Did Medium Succeed?

Yesterday Ev Williams stepped aside as the CEO of Medium, ten years after he founded the company.

Pundits’ reaction to this news ranged from muted to harsh. The media narrative around Medium turned bad over the past decade as the company repeatedly changed strategies, often impacting employees and users along the way.

This week we may have seen the climax of that narrative—in the form of the most brutal headline I have read in quite a long time:

“Ev Williams gives up” is how Casey Newton put it, thereby renouncing his right to ever go back to traditional media and quit writing his Substack newsletter 😅

Casey’s post is a classic retelling of the conventional version of the Medium story: they pivoted too much and screwed over writers, they raised too much money, and they failed to achieve their overly-ambitious goal of fixing the media. This story gets a lot right, but the full truth is more interesting and complicated. It’s worth taking seriously by anyone who wants to know how traditional publishers and open platforms collide, why so many well-intentioned efforts to “save media” seem to end in disappointment, and what comes next for writers and readers on the internet.

I don’t have any inside information on Medium, but I have been a close observer and user since the very beginning, and I was an early employee of their main rival, Substack. So I figured now is a good occasion to write up the Medium story from my point of view.

To begin, let’s go back to the beginning.

Medium was created to solve discovery

In August of 2012 Ev published a post introducing Medium to the world. In it, he pitched writers three reasons to use Medium: 

1. Focused writing experience
2. Collaboration tools (e.g. Google Docs style comments)
3. Distribution of your writing to new readers via curation and algorithms

The first reason, the writing UX, is a classic example of a product wedge. Come for the tool, stay for the network. Even if nobody you know uses Medium, you might want to publish there for the focused writing experience and overall high-quality design.

But this feature alone would be easy to copy. Which is where the next two selling points come in. By building a commenting system, they created a reason for writers to share Medium drafts with friends, pulling them into the network. And by offering distribution to new readers, it creates a potential network effect, where more readers lead to more writers, and more writers lead to more readers. 

To many writers, Medium felt back then kind of like TikTok and YouTube probably felt to video creators when they were getting started. It’s hard to overstate just how exciting and new this was in 2012. I still remember how pumped I was when I managed to get a beta invite. My first post was seen by thousands of people. I had no idea how or why that happened (I think it got featured?) but I loved it.

No way was I going back to Wordpress.

It worked!*

Over the next few years, millions of people read stories on Medium, published by hundreds of thousands of writers. They were able to attract pretty much every famous person and organization imaginable, up to and including the President of the United States.

I think a lot of this was due to one main reason, which wasn’t explicit in the original pitch: you could publish a post without starting a blog.

This made Medium the ideal place for people who had something to say but didn’t want to create an ongoing commitment. Great! There was just one problem (hence the asterisk in this section’s title): this feature was also where Medium got the reputation as the “PR dumping ground” of the internet. Anytime anybody wanted to promote anything, they’d post to Medium about it.

Lots of people posted to Medium and tweeted a link, or saw a link and clicked, but far fewer people followed writers or turned to Medium’s feed to decide what to read. Of course there were still hundreds of thousands, maybe millions of people who did this. But the growth of that type of user was much slower than the big social networks, and the amount of time they spent each week “engaging with content” was much lower on average, I am fairly certain. So it probably wasn’t looking to Ev and the team like they could aggregate enough attention to just slap on an advertising network and call it a business.

To make matters worse, there started to be growing signs of a quality problem. This isn’t an issue unique to Medium, but low-brow writing tended to perform frustratingly well. The same way people groan about cringey self-help threads on Twitter now, people griped about self-help Medium posts back then.

According to an article published in 2015, the top ten Medium posts of all time were:

1. Welcome to Medium
2. The 37 Best Websites To Learn Something New
3. How quitting my corporate job for my startup dream f*cked my life up
4. 8 Things Every Person Should Do Before 8 A.M.
5. 7 Things You Need To Stop Doing To Be More Productive, Backed By Science
6. 33 Websites That Will Make You a Genius
7. HOW TO LOSE WEIGHT IN 4 EASY STEPS
8. Advice from 30 year old me to 20 year old me
9. The Crossroads of Should and Must
10. 7 Rejections

Not exactly George Saunders or Anne Applebaum!

(Side note: I would love to read a George Sauders short story about a megalomaniac internet self-help writer who is addicted to making the numbers go up.) 

Anyway, to “elevate” the average sophistication of writing on their platform, Medium hired journalists and authors to write for them, and even had a strategy to get people to create entire professional magazine-like publications on their platform. Some of the writing that came out of this was truly incredible, but it also cost a lot of money, which was quickly becoming an issue.

I think we all know what happened next.

They pivoted to subscriptions

In 2017 Medium shuttered their advertising business and shifted their strategy to focus exclusively on subscriptions. This was a move that turned out to be fairly prescient, and I don’t think Medium gets enough credit for it.

