quality information at the time.
Fundamentally, that was the most interesting thing to me, is that you could take a subject matter expert and teach them how to teach. So for us, the problem is those people who have real information, they're kind of by definition too busy living to learn how to win on the internet. If you have a real career, you don't have time to go be a hustler. You can't be a growth hacker, you're not going to learn SEO, you're not going to build a mailing list. You got something else going on for you, but you want to share it. Sort of normal recommendation algorithms are not built for that. They're built for attention seekers and people who are good at gaming them. A big part of what we did to make it work was pick a type of author that's subscription worthy and then build a recommendation and curation system that could find and promote them. And that's essentially what turned Medium around as a business. A year ago we were 600 and some thousand subscribers down from our high.
Now we're above a million, but huge amount of growth. All of it driven by quality we're proud of this time and where the incentives work out. The authors are happy, we're happy, the readers are happy, and we weren't getting that when we were doing pure editorial.

‍

Danny Crichton:
I think people don't realize unless you're in the content business of how much work it takes to promote your own work. Back in the newspaper era you were a writer. You just went to city hall, you covered city council, you turned in some notes. An editor would turn that into an article, it went onto a front page. The whole sales team got those into every bin on every street corner, the hustlers on all the subway stops trying to sell copies of the newspaper, whatever. And then you go into the internet era and at least among my friend group I always ask, particularly the successful ones, how much time do you spend? And the answer is usually like 20 to 30 hours a week. It's like four to five hours a day, building lists, promoting content. And a lot of that is not just like I just have to tweet on the internet.
It's also finding exactly how to tweet. Should I put the link in now? No, the link is not going to get shared anymore because they're deprioritizing tweets with links. There's always this constantly revolving door of tactics that you have to be up-to-date on. That is why the content is bad for a lot of these folks because they spend so much of their working time and their creative energies trying to get attention in the first case that you finally capture the attention. And what are you sending them to? What is that piece that you did all that work? And I tend to be the opposite, I'm a terrible social media person. I'm terrible at everything, but I put all the time into the work itself. And so I always say there's something great if you can get the attention, but there is no one driving folks into there.
So to me that's sort of the magic that people have been trying to do I would say over the last five to six years in media. Is how do you balance, hey, I just need growth. I need people to read. It's great. But get me those people into the front door. Who are those people? How do you make it happen? And you got to do it with technology because it hasn't really scaled I think on the business side from an economic perspective, meaning human to human as I got producer or a business development.

‍

Tony Stubblebine:
You touched on something earlier too, that's the other part of it. It's not just how much time do you spend promoting it, but in what ways did you change the writing knowing that you're going to have to promote it. So it's like I usually use the example of a recipe online, why are there 10 pages of storytelling before the ingredient list? It's because the person that wrote that for SEO, like Google is rewarding it so they added a bunch of stuff that you as the reader don't care about. I did some writing with Tim Ferriss once, which was really interesting. I got to see his method for deconstructing an article and I was like, look, he cared a lot about the instructions and the accuracy at some level. Then there is the parts that make it viral, and that's this whole other section. And I think if he didn't need it to be distributable, he wouldn't include that stuff.
As soon as I felt like my writing was the product rather than a vehicle for marketing something else, just started writing differently. I didn't even understand how much my own writing had been corrupted by the system I was in, and some companies have to build alternatives.

‍

Danny Crichton:
When you think about Medium's business model, one of the big questions you always have with creator platforms is how much should a creator's earnings come from that particular platform? Is it just a part of a mix of earnings? You're launching a YouTube channel, you get some Twitter subscription dollars, you get some from Medium, maybe a Patreon or something like that. Do you want to dominate those earnings and say, "Look, if you put your whole investment of energy into our platform, you're going to have a full quote unquote, "career freelance". But a durable set of revenues that you can sort of rely on, do great work and you'll be rewarded." What's sort of the model when you think about creating that kind of class of workers on the platform?

