Yesterday’s news about Medium’s umpteenth pivot, and a buyout plan for the existing editorial staff, has come bundled with the news that the buyout was offered because of the failed unionization drive that the editorial and engineering teams spearheaded over these last few months.

Somehow this fact is being pitched in both an upside and downside way. That is, it’s being insinuated that the buyout wouldn’t have been offered but for the threat of unionization (even though that has, for the moment, passed) and it’s also being implied that founder and CEO Ev Williams is somehow exacting retribution on those who supported the Medium union by radically restructuring the company to get rid of them.

I have a hard time believing either end of this narrative is accurate. First, the union doesn’t have any particular power or impact on the company for the moment. The drive was defeated by one vote. (And to be clear, there does appear to have been a rather concerted union-busting effort at the company.)

Second, Medium is a well funded company led by a wealthy man. I know — I worked there, as the founding editor of one of the very publications that seem likely to now undergo a radical change. Because it is driven by its need for top tier engineering talent, everyone at the company benefits from a perk-laden, Silicon Valley lifestyle. Whatever Williams is or isn’t as a business leader, he is not cruel, and I do not believe he would not throw out dozens of staff on their ear without a reasonable, in fact, in this case, very generous, severance plan.

As for retribution — this is what, Medium’s fourth or fifth pivot or iteration? While Williams and his leadership clearly fought against having a union, it’s simply an unsound argument that he would pivot the whole company based on fears of having to deal with a union. What he is doing is pivoting the company based on the rapidly evolving media landscape. Just because it's well funded does not mean it can go on forever without making money.

Having left Medium over a year ago, I have no inside knowledge of the thinking behind the strategic pivot. However, even though I was given the opportunity to start a new publication there, I have to agree that the publication model wasn’t working for Medium — not because I think publications are dead, but because the tension between Medium the publisher and Medium the platform was one the company never truly confronted.

The old model of a house of publications, like Time Inc., or Condé Nast, leaves much to be desired in the 21st century. But the one thing that did work for these once-great companies was that the creative efforts of each title were siloed — and quite internally competitive. They battled for top-tier contributors, poached each others’ best editors and executives, and pushed each other towards greatness, even if they were competing in very different segments of the marketplace. Their business staffs were just as competitive, if not moreso, as their editors were, fighting over sponsorships, events, and any other piece of the financial pie they could.

And for media companies with just one title, or one that was outsized compared to all others, like The Atlantic, the mission was even more clear — them against the world.

Medium’s tension was that as much as had been invested in the publications, it never truly behaved like a publishing house, because the center of gravity of the actual business was so heavily tilted towards the platform — no matter how many journalists it hired.

There’s a reasonable debate to be had whether publications and the platform they're on should be in tension with each other, one that I believe the company just had. And finally, the answer is: No they shouldn’t.

And despite best intentions to grow and fund those publications — they couldn’t, by design, outshine the massive community of writers, both unpaid and paid, celebrity and anonymous whose work generates the vast majority of traffic (and paid membership) on the site — and do so at a fraction of the cost. The publications were less like new magazines and more like super-verticals of Medium. They were very expensive, nicely designed topic pages that employed dozens of well compensated staff to do really great journalism — that essentially only inflicted pain on Medium’s bottom line.

There’s a reasonable debate to be had whether publications and the platform they're on should be in tension with each other, one that I believe the company just had. And finally, the answer is: No they shouldn’t.

I left Medium just as Substack was making a splash as a place for writers. But it was already perceived as a threat, and many of Medium’s product enhancements over the past year (again, ones I have no internal knowledge of) seem directly designed to combat it. Most notably, Medium created a newsletter product, but it also allowed (again) writers to have custom domains and fine-tune the presentation of their stories.

Medium is in a battle for supply — not of editors and other middle-office workers, but for writing talent, and content, which is what readers actually consume and pay for. Any link in the chain “behind” the finished article that faces the reader is a cost center, even if — especially if — it makes the content better. The better that process makes the finished product — better editing, copyediting, graphic design, art direction, display, etc — the more expensive the finished product becomes.

And that is the precise problem that media workers are facing, and why unions won’t save their jobs. It’s actually a two-prong problem. First is automation — it simply takes less people to put out a glossy publication than it used to, thanks to enhancements in software, CMSs, distribution — thanks to the internet, basically (and not coincidentally, the internet that Ev Williams helped build with Blogger).

Second is the disruptive innovation that Substack represents (but is not the end point of, at least I don’t think so). Consumers are showing media companies they don’t care about glossy displays and pretty fonts and the branding of a publication. They care about reading interesting stories. And it doesn’t take an office full of professional journalists to create interesting stories, anymore (maybe it never did). 

