Content used to be king. Well, the king’s dead.
If you work in content marketing, congratulations on turning a bit of talent, a lot of luck, and a really expensive liberal arts degree into what passes for a career.
After all, student loans for private liberal arts colleges are pretty steep these days, and when finance majors and MBA grads are struggling to find jobs, it’s not like the Fortune 500 are competing to hire all the top comparative lit and philosophy majors out there.
Late-stage capitalism sucks, mostly. It’s gotten you trapped in a crushing cycle of debts, interest, and fees (hey, you played by their rules long enough, which is why you know more about Moliere than macroeconomics).
There’s one silver lining to it, though- without the collision of classism, consumerism, and capital markets, content marketing wouldn’t exist, and you wouldn’t be a decade into your Plan B of writing copy about crap you don’t care about for an audience you secretly loathe. You’d be writing the Great American Novel instead of, well, B2B newsletters or LinkedIn posts.
Now that you know what my internal monologue sounds like, please also know that I’m in the same boat as you – and while the medium sucks, and the message is worse (basically, “buy stuff from us” is a leitmotif in this business), we’re pretty lucky to have a job to begin with – especially right now.
It’s no secret that the content marketing function has recently been put under an unprecedented stress test that was, by most objective measures, a complete structural, spiritual, and systematic dismantling over the past two years.
Put less delicately, it’s been a total shit show ever since the world first saw generative AI and how “easy” content creation has become.
Because those of us who managed to survive as content marketers know that it’s actually really, really hard if you care about stuff like quality, or clarity, or extraneous emojis and em dashes.
And it’s getting harder every time an LLM throws out a new model, or another product promises that it’s capable of the alchemy that is turning mediocre prompts into memorable prose.
Not because AI is a threat – rather, it’s sort of the only moat left for content marketing careers. It’s largely because our audience believes the hype about AI that the bottom of the content marketing job market has fallen out, the illusion that productivity is more important than creativity, and the belief that emotional connections aren’t required in content marketing to create meaningful engagement.
That parasociopathy is the root cause for the fact that so much traffic, and so many jobs, have been sacrificed to an LLM that doesn’t need to cite its sources, fact-check, or rely on rewrites or redlines to turn a rough draft into a polished product.
Content marketing has always been about quantity over quality. It’s just that not too long ago, those two concepts could peacefully coexist before all that quantity created so much noise that no one even bothers listening to the signals anymore. Reach has become relevance, which is bad news for those of us in the business of content marketing.
The good news is that content is more important than ever before. They aren’t wrong; it’s just that the role of content – and that of the marketers who create it – has completely changed.
Which brings us to this week’s Mediabistro Weekly Drop. This week, we’re breaking down the state of content marketing jobs and looking at short-term trends and long-term career implications of what happens when you replace writers and designers with prompt engineers and project managers.
Spoiler alert — 💩💩💩
The Lead:
This week, SEMrush released a fascinating deep dive into content marketing job data, analyzing 8,000 US content marketing-related job postings from November and December 2025.
The results read less like workforce analytics and more like an obituary for content marketing professionals, which figures, given that it’s an Amazon subsidiary reporting on a data set largely developed from paid job ads indexed from Indeed. Yeah, we hate it here, too.
According to the report, job postings for Content Marketing Managers dropped by 73% from 2023; in those two years, postings for Content Marketing Specialists fell by 74%. That’s not a rounding error – that’s a body count, considering these are by far the two most prevalent job titles in the content marketing profession.
It also explains why we’re stuck writing weekly SubStacks for a living (and feel lucky to do so): there really aren’t many options for anyone who’s built a career (or something closely resembling one) in this business.
These jobs have been replaced by two roles that, on the surface, seem to exist at opposite ends of the content marketing spectrum.
At the tactical, execution, and operational end, listings for “Content Producers” jumped by a staggering 1,261% in the past 48 months, which makes it the professional equivalent of buying stock in NVidia or buying up Bitcoins. “Content Creator” jobs, similarly, rose 410% in the same time period.
These two closely correlated roles accounted for 34% (or about one out of three) of the total number of content marketing job posts analyzed in the report, a significant plurality that’s only continuing to grow, albeit at a slightly slower rate.
At the other end of the content careers spectrum, postings for “Head of Content Marketing” grew by 376%, and “VP of Content” (and equivalent titles) rose by 308%. This is reason for cautious optimism, but the growth in these divergent roles- at the expense of most other content marketing jobs – can also be interpreted a bit more cynically.
Companies want people who will do the work and people who will own the strategy. They aren’t willing to pay for anyone who exists anywhere in the middle, which is bad news for anyone looking at content marketing as a long-term, viable career choice. These experienced, mostly mid-career sort of roles are unlikely to be saved by the bell curve, even one that’s so patently problematic.
Another interesting finding, for anyone working in media or entertainment – beyond content marketing, across industries and markets, the number of senior and executive level postings (defined as director level or above) listing “storytelling” or an equivalent soft skill as a basic job requirement rose from 8% in 2024 to 29% today.
By comparison, that’s one percent higher than executive roles listing AI expertise or experience, which is great news for anyone in this industry – for now.
