TheEconomicsof the Feed

Attention Is a Currency

A currency is any medium that stores value and can be exchanged. We tend to think of currencies as tangible: coins, bills, numbers in a bank account. But the defining feature of a currency isn’t its physical form. It’s that it’s scarce, it’s valuable, and people make decisions about how to spend it.

Attention meets every criterion.

It’s scarce. Gloria Mark’s research, which we covered in the previous chapter, puts focused attention on a single screen at forty-seven seconds. But that number describes the container, and it overstates what any individual piece of content gets. Inside a feed, a single image holds the screen for about nine tenths of a second at the median, which is roughly the time it takes a thumb to move a post from the bottom of the viewport to the top and off it.

It’s valuable, and it is worth being precise about how little of it there is. Lumen Research, the attention measurement firm, has spent over a decade finding that average dwell across digital ad formats falls between one and two seconds, and their work with Ebiquity (Ebiquity and Lumen Research, Maximising Profit Through Attention, November 2024) showed a measurable correlation, across media channels, between attentive seconds and incremental profit. But average is doing a lot of work in that sentence. Attention time is a heavily skewed distribution: in Parry and Masur’s feed data (PsyArXiv preprint, 2026) the mean was 2.79 seconds and the median was 0.90, dragged apart by a long tail of rare, long looks. Simonov and colleagues, publishing in the Journal of Marketing Research in 2025, found the same shape from the other direction: an ad that was on screen for 19.03 seconds was looked at for 2.76. So the honest version is two numbers. Averages land between one and two seconds. The typical piece of content gets about nine tenths of one.

And people spend it. Every time you scroll, you are making a series of micro-transactions. Each piece of content that enters your viewport makes a bid for your attention. You either pay, by slowing down, focusing, or stopping, or you don’t. The scroll itself is the transaction mechanism, the marketplace in motion.

Simon understood this in 1971, though the marketplace he was describing was organizational, not digital. He argued that designers of information systems kept framing their problem as information scarcity, how to get more data to decision-makers, when the real problem was attention scarcity: how to make sure the right information got noticed amid the noise. As he put it, progress isn’t about producing more information. It’s about developing better systems for filtering it.

Fifty-five years later, every social media platform, every algorithmic feed, every content management system is an attempt, sometimes brilliant, sometimes predatory, to solve Simon’s problem at planetary scale. And every piece of creative work that enters those systems is subject to the same economic logic: you are competing for a finite resource that depletes with every impression.

Creative Is the Product

If attention is the currency, then what is the product?

In a traditional economy, the product is the thing being sold. In the scrolling economy, the product is the thing being traded for attention. That means the creative is the product. The ad, the post, the video, the carousel, the reel: every piece of designed content that enters a feed is a product that either earns the attention it needs to survive or doesn’t.

This reframe matters because most creative teams don’t think of their work as products in an economic system. They think of it as communication: messages to be delivered, ideas to be expressed, briefs to be fulfilled. In a broadcast world, where the audience was captive and the delivery system guaranteed exposure, that framing worked. You made the thing, you placed the thing, and people saw the thing. The quality of the creative affected how well the message landed, but it didn’t determine whether the message was seen at all.

In a scroll environment, it does.

A piece of content that fails to earn attention in the first one to two seconds doesn’t just underperform. It effectively doesn’t exist. It was manufactured, distributed, and delivered to a viewport, but it generated zero cognitive engagement. The impression was logged. The attention was not paid. In economic terms, the product was placed on the shelf, but nobody picked it up.

The published measures make this uncomfortably clear, and they are narrower than the industry’s own vocabulary. The standard for a viewable impression is fifty percent of an ad’s pixels on screen for one continuous second (Media Rating Council, Viewable Ad Impression Measurement Guidelines, 2015), which is an audit threshold rather than evidence that anyone looked. In a controlled eye-tracking experiment, ads met that bar for just over nineteen seconds on average and were looked at for 2.76 seconds of them (Simonov, Valletti and Veiga, Journal of Marketing Research, 2025). On television, the average viewer attends to 7.7 percent of the ad seconds served (McGranaghan, Liaukonyte and Wilbur, Marketing Science, 2022). Whichever measure you take, most of what is bought as attention is never given.

Think about that in economic terms. Imagine a factory where almost everything it manufactured was never picked up by a single customer. You wouldn’t call that a messaging problem. You’d call it a product problem. You’d redesign the product.

That’s the opportunity hiding inside the data. The scrolling economy doesn’t punish creative work. It punishes creative work designed for a different marketplace: work built for the presentation room instead of the feed, work that leads with the message and assumes the attention it should be earning.

The Feed Is the Marketplace

Every marketplace has a structure. Physical retail has aisles, shelf placement, and eye-level positioning. E-commerce has search algorithms, recommendation engines, and product grids. Each structure creates its own rules for what gets seen, what gets ignored, and what converts.