Substack built the bare minimum publishing suite, and then recruited entrepreneurial writers with big audiences, paying them generously, and giving them just enough of the tools they needed, with no fancy junk to distract from that mission. Importantly, they gave them tools from the get-go that made those writers invested in mutual success — something that is a bit more opaque in Medium's publishing system.

Not to bog down in theory, but Substack as a response to Medium is disruptive innovation at work — a Clay Christensen speciality and one we see everywhere, when we learn to look for it. The expensive, fancy thing gets displaced by the inferior, cheaper thing, because, even though it’s a “worse” product, it actually does enough of what customers want to fit their needs. 

But in this case the customer is not the reader, it’s the writer — and the product is the platform they choose to put their writing on. The fact that Substack has played into the star system of talent and personal branding is the icing on the cake. By getting out of the way of writers who want to write and offering guaranteed incomes and creative freedom to the ones who can draw paying customers, Substack did what Medium had set out to do — disrupt traditional publications. They just did it by ripping the guts out of publications all together. And they are stealing Medium’s supply — good writers. For the moment they have out-Medium’d Medium. And Medium had to respond.

This is how crappier, cheaper products beat more expensive products that have grown stale. It’s exactly what the Japanese auto industry did to Detroit’s Big Three in the 1970s — moved to where the market was going (rising energy costs meant Americans needed cheaper, smaller cars), and used profits from the massive demand for their then-inferior cars to fix issues that would unlock their next level of growth — better quality, and more and bigger models to compete with American brands, when energy prices dropped again.

This isn’t going to turn into a rant about how unions crippled the Big Three’s ability to innovate — no, greedy and lazy management did that. But unions weren’t able to turn back the tide and protect the jobs that were disappearing due to a fundamental shift in the economy. Nothing and no one could do that. The Medium union — a movement spearheaded by engineers, according to reports — wasn’t going to save the publication business model. And if it wasn’t going to save jobs or change the company, and Medium itself has offered generous and voluntary severance packages, it’s hard to understand why so many journalists are calling this an anti-union, union-busting, or illegal activity.

I am not anti-union, and not some management goon, though I’m sure I’ve already lost many people by touching the third rail of “are unions good?” at all. I’ve been in a union, worked alongside unions, been in management, and even sat on the management negotiating committee when a magazine I worked at was organized. There’s nothing inherently bad about the desires of workers who want to unionize. Their aims are often noble. I do believe one thing that gets overlooked is that unions are themselves a business — ones with paid staff, a product, expenses, revenues that are reinvested, growth targets, etc. Just because they have a different financial structure and use different terms for this does not mean that the people who run them don’t need to make money, get ahead, succeed, etc.

What unions can’t do is change the water that media businesses are swimming in. And that water is very murky right now. They can’t change the fact that readers choose to consume work from “star” writers because they are very good at creating interesting content and need less of the support and infrastructure that traditional publications, and now, even Medium, offer. They can’t change the fact that the media business model is broken, and that media companies, whether 10 years old, or 110 years old, need to innovate to survive.

As individual stars flourish and haul in big (for journalists) paydays on Substack (and even on Medium) they may decide they need to hire editors and designers to increase the quality of their work — and that may be something that Substack offers them, and that Medium does too. But let’s be clear — that kind of work is commoditized. 

As great and necessary as an editor may be, they are the workers on the factory floor putting the car together. The job can be very skilled, take a lot of training, be really hard, and leave a worker both exhausted and proud of their efforts. And some workers will be far better the task than others. But it is never going to be the reason people buy the car. If they have that job, they should be paid fairly. But there’s no reason that job should exist when customers have shown they don’t care about it.

As great as an editor may be, they are the workers on the factory floor putting the car together.

This isn’t a defense of union busting, layoffs, pivots, or Medium’s increasingly long history of indecisiveness about its core product and competency. It is, however, a defense of a business’s need to innovate to compete, survive, and avoid being destroyed by forces outside of its control. And as knowledge workers who increasingly are being subject to the same pressures, we journalists have to do the same.

Medium is retooling their assembly line to try to compete with a cheaper, crappier upstart, before its lunch gets eaten. If it does it right, that means many more jobs — different jobs, different kinds of editing, different kinds of opportunity, but opportunity nonetheless — for those who have made journalism and writing their career. Medium isn’t to blame for that — consumers are. And there was nothing a union was going to do to change that fact.

About the Data:

Thinknum tracks companies using the information they post online, jobs, social and web traffic, product sales, and app ratings, and creates data sets that measure factors like hiring, revenue, and foot traffic. Data sets may not be fully comprehensive (they only account for what is available on the web), but they can be used to gauge performance factors like staffing and sales.