Right now, in newsrooms and marketing departments and university communications offices, the same decision is being made over and over with the same confidence: we need to be doing vertical video.

Reels, TikTok, Shorts. Someone senior has seen the engagement numbers, watched a competitor's feed fill with nine-by-sixteen clips, and concluded that the organization needs a short-form video strategy, usually with a hire and a budget attached. The trend reports agree. The Reuters Institute's 2026 outlook found three-quarters of publishers planning to make their staff "behave more like creators," and named efforts from the Daily Mail's sixty-person creator unit to the New York Times' new vertical video tab as the shape of the moment. The feeds reward the format, the whole industry is moving, and standing still feels like malpractice.

I want to suggest that a great many of these initiatives are about to fail, and that they will fail for a reason nobody in the meeting is naming, because the sentence that launches them has no object. A vertical video strategy for whom? Serving what? Toward which outcome, measured how? The sentence sounds like a plan and it is actually an anxiety, the feeling that everyone else is doing something we are not, dressed in the vocabulary of decision-making. And we know how it ends, because we have watched this exact movie before, at enormous cost, with a slightly different aspect ratio.

We have done this before

Nine years ago the industry ran this experiment at full scale and published the results, mostly as layoff announcements. In the mid-2010s Facebook told publishers that video was the platform's future and backed the message with the most persuasive evidence there is, engagement metrics, and newsrooms responded rationally to the data they were shown: writers and editors out, video teams in. Mic, Vice, MTV News, Fox Sports, Vocativ, Bleacher Report, and Mashable all cut editorial staff to chase the format, and the phrase "pivot to video" became so tightly bound to job loss that it turned into gallows humor, a euphemism for layoffs.

Then the floor gave out, in two stages. First the audience reality: Digiday reported, using comScore data, that publishers who pivoted saw traffic fall roughly 60 percent year over year in August 2017. The audiences these outlets had spent years building were largely readers, and the pivot had, in effect, fired the people who served them in order to chase people who did not exist in the numbers promised. Then the revelation about the numbers themselves: Facebook had overstated average video viewing time by 60 to 80 percent, and a later advertiser lawsuit alleged the real inflation ran into the hundreds of percent. Facebook settled the suit for 40 million dollars while denying wrongdoing. The single metric most responsible for convincing publishers to gut their text operations had been inflated the entire time.

The comfortable lesson to draw is that Facebook lied, and Facebook did. But that lesson lets us off too easily, because it implies that with honest numbers the pivot would have been sound, and it would not have been. The pivot to video was a format, mandated by a channel, installed at the top of thousands of strategy documents with no product beneath it and no reader job named anywhere in the chain. Nobody restructuring a newsroom in 2016 could tell you which specific reader need the new video would serve better than the text it replaced, because the strategy had not been built from a reader need. It had been built from a platform's appetite, and when the appetite changed, as appetites always do, there was nothing underneath the work to hold it up. The bet did not fail because video was wrong. It failed because it was placed on the wrong layer, and it would have failed on any format placed there. Which is the uncomfortable thing to notice about the vertical video initiatives being greenlit this quarter: the aspect ratio is new, the structure of the mistake is not.

To see why, you have to name the structure everyone keeps skipping.

The hierarchy nobody names

Any organization that communicates, which at this point is every organization, is operating within a four-layer structure whether its leaders know it or not.

At the top of the structure is a person with something she is trying to accomplish. Product strategists, following Clayton Christensen's jobs-to-be-done framework, call this the job: the progress someone is trying to make in her life, the thing she would hire a product to do for her. A commuter wants to arrive at work already knowing what happened yesterday. A new employee wants to get fluent in an unfamiliar field before anyone notices how lost he is. A patient wants to understand a diagnosis well enough to make a decision she will not regret. In my own work I define this top layer with TCI, Task, Context, and Intent, because the task alone tells you what to build, while context and intent tell you which form it should take and when it needs to arrive, which happens to be exactly the question the rest of this essay is about.

One layer down is the product: the thing an organization builds to do that job. A morning briefing. An onboarding course. A plain-language guide to a procedure. Products are where organizational effort should concentrate, because products are the layer where a stranger's need and an institution's capability actually meet.