If they had continued down the path of advertising, then all the clickbait and low-quality content they were struggling with would have a fairly powerful incentive to flourish on the platform. By shifting to subscriptions, the theory was that it would decrease the rewards for publishing a large quantity of shallow content, and increase the rewards for publishing a more manageable volume of deeply interesting content that could reach for a wider audience and begin to encroach on the territory of book publishers and magazines. 

The goal was to kick off a Netflix-like virtuous cycle where subscription revenue could fund new writing from world-class authors and journalists, which would create more subscription revenue, etc, etc. This sounds like a pretty good flywheel in theory, the challenge is getting it to work in practice. God in the details, and all that.

The way their paywall worked was unique. Anyone could still publish anything on Medium for free, but optionally you could put your story behind their metered paywall and get paid. How much you got paid was determined by how many times people clicked on a “clap” button, amongst other variables.

This system had the advantage of being low-commitment and simple. You didn’t have to launch anything, craft a coherent value proposition, or even ask anyone to pay you. Medium handled all that for you, and all you had to do was check a box. This ease of getting started was reminiscent of what made blogging on Medium take off in the first place: in the same way that you could publish a single post without having to start a blog, you could also paywall a single post without having to launch a subscription media product.

The problem was: what kind of writers do you attract by building the easiest way to make money from writing on the internet, rather than the best way? Furthermore, the value proposition for readers was always murky. What kind of stories do you get when you become a subscriber? Why are they worth paying for? Who’s it for?

The answer then (and now) seems to be: all types of stories, for anyone and everyone.

This kind of broad value proposition usually doesn’t work, especially for something as cheap to produce and costly to consume as writing. But it seems like Medium actually built a pretty good subscription business out of this! Not amazing, maybe—especially if you factor in their reported $600m valuation—but way better than most people realize.

It might be surprising to some, but I would estimate Medium actually earns more revenue than Substack does, although they likely are not growing as fast. In early 2021 Medium said they had 725k paid subscribers, which equates to roughly $43m, of which 50% is reportedly paid out to writers, leaving over $21m in revenue for the company. Substack is said to have earned $9m in 2021. So even if my estimates are off, that’s a pretty healthy margin of safety.

There are two key drivers of Medium’s subscription business:

1. Huge top of funnel. They have a long tail of articles produced over nearly a decade of operating that show up consistently in Google and elsewhere, giving them an estimated 170m monthly users.
2. Superior take rate. Medium can keep 50%, unlike Substack’s 10%, because they drive discovery for writers. All you have to do on Medium is press publish and you have a chance of earning money. On Substack you bring your own audience, giving writers greater power to leave, and therefore greater pricing power.

There is a downside to this wonderful take rate, though, and it is the quality of writers that are willing to throw their weight behind Medium’s subscription. The most valuable writers seem to mostly be choosing to publish either through a traditional publication, or a platform like Substack or something similar like Ghost, ConvertKit, etc. Turns out, creating the easiest way to make money from writing online doesn’t necessarily mean you get the best writing. As with most types of media, most of the value in writing comes from the hits, and Medium doesn’t seem to have them.

So now it seems the flywheel they dreamed of—where Medium uses subscription revenue to attract great writers, and great writers to attract subscription revenue—isn’t accelerating much anymore.

Where do all the good writers go?

The narrative is that Substack ate Medium’s lunch, but I don’t see it that way yet. The vast majority of the revenue that gets generated by all forms of writing still gets captured by the top publishing companies: newspapers, magazines, book publishers, etc. The New York Times is thriving. So are Bloomberg and many other subscription publications. Book publishers are holding steady—their authors mostly haven’t fled to go direct on Amazon or anything. Even Substack, which has a lot more momentum than Medium, seems to be attracting fewer new big-name authors now than they did a few years ago. (I could be off, they don’t report a metric as detailed as “revenue from new writer launches” or anything like that—this is just my perception and I’m happy to let the data correct me.)

We are now years into the “creator economy” hype. If we’re comparing the shape of the curve to the adoption of PCs, the internet, smartphones, etc, it would probably not look nearly as interesting.

Why?

Of course, this could be a temporary lull before one of these online platforms figures out the next unlock, similar to how Substack figured out Substack Pro back in 2020, which helped them reach a new level of writer that would never have considered going independent before. But still, the current plateau is kind of weird. You’d think that more writers would go somewhere they could make the most money and have the most control. Some do. But not the majority.

Why? Independence is not the only thing that matters to writers.

