319. That’s the number of discrete advertisements, both online and off, served to me over the course of one Tuesday in July. I know because I counted each and every one.
Why catalog the informational noise that, in many cases, our brains have quite literally trained themselves to ignore? Hidden in plain sight, ad money is the invisible force that subsidizes many of the services we depend on—especially online, where keeping up with friends, reading the news, or streaming music is ostensibly “free.” I wanted to to find out exactly what my eyeballs are worth.
After turning off ad blockers on both my work and home laptops, I did my best to have a normal day, making no effort to avoid or intentionally expose myself to ads. The act of writing down each instance of advertising may have slowed down my usual browsing pace slightly, but otherwise this was a regular Tuesday: working from home for a few hours, commuting from Brooklyn to Manhattan, walking to meet a friend in the East Village after work, taking a cab back, and catching up on more work at home while watching TV. Bland as this routine is, it put me in front of a variety of digital ads on desktop and mobile, as well as what are called “out of home ads,” which include things like physical posters and billboards.
Forcing yourself to notice and comprehend each ad you encounter in a day is, it must be said, a thoroughly unrewarding experience. It had been over a decade since I really browsed the web without plugins like uBlock Origin and Canvas Defender. Boy, has it become a cluttered and ugly place. Amazingly, around 75 percent of Americans look at the internet this way. Given the obnoxious tactics used by advertisers (popovers, ad units that follow you on scroll, auto-playing videos with sound), irrelevant products promoted, and ever-growing volume of ad units per page, it’s no wonder users have learned to ignore these impositions.
To make this experiment achievable, the only ads I counted were ones where a company paid for time or space someplace they did not already control. Store awnings, branded delivery vehicles, product reviews, and HBO’s self-promotional spots were all excluded for this reason. I applied the same exception to incidentally advertorial merchandise, like the guy wearing a Pepsi shirt on my train ride. Sure, these are powerful tools for raising brand awareness, but they’re difficult (if not impossible) to ascribe direct monetary value to—not that any part of this experiment is quite up to scientific snuff.
One more note: Most ad companies do not make their rate sheets public and aren’t keen to share those figures with reporters or the general public. Unless otherwise stated, all companies brokering these ads were contacted and declined to respond.
Desktop advertisements made up a little less than half of my daily ad total and (despite some distracting news dropping that day) I expected the percentage to be much higher. Maybe it shouldn’t have come as a surprise, given the research Pew and other organizations have put out regarding the prevalence of Americans getting their news from social media. In my case, I was also getting the day’s news by scanning headlines on RSS, Google and IFTTT alerts, and push notifications from the New York Times, Reuters, CNN, Washington Post, and Wall Street Journal apps. Clicking into an article was pretty strictly reserved for things I had a strong inclination to believe would be relevant to the beats I write on, and each story spat out between one and four ads, based mostly on how deep on the page I scrolled.
One of the most common ways online ads are priced—especially the traditional banner ad—is called CPM (cost per mille), the amount of money needed to serve that ad to 1,000 people. I brought a list of these ads, including information about where they appeared on a page and what elements they contained, to Jillian Schulz, Gizmodo Media’s VP of Media Operations and Ryan Brown, our SVP of Business Development. The two of them provided what they felt were conservative market rate CPMs for each, with the consideration that the sites I’d visited were largely established, well-trafficked news sites. Their estimates ranged from $5 for static banner ads along the right “rail”—a section of the screen many users ignore outright—up to $30 for autoplaying video ads.
Included in this section are ads from Reddit, Google, and the RSS reader Feedly (R.I.P. Digg Reader). Reddit browsing for me usually involves a peek at the site through some multi-Reddits I have for specific subjects, which aren’t equipped to promote posts. The absolute minimum CPM on Reddit’s self-serve ad product is 50 cents, but frontpage targeting appears to only be available through managed campaigns, which require a minimum spend of $50,000. Jillian and Ryan felt $1.50 was a safe CPM to use for these posts. Feedly’s text-only ads, which are slotted into the feed, they pegged at $0.75 per mille. Google’s sponsored search results are most commonly priced per click, abbreviated CPC. I didn’t click any of them, so those likely didn’t cost their advertisers a thing to show me.