‍

Tony Stubblebine:
We've really backed away from the creator economy. And I sort of don't like to talk too much about it because I don't get a great joy out of bringing bad news to people, because the creator economy is filled with all of this hope that you could quit your job and live this creative life. But for the most part, I've found it to be a failed get rich quick scam. Even when it's working, it's working towards this local maxima that has this much bigger mountain right next door. So I feel like the creator economy lost track of the expert economy. Before the creator economy existed, there was this huge amount of writing that was paid for not with money, but by reputation and distribution. The people who make a living in creating kind of hate this like, oh, I get paid in exposure. But they're just wrong because there's a lot of people out there that get paid extremely well based off of exposure and reputation. And they're people that make some living based on expertise.
If you go back to my O'Reilly experience, the best case scenario you're going to make $40,000 back then a programming book, probably more like 10 or 20,000 bucks. So for the salaries of an engineer, basically nothing. Like a nice little bonus, you might buy yourself something but it's not going to change your life. But that book on your resume changes all your career prospects. By far the biggest earners on Medium are consultants and professionals who are using some small amount of writing as essentially a form of resume or portfolio. I just think there's something so much healthier about that because the incentives are aligned. Again, I need to do the research in order to be smart enough to write this piece that shows the world how smart I am, because that's going to generate new career opportunities for me.
I think almost everyone in the creator economy would be better off thinking that way because that's so much more lucrative and it's so much more fulfilling. Like turn yourself into an expert rather than a blowhard. We don't need more opinion. We need more real deep information from people who know what they're talking about. So to answer your question, we're not in the business of giving anyone a living. We pay our authors major portion of our subscription revenue as part of the trade. But I think the people that are most motivated probably also feel rewarded and motivated by other things.

‍

Danny Crichton:
Well, I think when I have seen other media companies build successful creator economy business models, I think of Forbes starting in around 2014, 2015, Expert Networks, LinkedIn building out a newsfeed. I follow a lot of people who do not post interesting stuff, but there are people who do and it's a very rewarding in that way that if you're in sales and you have interesting things to say about sales, you can post in LinkedIn. It works well, it cycles through. I think the challenge has been, look, if you're a professional in one of these categories, it's very hard to be original. You're a marketer, you're in SaaS, you're in sales, you're a consultant, you're in the healthcare industry. The reality is most of the stuff is not brilliance. It doesn't not require a massive originality. The challenge is that if you're a large company, you need 50 of these people and there's plenty of work to go around. But it's very hard to differentiate on writing.
And so I'm just curious if you in any way that you've sort of constructed it as a professional network, how do you deal with that kind of originality problem where it's not just, hey, here's five tips a day to improve your email deliverability, whatever the case may be? But how do you try to create a kind of a not winner takes all, but sort of a race to the top kind of model for editorial?

‍

Tony Stubblebine:
I would sort of push back on that, it's the... Well, I'll give you some venture capital examples because thinking about venture capitalists that have written on Medium. One of the examples I was using for a while was how to run a board meeting. I think there's literally 30 versions of this article on Medium, and Mark Suster has himself written 10 versions. So he is one of my favorites. On the one hand, one of them is the best but they all tell you something interesting about that investor. I wish every investor I ever had to interact with had their sort of portfolio of here's my manual for how I write up, because you get these little different nuances. I think people really underestimate, especially in this modern era where everyone is going to convenience, they underestimate the value of the context of a real story. Someone's whole life is kind of exposed in something that you might think is pretty trite, how to run a board meeting.
You get these little wrinkles that are different for each person, and actually it tells you a lot about the subject which matters to me and maybe not to some of your listeners. If you go to someone like Mark Suster, it's probably 90% of what he wrote on Medium, 10 other VCs have written the same thing. There's something that's unique about his version that helps you understand, hey, is Mark someone that I would want to work with? And then every now and then there's just the muscle of writing, one out of 10 things that you write actually is unique.

‍

Danny Crichton:
So one of the questions I have for you is obviously there's been a big question over the last couple of years around how much should publications be writer driven? So there's some new startups that are like this. You can also include Substack and Patreon as that model of the creator is the CEO platform that's designed to empower them. The reader relationship is to that writer, et cetera, versus the media model and I think some others in your market, which is like, look, the publication owns this. It's a basket of goods, so to speak. A reader is subscribing to Medium, they're getting access to a lot of different things. You don't have that relationship, but the benefit is that you have a lot more readers who are coming in through the front door who otherwise may not see your work, whatever the case may be. Where do you think 2024... A lot of things have changed and Substack I feel like had this parabola of tension to success. And now it feels like every other media company, although I'm not sure they would agree with that but certainly what I view it as.
How do you see the world in 2024 from that perspective? Is that rebalancing back towards the platform or is that sort of individual creator still kind of the linchpin for building great publication?