One layer below that is the format: the material a product is expressed in. Text, video, audio, a live room, a data visualization, an illustrated explainer. And at the bottom of the structure is the channel: the pipe a given format travels through on its way to the person at the top. Email, YouTube, Instagram, a podcast app, a search result, a conference stage.

Jobs are served by products. Products are expressed in formats. Formats travel through channels. The hierarchy only functions in that order, because each layer exists to serve the one above it, and each layer's questions can only be answered by looking up. What format should this product take? Depends on the product and the person it serves. Which channels should carry it? Depends on where that person actually is when the need arises. Run the questions top to bottom and they answer themselves with surprising speed. Run them any other direction and they cannot be answered at all.

"We need a vertical video strategy" reaches down into the third layer of this structure, grabs a format, and installs it at the top as though it were a goal. And a format cannot be a goal, for the same reason that ink was never a publishing strategy and buildings were never an education strategy. The moment a format floats to the top of the hierarchy, every decision beneath it loses its reference point. What should the videos be about? However long? For whom, in what tone, on which platforms, judged by what? These questions have no answers, because the things that would answer them, the job and the product, were never named. So the team guesses. The output wobbles between imitations of whatever is currently performing on someone else's feed. The metrics disappoint, because metrics without a job attached can only ever measure activity, not progress. And eventually the organization writes off a perfectly capable format because it was asked to do a thing no format can do, which is to matter on its own.

Before I go further I should be clear about where I sit, because everything above could read as an argument against vertical video, and I have skin in precisely the opposite game. My company runs an entire division built on exactly this: we partner newsrooms with creators and influence journalists whose work is native vertical video, feed-first, nine-by-sixteen, the whole vocabulary this essay keeps warning about, and a real part of my consulting practice is helping newsrooms build short-form video operations.

Vertical video, in the most literal sense, pays my bills. That is not a caveat to the argument. It is the reason I hold it. I built that work as a division of my company rather than as the company itself for the same reason I am writing this: vertical video is a powerful avenue today, I have no idea whether it will be one in ten years (even two years!), and I do not put all of my marbles in a basket whose bottom is controlled by someone else's algorithm. We have also told clients, more than once, not to invest in vertical video at all, because they lacked the capacity to do it well or the audience case to justify it, and a firm that sells a format should be willing to say when the format is wrong. The people who build vertical video for a living should be the most disciplined about where it sits in the hierarchy, because we are the ones who intend to still be standing when the aspect ratio changes again.

The same format, in its right place

Here is what makes the story genuinely useful rather than just cautionary: the same format that presided over the collapse is, right now, at the center of one of the healthiest growth stories in media, and the difference between the two moments is exactly the difference this essay is about.

Podcasting spent its first two decades as an audio medium, and over the past few years it has been visibly becoming a video one. YouTube now reports more than a billion monthly viewers of podcast content, which would make the world's biggest video platform also its biggest podcast platform. Edison Research has found YouTube leading Spotify and Apple as the most-used podcast service among American listeners. The share of new weekly podcast listeners who prefer watching a show to merely hearing it has climbed to over half, up from under a third in 2022, and roughly seven in ten podcasters now incorporate video in some form. Shows that once existed only in earbuds now ship with filmed versions, clipped highlights, and channel art, and the ones that made the move early have been rewarded with discovery at a scale audio-only distribution never offered them.

Notice what did not happen here. Podcasters did not wake up one morning and declare a video strategy. The product already existed: a show, with a host, serving a defined audience's defined job, whether that job was staying current on technology or feeling accompanied on a long commute. The listener was already named. What changed was the answer to a question two layers down the hierarchy: where does the thing we already make need to travel, and in what form, now that the place where new listeners discover shows runs on video? That question has answers, and the answers were cheap relative to the value they unlocked, because the expensive layers, the product and the audience relationship, were already built. Video, held in its correct place, turned out to be one of the best distribution decisions available in media. Video, installed as the strategy itself, had been one of the worst. Same format, different layer, opposite outcomes.

The same logic redeems vertical video, in its place. Consider the newsroom covering a ballot measure, or a benefits change, or a school closure, something a real person needs to understand and act on. That reader arrives on a phone, mid-feed, with a few seconds of patience and a genuine question, and a tight vertical explainer meets her exactly there: right person, right need, right form, right channel. That is not a vertical video strategy. It is a product, an explainer serving a nameable reader's nameable job, that happens to be expressed vertically because vertical is where she is. Built that way, from the job down, the same nine-by-sixteen clip that fails as a strategy succeeds as a distribution decision. The format was never the question.