Estimated Cost to Advertisers: $0.824
With a little under an hour to kill, I tend to listen to podcasts on my way into work—in this case, two short ones (NPR’s The Indicator and Death, Sex & Money) each with one ad, and half of a longer one (Maximum Fun’s My Brother My Brother and Me) with two. According to David Raphael, the President of Public Media Marketing—a company that works with Maximum Fun, as well as shows like This American Life and The Joe Rogan Experience—podcast product spots tend to be “stickier” than many other forms of advertising, and command relatively high prices as a result. In his telling, typically market rates span from a $20 CPM up to $60. Without knowing the listenership numbers of any of these shows, I went with $20 across the board.
The next part of the equation is public transit, and unlike digital ads, where eyeballs and clicks can be directly tracked, out of home ads tend to be a little squishier. Where cost is concerned, the only person to respond to my inquiries on subway ads was Danny Pouladian, the director of an ad broking firm called Blue Line Media, who sent over when he called “very rough numbers” for various four-week campaigns—typically the shortest buy for out of home—that were priced to reach 25 percent “share of voice” (SOV) meaning a quarter of all people in Manhattan. (According to a study from NYU’s Wagner Graduate School of Public Service, the total number of people in the city, including commuters, is around 4 million people daily.)
Going off his figures, the daily cost for an ad plastered to the side of a bus is a little more than $35, while interior subway ads—of which there are often a dozen or more per car—are less than $2 each. Given the audience scale we’re operating with here, these all amount to an absolute pittance. Even using the figures listed on Blue Line’s website, which don’t specifically mention a target SOV percentage, dividing those costs by the MTA’s estimated ridership for my stop (15,269) still did little to unbury the individual pricing from beneath a slew of zeros.
Most of these campaigns are expensive on their face, and a far cry from Reddit’s self-serve minimum daily spend of $5. A “station domination,” the horrendously named practice of a brand plastering itself over nearly every square inch of a subway stop (as Taskrabbit did to Union Square on this particular day), is estimated by Pouladian’s firm to cost “$25,000 - $75,000 or more” for a four-week run. Account for an average weekday ridership at Union Square of 107,141 people, and each of those commuters is worth between an eighth of a cent and two-and-a-half cents.
One of the most frequent out of home ads I (and probably every other New Yorker) sees are the illuminated miniature billboards on the tops of cabs, called taxi toppers. John Schweikert, President of Strong Digital Media—one of the Taxi and Limousine Commission’s largest brokers of such placements—was kind enough to provide an approximate CPM of $1. Bus shelters also double as ad spots, and without anyone willing to comment on local costs, I opted to use the CPM for Los Angeles (designated “Market Area #2,” right behind NYC) from Clear Channel’s rate sheet. Many sidewalks in New York now feature Link NYC terminals for phone charging. The sides of those function as targetable, digital banners, with average ad CPMs of between $8 and $12, a figure the company provided me over the phone and then attempted to walk back on in an email. Even the elevator in the building Gizmodo works out of has a screen that continuously cycles through ads, which are served through a company called Captivate. They provided me an estimated $10.60 CPM for the buildings in our neighborhood, making it one of the more expensive offerings, but, well, what else is there to do in an elevator?
On the cab ride home I also came across one traditional billboard, located on Delancey Street heading onto the Williamsburg Bridge. Lamar, the company that rents out that billboard, estimates a cost of $10,500 to $17,500 for a four-week campaign on that size and style of billboard in the New York metro area. The most recent New York City Bridge Traffic Volumes report states that 105,645 vehicles go over the Williamsburg Bridge each day, and the Federal Highway Administration estimates 1.7 passengers as the average vehicle occupancy. Some light arithmetic, and that’s $0.002 to $0.0036 per rider.
The last out of home category here is probably the oddest, in that it accounts for both the ad and its own infrastructure: Citi Bike. Each bike and docking station is covered in Citi branding, and Citigroup has invested a total of $11.5 million to support the bikesharing project it started in 2013 through 2024. Rather than than calculate this per bike, I went by station: I walked past two of the 750 that currently exist. The New York Times estimates walking a city block takes approximately one minute, and looking at said branded bike dock for that minute costs Citigroup just over two and a half tenths of a cent. Does anyone at Citigroup calculate the value of a bikeshare this way? Probably not, particularly because one person noticing a rack of indestructible blue bikes doesn’t preclude anyone else from noticing it. But for this particular goose chase I do could not come up with a better estimation, because what little foot traffic data I could find for New York is woefully incomplete.