The feed is no different. It’s a marketplace with a specific structure, and understanding that structure is the first step toward designing work that succeeds inside it.

The feed is vertical. Content enters from below and exits from above. The direction is fixed: you scroll down (or, on some platforms, right), and information passes through a narrow viewport window. Unlike a magazine, which you can flip through and return to, or a website, where you can move laterally, the feed is linear and temporal. What’s gone is gone.

The feed is infinite. There is no bottom. The supply of content in any social feed is, for all practical purposes, unlimited. The number usually quoted here, five hundred hours of video uploaded to YouTube every minute, is retired. YouTube published it until May 2025 and now states only that “on average, there are over 20 million videos uploaded daily.” The two cannot be converted into each other, because YouTube has never published an average video length, and the conversions circulating online are invented. A better measurement exists in any case. McGrady and colleagues built a genuinely random sample of YouTube by guessing video IDs, roughly eighteen trillion attempts for 10,016 hits, and estimated 9,881,141,822 public videos as of late 2022, of which 25.91 percent had been uploaded during that year alone (McGrady, Zheng, Curran, Baumgartner and Zuckerman, “Dialing for Videos: A Random Sample of YouTube,” Journal of Quantitative Description: Digital Media 3, 2023; DOI 10.51685/jqd.2023.022. Daily uploads: blog.youtube/press, accessed 23 September 2026). A quarter of everything ever posted to the platform arrived in the final twelve months of the count. Buffer’s analysis of more than fifty-two million posts (The State of Social Media Engagement in 2026: 52M+ Posts Analyzed, 5 March 2026, and limited by Buffer’s own statement to “Buffer users and Buffer-posted content only. It’s not a full-platform view of any network”) found that even as median post volume on Bluesky nearly quadrupled year over year, median engagement per post slipped. Supply grows. Attention doesn’t.

The feed is competitive. Every piece of content in a feed is in direct competition with the content above it, below it, and everything the viewer has already scrolled past. There are no captive audiences and no obligation to look. The feed is a pure attention market, and in a pure market, only the product that earns the most value for the least effort survives.

And the feed is algorithmically sorted. On most platforms, content doesn’t appear in chronological order. It appears in predicted-engagement order: the algorithm surfaces what it believes will generate the most interaction, based on past behavior. The feed isn’t just a marketplace; it’s a marketplace with a gatekeeper. And the gatekeeper’s only criterion is attention.

For designers, this creates a dual challenge. You have to earn attention from the human viewer, and you have to signal value to the algorithm that determines whether the viewer ever sees your work. These are not always the same thing, but they converge on the same principle: if your creative doesn’t generate engagement in the first moments of exposure, it gets pushed down, shown to fewer people, and eventually buried.

The Four Laws of the Scrolling Economy

Law 1: Attention depreciates with supply.

Law 1 is the one nobody argues with and almost nobody counts properly. Content supply is growing without limit; the attention available to absorb it is biologically fixed. The numbers usually quoted for this are worth handling with care. Metricool’s 2025 study covered twenty-one million posts from more than a million accounts across eight networks; the forty-million figure belongs to its 2026 edition. And the population number usually attached to those studies is not Metricool’s at all. The 5.66 billion social media users, each holding accounts on about seven platforms, come from DataReportal and GWI, and DataReportal warns in its own pages that its “figures for social media users may not represent unique individuals,” which is why it calls them identities rather than people. They are accounts declared to advertisers. The ITU, which counts through household surveys instead, reports that almost three-quarters of the world’s population is now online and that 2.2 billion people are still not. The gap between the two counts is the point: the industry counts accounts, and attention belongs to people. Either way, the direction does not reverse.

For designers, this means that creative quality is no longer a competitive advantage. It’s the price of entry. Average work didn’t used to disappear. It just performed averagely. In the scrolling economy, average work is invisible work.

Law 2: The first impression is the only impression.

In a scroll environment, you do not get a second chance. A viewer who scrolls past your content will almost certainly never see it again; the feed moves forward, and so does the thumb. The first moment of exposure is the entire customer journey compressed into a single interaction. If your creative doesn’t earn recognition in that window, it earns nothing.

This is the fundamental difference between broadcast and scroll environments. Broadcast media could afford to build. A thirty-second commercial could open with an intriguing image, develop a narrative, and deliver a punchline. The audience was sitting there. In the scroll, there’s no guarantee anyone stays past the first frame. The first frame has to be the argument, the brand signal, and the emotional hook, all at once.

Law 3: Cumulative signal beats transactional impact.

The most important shift in the scrolling economy is from transactional thinking to cumulative thinking. The goal of a single piece of content is not to close a sale, change a mind, or deliver a complete message. The goal is to deposit a signal, a fragment of brand recognition or emotional residue, that accumulates over repeated exposures into something durable.