This is not a media problem

I have told this story with journalism's receipts because journalism paid for the clearest ones, but the error is not a media error, and if you work anywhere near a communications, marketing, or engagement function, you have been in the meeting.

The hospital system that decides it needs a TikTok presence, before anyone has articulated which patient or caregiver need its short video would serve, is making the pivot-to-video bet at smaller scale. So is the university advancement office that launches a podcast because peer institutions have podcasts, and quietly stops publishing it eleven episodes in, which is close to the average lifespan of an abandoned show. So is the nonprofit that hires for "digital content" without being able to say what job its digital content does for a supporter that its existing communication does not, and the B2B company whose leadership wants "a bigger presence on LinkedIn" without a theory of who on LinkedIn needs what from them. In every one of these rooms, an attribute from the bottom half of the hierarchy, a format or a channel, has been promoted to the top, and everyone downstream of the decision is left trying to reverse-engineer a purpose for it. The people assigned to execute these initiatives tend to be junior, and they tend to get blamed when the initiative produces activity without progress, which is the only thing it was ever structurally capable of producing.

Two questions that sort everything

The practical version of all of this fits in two questions, and they are portable to any organization, any format, and any meeting.

The next time the sentence arrives at your table, we need vertical video, we need a podcast, we should really be on TikTok, ask first: which product would this express? Not which topics, which product: what existing or planned thing, built for a nameable person, would this format be a version of? And then ask: what job does that product do, and for whom?

When I introduced TCI, I wrote that institutions adopting situation-based audience thinking stop launching products in search of audiences and start identifying audiences in search of products, and that the shift kills more bad product launches than any board oversight ever has. The two questions above are the same discipline applied one layer further down the stack. A format in search of a product is the little sibling of a product in search of an audience, and both die the same death, just at different budgets.

If the room can answer both questions, then what you are looking at is not a format initiative at all. It is a distribution decision, and it is probably a sound one, because a real product traveling in a suitable form to the place its audience already spends time is close to the definition of distribution working. The podcaster adding a camera because discovery moved to YouTube is answering these questions. The newsroom turning a genuinely useful explainer into sixty vertical seconds because the people who need the explanation arrive through feeds, on phones, with four minutes of attention, is answering them too.

If the room cannot answer, then what you are looking at is platform anxiety wearing a strategy costume, and the honest response is not to kill the idea but to send it back up the hierarchy where it belongs. Find the person. Name the job. Build or identify the product that serves it. By the time you have done that work, the format question will usually have answered itself on the way back down, and it will sometimes surprise you: the audience you actually need to reach may turn out to live in an inbox, or a conference room, or a search result, and the vertical video budget may be better spent there.

There is a discipline that follows from the test, and the organizations that communicate well practice it whether or not they have language for it: format volume follows purpose, not platform appetite. You make more video when the things you are building call for more video, a launch, a season, a moment your audience is living through, and not because an algorithm is currently generous or a competitor's feed looks busy. Platform appetite is real, but it is weather, and it changes without notice or apology, as every publisher who restructured around a single channel's preferences learned in the span of one algorithm update. Products serving named jobs are climate. Organizations that plan around the weather rebuild every time it turns, and the rebuilding is always described, in the moment, as a bold strategic pivot.

I have been circling this error from different directions all year. In What Demographics Miss, I argued that audiences are not demographics, that who someone is on a spreadsheet predicts almost nothing about what they need from you. In When Audience Goals Become Performance, I argued that goals built on those demographics are performances, satisfying to announce and structurally impossible to be held accountable to. This essay is the same argument pointed inward, at the organization's own output instead of its readers. In every version, an attribute gets promoted into a strategy, an age, a format, a platform, and in every version the tell is identical: nobody in the room can name the job at the top of the chain. The fix is also the same, and it is deliberately unglamorous. Start from what a specific person is trying to do. Build the thing that does it. Express that thing in whatever material suits it, and send it down whichever pipes reach the person it was built for. Nothing in that sequence will ever sound as exciting as a vertical video strategy in a Monday meeting, and that is precisely how you can tell it works.