Estimated Cost to Advertisers: $0.1852388
An outsized portion of my smartphone’s usage boils down to checking social platforms. And unlike desktop client Tweetdeck, Twitter on mobile is littered with ads. Instagram and Tumblr still cross my tiny screen, if less often these days, but I rarely check Facebook at all, thanks, in part, to a similarly harebrained experiment.
A Tumblr representative was kind enough to confirm to me that the network’s ads are all priced on a CPC basis. Once again, I don’t really click (or tap) ads ever, so that’s worth nothing. Surprisingly, Jillian and Ryan indicated that that was the case with many static social ads—the outliers being promoted tweets with autoplaying videos embedded, and Instagram stories (which also autoplay.) They suggested a $20 CPM for anything of that nature. Instagram advertisers tend to bundle two or three stories into a single ad package, and since I find tapping faster than swiping, those all factored into the total.
The biggest headache in this otherwise breezy chunk of the project was the presence of “promoted trends”—Twitter’s way of letting brands buy their own hashtag campaigns. On this particular day, #christopherrobin was trending at the behest of the Walt Disney Corporation (thanks, guys). The only public data on the cost of these campaigns comes from early 2013, when Twitter’s user base was about half what it is now, and the company had not yet gone public. Back then, a promoted hashtag was reported to cost around $200,000. Around that time the service had recently hit 200 million users, giving us a nice, round $0.001 each.
Estimated Cost to Advertisers: $0.401
Being a millenial, it’s my moral duty to avoid cable at all costs and replace it with a series of equally janky video services. Over the course of that Tuesday, my video regimen consisted of YouTube, Twitch, and the lower-tier ad-supported Hulu membership I borrow from a friend.
For pre-roll ads on the former two, Jillian and Ryan again went with a $30 CPM—although YouTube also uses small, semi-opaque banner overlays inside the video player ($1.50) and standard right-rail banners above suggested “up next” videos ($5).
Hulu is fairly secretive about their ad rates, but a report from Digiday published last November cites CPMs “in the mid-$30 to $40 range.” Conservatively, we’ll call that $35, and given how aggressively the service inserts ads, binging a few episodes of reality knife-making show Forged in Fire was worth over a dollar.
Estimated Cost to Advertisers: $1.2825
Of course, there were a few things I couldn’t even venture a guess on. The cost of the vinyl banners around Union Square for a music showcase series (also sponsored by Citigroup) remain a mystery, as do things like subway escalator wraps and the posters glued to the sides of New York’s dwindling public payphones. And I got to $2.69 using conservative estimates. Even if we assumed that the true cost is closer to double that due to lapses in cataloging and pricing these ads, it would still barely buy you a coffee and a bagel.
More than anything else, trudging through the numbers on these various campaigns got me thinking about information pollution, and the sheer amount of space in our day that’s monetized. Introducing limits sounds pleasant enough—who wouldn’t want fewer ads?—but that might also tip the balance even further in favor of huge brands with budgets to burn on the ad spaces that would become costlier through artificial scarcity. Maybe aggressive ad density could be taxed like cigarettes, air pollution, or anything else with negative externalities. I’m not the first or last person to see the reliance on advertising—especially online—as flawed, and we should all be deeply concerned by the way it incentivizes the sort of mass data collection Facebook and Google do.
The low total I reached seemed, at first, to be an argument in favor of paywalling the world. Would I spend $3 a day to never see an ad for the rest of my natural life? In a heartbeat. But this total really only represents half the story. $2.69 is what it costs brands to buy that real estate and have it reach one living, breathing, spending human, not the value of the advertising to those brands—and to know that number would require proprietary data on each campaign as well as their specific goals. I would imagine the value figure is considerably higher. And in any case, paywalls on every major site would a) be a consumer nightmare and b) is an inherently classist approach that would likely harm those who already have the least access to libraries, good public schools, and other traditional sources of information. Advertising isn’t my area of expertise, nor is economics, and for the time being I can’t offer a picture of a better path forward.
As a small, semi-scientific experiment, however, I did go into this assuming my eyes and ears would be more expensive. It’s left me feeling a little short-changed, and given that I write online for a living, somewhat guilty that my paycheck is at least partly contingent on the irritation and inconvenience websites inflict on people just trying to find out what the hell is going on in the world. If some measure of self-worth is what you’re after, I don’t recommend looking into advertising as the avenue to find it.