The moral of the story: content is no longer considered brand marketing, but increasingly, as operational infrastructure. Content marketing is entering its second act, and with one helluva plot twist, apparently.
You can read the full study at http://www.semrush.com/blog/content-marketing-job-market-study
Search Media & Creative Jobs on MB
Press releases are like participation trophies for corporations. No one asked for them, nobody really wants them, but they’re still an integral part of content marketing and corporate communications strategies – even though those same execs all pretty much know damned well that when they want to cover a story, the press actually reaches out to you directly.
They are, like most AI-generated content, ridiculously formulaic and mid, for lack of a better word. They all open with some variation on the fact that they’re “thrilled to announce” whatever the headline already has, which is corporate speak for “you can applaud now.”
Then, they go back to ignoring corporate press releases, just like everyone else.
Like Harlequin is to bodice-ripping paperback romances, “news” wire service PRNEWS is to this most generic of genres. For a nominal fee, they’ll let corporate comms teams upload and format their own press release, then “distribute it” to hundreds of outlets (which is technically true of an RSS feed).
Their business model is predicated on press releases somehow persisting, even in the face of budget cuts, media distrust, and the general collapse of attention spans or intellectual curiosity. They’re like cockroaches or Toyota trucks, but with boilerplates.
PRNEWS (the all-caps naming convention being very on-brand), likely realizing that their entire business is predicated on perpetuating the myth that press releases are not only a best practice but also carry some sort of cachet and prestige, has invented a similarly specious industry awards program. If the Cable Ace Awards or Daytime Emmys taught us anything, it’s that awards provide market validation that even the most cringeworthy content is worthy of recognition and praise.
So, on February 19, PRNews announced the 2026 winners of its annual Content Marketing Awards, which are like the other CMAs but somehow even less relevant to enterprise businesses. This version of the CMAs celebrates the fact that PR pros invented content marketing, and that entitles them to be independent arbiters of what quality content looks like.
The awards this year included categories such as branded podcasts (why is this a thing), “thought leadership campaigns” (definitely not a thing), AI-powered storytelling (God help us all), and “owned media platforms” (like Pravda in the former Soviet Union). The winners this year recognize some genuinely impressive accomplishments, all for the low, low entry fee of $550 per category.
For example, rbb communications (the ee cummings of brand naming conventions) took home two of the night’s top honors: Most Innovative Use of AI in Content Creation, with another win for “Best Evergreen Content Strategy.” Let’s hit pause on that, uh, interesting combination. Evergreen content and AI innovation are diametrically opposed concepts; one lives or dies on its longevity and utility, and the other completely ignores these factors while converging all content toward the median.
Winning both awards is a clear sign that content marketing success doesn’t take talent or require quality. It requires cash, connections, and the ability to keep a straight face. Here’s hoping they put the award next to their “Best Places to Work” plaque – they’re equally reputable, objective, and definitely not pay for play.
Check out the full list of winners here for inspiration (or proof that content marketing’s collapse continues unabated).
In mid-February, LinkedIn publicly disclosed that it had seen a nearly 60% decline in traffic over the last year, largely due to the rise of AI-powered search. This news was released at about the same time that many marketing agencies and analysts issued public guidance to double down on LinkedIn spend, particularly given TikTok’s imminent demise and, well, the decline in search engine volume, which is a real drag on sponsored results.
Man, you gotta love it when irony does its thing.
To be fair, this isn’t a LinkedIn-specific problem. It’s what happens when queries that used to drive clicks to websites now move to platforms built on AI summaries and preview panels that don’t require clicking any external links. News publishers have seen referral traffic drop by roughly a third since Google first launched AI Overviews, according to Chartbeat data of over 2500 major news outlets.
That’s bad, but LinkedIn’s 60% is notably way worse, probably because, unlike actual news sites, its content is already the same sort of generic copy stuffed with listicles and broken backlinks that AI now produces itself, without requiring users to leave their platform at all, much less for the world’s largest “professional network.” That’s what they call spam factories these days.
This doesn’t mean that LinkedIn is dead, as much as we’d all love to see that happen. It means that in content marketing, reach and traffic are no longer concentric concepts – they’re increasingly decoupled, and completely unattributable as far as source tracking is concerned.
Metrics built around click-through rates are now even less meaningful than ever before, since what they’re measuring is less meaningful than ever before, too.
Whether that’s good or bad news depends entirely on whether or not your content marketing can consist exclusively of AI-generated summaries and aggregated social media posts, which is a convenient segue into the next story.
Even in proposal form, these new requirements would establish a clear precedent that other regulators will surely notice, and that requires a public response – and public stand – by Google, whose Gemini LLM is quickly lapping most of its competitors in both sophistication and user volume.
For content marketers, the immediate implications seem clear: if brands can elect to meaningfully opt out, the value of continuing to voluntarily provide LLMs a high-quality source should significantly increase, as the inventory of available content that can be leveraged by AI will shrink, while the demand for AI-generated answer engines will continue its unabated growth.
Being the brand that AI wants to cite will be the ultimate competitive advantage in content marketing – provided that most companies, predictably, decide that proprietary IP is more important than perceived authority and brand reach.