Havas, Lumen and Brand Metrics showed this in the largest study of its kind: across 9,089 campaigns and 1.9 million people, the strongest movements in brand lift came when people engaged with several ads, each for a “good enough” amount of time. The finding comes with a condition, and the condition matters. It holds for awareness and consideration. For preference and purchase intent the same study points the other way, toward a single high-attention placement. So the law is narrower than it first sounds: accumulation is how a brand gets recognized, not how every objective gets met. The point is repeated recognition, across different contexts, until the brand occupies a distinct and persistent space in the viewer’s mind.

This is how the strongest brands in the scrolling economy operate. They don’t try to say everything in every post. They say one thing, the same thing expressed differently across dozens or hundreds of touchpoints, until it becomes automatic. Color becomes association. Shape becomes identity. Repetition becomes recognition.

Law 4: Signal needs no consent.

Call it the governing law of this book. In the scrolling economy, a visual signal reaches the viewer on terms a verbal message cannot match, and the reason is mechanical rather than rhetorical. A complex natural scene can be categorized in under one hundred fifty milliseconds, inside a single fixation; Thorpe, Fize and Marlot showed that with photographs held on screen for twenty milliseconds (Nature, 1996). Larson and Loschky then asked where in the visual field that categorization comes from and found peripheral vision more useful than central vision (Journal of Vision, 2009), which means an image can register without the eye ever landing on it. Text has no such option. Reading requires foveation: fixations of roughly 200 to 250 milliseconds, moving seven to nine letter spaces at a time (Rayner, Psychological Bulletin, 1998), at around 238 words a minute for nonfiction (Brysbaert, Journal of Memory and Language, 2019). An eight-word headline is two seconds of work the viewer has to agree to do first. The image underneath it is already finished. Signal arrives before meaning, and that is a fact about eyes rather than a preference about design.

And yet, most creative work is still designed message-first. The brief starts with what we want to say. The creative starts with copy. The brand guidelines start with tone of voice. In the scrolling economy, this is building the house from the roof down.

Designing for the scroll means designing signal-first. It means leading with what the composition says as a whole, the palette, the silhouette, the contrast against the feed around it, and trusting that to do the heavy lifting before a word is read. Image before text. Composition before copy. The signal is what earns the right to deliver the message, not the other way around.

The Compound Effect

There’s a concept in investing called compound interest: small, consistent returns that accumulate exponentially over time.

The scrolling economy has its own version of compound interest, and it works the same way.

Every time a piece of brand communication enters a viewport and deposits a recognizable signal, even if the viewer doesn’t stop, click, or consciously register the brand, something accrues. The signal adds to a cumulative store of recognition. And over time, that store produces returns that individual impressions never could.

That’s why brand systems matter more than individual executions. A campaign that produces one brilliant piece and nine forgettable ones accumulates less signal than a system that produces ten pieces of consistent, signal-first work. The brilliant piece might earn more attention in isolation, but the system earns more recognition in aggregate.

It’s also why the scrolling economy rewards the patient over the spectacular. The brand that shows up with the same visual DNA across hundreds of touchpoints, the same palette, the same spatial logic, the same recognition cues, builds an asset that compounds. The brand that reinvents itself with every campaign starts over every time.

Newsworks, Lumen and the effectiveness analyst Peter Field put a number on this at the media level in their 2025 Attention report: high-attention media plans, campaigns that prioritized quality attention over sheer volume of impressions, grew market share twelve percent more than low-attention plans. The investment in being seen properly, rather than just being seen often, paid measurable dividends. Read the number for what it is. It comes from the IPA Effectiveness Databank, the Institute of Practitioners in Advertising’s archive of case studies submitted to its effectiveness awards since 1980, so the campaigns in it are the ones their agencies chose to enter and believed had worked. There is no peer-reviewed replication of the twelve percent. It is the best evidence the industry has produced on this question, and it is industry evidence.

For creative professionals, the compound effect reframes what “success” looks like. A single viral post is not the goal. A system that accumulates recognition over time is. The feed rewards consistency over spectacle, and investment over transaction.

What This Means for the Rest of the Book

Part I of this book has laid the foundation. We’ve reframed the scroll as the medium, not the obstacle (Chapter 1). We’ve examined the research on how attention actually works and how dramatically it has shifted (Chapter 2). And now we’ve mapped the economic logic of the feed: attention as currency, creative as product, the feed as marketplace, and the four laws that govern everything inside it.

The scrolling economy is not a crisis. It’s a context. It has rules, and those rules can be learned, applied, and used to your advantage.

Part II introduces the framework for doing exactly that, one principle per chapter, starting with the micro-moment: the fundamental unit of design in the scroll.

Each chapter opens with a data point and closes with a practical principle. Each is built to respect the thing it’s about: your attention.

The rules are on the table. The rest of this book is how you work inside them.