‍

Tony Stubblebine:
For us, I just really think of it mostly in terms of us and our target author. Our target author is not in the business of building an audience. It's almost that's an anti pattern. If they are, it probably might not reflect that well on the quality of the writing. We're looking for people living interesting lives and want to write about it. So I'd rather that they be out there living than hustling. Unless they're writing about hustling, in which case the hustling is living. For that person it's this huge benefit. The main reason to write on Medium is to get distribution. And there's a lot of ways in which if you write something great on Medium, we're actively looking to give you distribution even if you have no followers at all. Think about two different markets of authors. The market of authors who need distribution is one million times bigger than the market of authors that have the time to build an audience for themselves. So maybe you and I are in this elite club of people that are dedicated media personalities for the rest of our lives and it's worth-

‍

Danny Crichton:
Oh God, that sounds like total hell.

‍

Tony Stubblebine:
Yeah. You don't even like doing the distribution, but there's some argument that you should be, like you Danny should be or maybe me. But your average person, that's worth hearing from. Now platforms are really, really valuable and it's good to have platforms that have different distribution characteristics [inaudible 00:20:07], attention, rage bait, clickbait. It's good to have an alternative.

‍

Danny Crichton:
We started the conversation around incentives and algorithms have been in the news very heavily over the last two years. Instagram, TikTok obviously getting a lot of attention from Congress, a lot of legislation. California trying to pass laws. I think New York just passed a law that says we can't algorithmize beyond date for people under the age of 16. So there's no algorithmic choice when people are like, everything's an algorithm. And it's true, no one knows how to do this. When you think about Medium, how do you both create... You said you want to have an economy where no one has followers, can still get their stuff seen, but obviously having followers is one of the key signals that a lot of platforms use in order to prioritize content in the feed. So when you think about the algorithm which is running that front page of Medium, and I'm logged in and I'm seeing top articles about bird poop, what you don't know about bird poop. So I am intrigued by that. As a side note, Tony, I don't know if that was one of yours or someone on your team chose that.
But nonetheless, how do you sort of pick those articles and what sort of signals are you taking to create kind of a positive economy there and less of this, hey, I know all the growth hacks. I know how to optimize for the algorithm and know how to get ahead.

‍

Tony Stubblebine:
People are maybe less aware of this as a trend. So I just want to point this out, it's not just a Medium trend. Is that social media platforms have been fragmenting and people have been running behind human moderation walls. So Reddit would be the most obvious example, right? And Medium very much is the same. So if you have followers, that's one of the reasons that will recommend us. If I follow you, your stuff will be in the mix to be recommended to me. Everything else, we're using human moderation for it. On a given month, there's about 9,000 active editors on Medium who are curating and working with authors. Then we have a tier above that, which we call the Boost program. That group is nominating the best of the best to our internal curation team, and we're giving the best of the best an extra boost.
Essentially, we believe in human curation because you can get an expert signal into the algorithms that you wouldn't otherwise get. It's the way to escape the attention economy because otherwise you're just keying off of attention signals. And attention is very different from, is this satisfying? Is this going to make you a better entrepreneur or a better investor? What we were tending to get before we made this change is recommendations that would make you an angrier investor. And so a thing that we had to learn the hard way is there's a huge difference between what someone will click on and what they'll be happy to have paid to read. We can recommend you really upsetting hot takes on investment that make you angry and that you click on and read and you hate to read. But then at the end of the day you think, why on earth did I pay five dollars for this.

‍

Danny Crichton:
We've covered a lot of ground, Medium's around 12 years, famously has pivoted a couple of different times over the years. You came in two years ago, got into cash flow breakeven, building out sort of a new model around membership and what we just described with algorithms. Project me forward, we're in 2024, what's the next couple of years here for Medium? Is it just sort of continuing the same story over and over again or is this empowering you to begin the next phase of sort of Medium's evolution?

‍

Tony Stubblebine:
I think it's really nice to be in what I would call an early moment of, oh, this model works. It's we're sustainable, it's working for all the participants. But it's really not well known yet to the rest of the world that it's working. So that's why I say there's two turnarounds. The first turnaround was financial, the second is relevance. And I just think we have not even scratched the surface of pulling important stories out of people. If we can scale that across all topics, across all languages, you get this opportunity. This is the thing that excites me. I don't know if we can do it, but this thing that excites me is great library of human stories that's on par with a Wikipedia. So where Wikipedia might cover facts, we can cover understanding how to do something, why to do something.
And I just have a lot of faith that there's a really big market of people who place a high value on being accurately educated. I don't need it to be the majority of people. I just know that there is a market out there of people that are going to care about that, and I think it's a big opportunity to serve them.

‍

Danny Crichton:
Tony, you are way more optimistic about people. I agree with you, but certainly more optimistic than I am. All right, take care.

‍