Long term, though, the implications are much more uncomfortable for content marketers. The entire profession is built on a 20-year-old business model that assumes good content inevitably leads to increased traffic.
Just like 20 years ago, when everyone assumed that home values never went down, that R Kelly was a really nice guy, and the Lions would have won a Super Bowl by now. But now we know better: AI Overviews breaks that virtuous cycle by breaking the entire logic behind links. So, what replaces this familiar but terminally ill business model?
Hell, at this point, no one knows. Not even Grok, Claude, or Clippy.
But when models shift, content marketers tend to figure it out quickly. After all, survival is the most important soft skill you can have in this business. For example…
According to Adweek data published this month, YouTube has overtaken Reddit as the most frequently cited social platform in AI-generated content. The one-time P2P video sharing service now accounts for 16 percent of LLM citations during the last 6 months; Reddit, the Internet’s self-proclaimed frontpage, only accounted for around 10%. That’s a pretty big reversal from where things stood just a year ago.
So, what’s going on? Well, the result is more structural than anything to do with machine learning or consumer preferences. It’s because YouTube content is loaded with automated enhancements like transcripts, captions, timestamps, and keyword-rich descriptions.
That structure makes it easy for AI engines to parse, understand, and reference – exactly what you’d expect from a Google subsidiary that’s definitely not evil (trust them). Reddit, by contrast, tends to have content that’s messy, user-generated, community-moderated, and more reliant on context than the content itself. AI models, obviously, strongly prefer the former approach when it comes to training data, creating the emerging discipline (and impending Silicon Valley gold rush) that’s called Generative Engine Optimization, or GEO.
Because if there’s one naming convention that instantly conjures up images of quality and permanence, it’s “Geo.” This emerging discipline (and cottage industry of commoditized crap) is basically where SEO was back in the dot-com bull market. Everyone knows it’s important; no one knows how it really works, but everyone’s willing to throw money at the problem rather than risk falling behind the competition.
Recently published data suggests that content that’s deliberately optimized for AI citation is cited 43% more in AI Insight and LLM results than content that’s just optimized for traditional, boring old SEO.
So, if you’re one of those people, like me, who prefer to stay off camera and believe the power of the written word is stronger than the powers of post-production and after effects, well, it’s finally time to trade in our bylines for hyphenites.
YouTube – and video content in general – is no longer optional, or nice to have just because you’re in B2B marketing. The days of professional audiences preferring long-form content are long gone – the case for skipping a cross-platform content approach is getting harder to make every day. Your ICP may not be watching your videos; your buyers likely don’t like, subscribe to, or comment on your YouTube channel.
On the other hand, the AI models responsible for creating the summaries that now form the foundation of how knowledge workers acquire that knowledge are constantly tuned into your channel. And if you’re not tuned into that fact, you’re probably going to find yourself out of work.
Or at least, out of an audience.
That’s a ton of content about content, and almost all of it’s as uplifting and optimistic as your standard 18th century Russian novel. But here’s the bottom line: content marketing is splitting itself down the seams, dividing its workforce into two distinct camps. There are those people who can produce content on time and at scale, and those people who oversee content marketing strategy, distribution, and monetization.
The middle is no longer content marketing managers; rather, these often antagonistic (and codependent) polarities are connected by AI and algorithms; in fairness, this is one AI use case with promising early returns.
At the same time, the platforms that content marketers used to rely on to distribute content to targeted audiences and qualified decision makers are the same platforms that are losing users and search volume to black box systems that don’t care about protecting IP, enforcing objectivity or even reasonable accuracy (thanks, Grok) – and doing so without any attribution or referral revenue, as an added kick in the content marketer’s ass.
The skills that mattered most in content marketing for most of the past few decades have seen an empirical collapse in employer demand over only a couple of quarters; at the same time, regulators are now seriously scrutinizing what the implications might be for AI companies whose products are predicated entirely on aggregating other people’s work. We used to call it plagiarism, but today, it’s just the cost of doing business in the information economy.
The content marketers who will have a place in the new normal will be the ones who never approached content as a product, or publishing and distribution as “output.” They think about it as a means to an end, and see content as part of a system rather than something created in a silo and optimized for search engines.
If you know enough about analytics to be able to add quantitative evidence to that content and deliver data points instead of vibes, you’re probably going to have a very lucrative career in whatever this next iteration of content marketing is. If you can tell a story with data and know enough about AI to understand its impacts on traffic, conversions, and ultimately, revenue, then you’ll always be in demand.
If you’re more concerned about driving onsite engagement for your LinkedIn posts than understanding why LinkedIn posts are suddenly driving 60% less conversions than they did a year ago, or if you’d pick your byline over business outcomes or brand equity, well, it’s probably time for you to find some other way to put that BFA or liberal arts degree to use.
The good news: Hourly hiring demand has never been higher, so unlike content marketing managers, you should have plenty of options. Most are likely more lucrative, too.
Until next month, keep your heads up. The people who figure out what content marketing becomes next will write the playbook that everyone else copies. Might as well be you.
Matt Charney
Executive Editor, Mediabistro