Here are a few examples:

1. Access. It’s harder to reach out to random prominent people and say you write on Medium or Substack and have them spend time talking to you for a story. To be clear it is not impossible, but you need to be quite successful and have built your own brand in order for it to work. Compare that to writing for an established publication, even new writers can enjoy some degree of baseline access that far exceeds what they could get on their own.
2. Risk capital. When you get a job at a newspaper or magazine, they pay you a salary. When you sign a book deal, you get an advance. If your writing ends up being popular, you get leverage to ask for more money down the line. But what if your writing isn’t that popular at first? This is unfortunately the most common case. If you’re on your own on the internet, you’d have to do something other than writing to make a living. But when you have a publisher, you get to keep going a bit longer. Publishers are like venture capitalists for writers. They make early bets and give people a chance.
3. Prestige. Maybe this will change in the future, but it still means something to a lot of people to be published by an established institution. Seeing your subscription newsletter’s revenue chart grow is nice, but it’s also nice to see your book in a bookstore, or to see your name on the front page of the New York Times. It’s a lifelong aspiration for many writers that is hard to erase—even if some may consider it irrational.
4. Support. Creating an independent writing business on the internet is about a lot more than just writing. When you work for a traditional publisher, they handle everything for you—graphics, fact checking, copy editing, legal, liability insurance, etc. Not to mention social media, paid user acquisition, events and conferences, swag, etc. The list is long. 

And this is leaving aside publishers’ most important current advantage: distribution to an audience. 

I left this out intentionally because both Medium and Substack provide it, and theoretically an online platform could provide even more distribution than a traditional publication could. But really distribution should be a 5th reason in the list above, because traditional publishers currently still beat online platforms at this, at least for writing. (The same can’t be said for video.)

Sacrificing some independence to work with a traditional publisher obviously has its downsides, but just based on the fact that more writers haven’t fled by now, it’s clear that the downsides are outweighed by the upsides to most writers.

What’s next for internet publishing?

Being able to compete with traditional publishers on some of these value propositions is the most important thing to look for. Substack has expressed interest in offering a lot of these types of services to writers through their “Pro” offering, but it’s still young, as far as I can tell. I will be very curious to see whether Medium tries to go this route.

There are also new experiments like my company, Every, which tries to blend the upside and control of writing independently with the distribution and support of a traditional publisher. But we are still tiny.

Meanwhile, Substack shows signs of developing a type of network effect that doesn’t exactly rely on algorithmic recommendation of content. As I’ve written before, they have an ideology that is at tension with their commercial incentives, and it will be fascinating to watch them navigate it. But I think there’s a real possibility they will be able to create the first true YouTube-esque network for writing. Especially if they open their platform and APIs so that traditional publishers decide it’s not worth it to live on their own islands anymore. Perhaps the same way everyone, even large media companies, publish video on YouTube rather than on their own site, maybe writing could live in a central walled garden someday.

(Is this a good thing? That’s a separate question, but probably not.) 

So, has Medium succeeded?

The question is: succeeded at what? There was a lot of lofty rhetoric in the early days about fixing the attention systems that feel so dysfunctional on the internet, and I think it’s clear those systems are going as strong as ever. But was it ever reasonable to expect otherwise?

Medium raised a lot of money in its heady early period of rapid growth, and it will probably be hard for those investors to ever feel like Medium was a grand slam financially. But that’s just how it goes a lot of times—especially if you’re doing anything related to writing. I think those investors probably all knew that going in. What they expect of entrepreneurs is not guaranteed success, but courage and skillful effort. And from what I can tell by the public tweets, they all seem to be pretty happy to have supported Ev and the team.

The one thing nobody can say is that over the past 10 years Medium didn’t try hard to get the model right. You could tell a story about the pivots that blames indecision or waffling, but I see it another way. I think it’s a story about trying to do something really hard, holding yourself to the highest standard imaginable, and refusing to give up and let the thing just coast even when it was hard.

I’m sure a lot of the employees and customers who were negatively affected by the pivots might find that argument disagreeable. But what else can we ask of entrepreneurs? Do we really want to live in a culture where people are afraid to try new things or change directions because they’ll be shamed? What is the reasonable expectation that customers and employees can have of startups they choose to work with? If you’re joining a startup, you have to know there is a good chance the company could fail or shift directions in ways that make your services no longer necessary. If you decide to use a startup’s software, you obviously also are taking some risk. If that risk isn’t appealing to you, fine! Work at Google! Use Wordpress!

Of course I don’t know all the details here. I’m sure there are components of the pivots that could have been handled better. I’m not here to defend everything Ev or Medium ever did. What I am here to argue is that it’s more complicated and interesting than just “they pivoted too much and screwed over people and so they failed.”

It’s easy to be snarky about how the story is over, and probably in all likelihood Medium will never grow as fast or attract as much hype as they did in the past. But it’s not a 100% done deal, and I think it’s a good thing that there is a new CEO at the helm with a new vision and fresh energy. I’m still rooting for them.

Thanks to Rupam Grimoeuvre for supplying the original artwork for this week’s post!