Defense in the War on AdTech: Case 2

Why decentralized ad networks will become more important than ever

This story is a continuation from my two-part post, A Defense in the War on AdTech. You can read the first post here.

In my first piece, I covered why AdTech is a great tool for content strategists to identify audiences according to interest groups, defined by a publication or magazine’s readership. However, using ad placement tech for audience insight is not the traditional use of these technologies. Most teams use it to simply direct traffic to product or site pages.

I ask a fundamental question to the Eisenbergs (Michael A. Eisenberg: Six Kids And A Full Time Job) and other AdTech skeptics, which will drive this piece: If AdTech is gone, what’s the alternative method to reach consumers? Facebook and Google, right?

Screeeeech the brakes.

Of course, Facebook and Google work when a brand wants to reach customers. In fact, they work so well that they have what eMarketer calls a duopoly on all digital ads. In their analysis of 2018 ad spend, eMarketer reported Facebook and Google took 57.5% of the digital ad spend market, or $64.2 billion. In 2020, they predict Amazon will come play on the playground as well, where these three would own 62.9% of the ad spend pie, and probably be hosting about that much of all digital ads (number and scale of ads only correlates with dollars spent, does not predict it).

With those rates, the argument in favor of ad placement technologies is that they promote decentralization of traffic. This means that putting all the ad eggs in the Google and Facebook network basket can be dangerous, because those networks can cut off any advertisers’ access to their customers with the flip of a switch.

This is referrals to’s site back in 2017. The hold these aggregators have on attention industry-wide is even worse today. (via Joe Lazauskas at Contently)

This is exactly what happened in 2017 and in January 2018 when Facebook drastically changed its News Feed algorithm in favor of friends’ posts, not page or brand posts, because Zuckerberg wanted to “bring people together”. Adam Singolda (adam singolda), the founder of Taboola, cited this risk back in 2014 as part of the reason Taboola could and should exist. The topic is still relevant, especially with FB’s algorithm changes. Joe Lazauskas, Director of Strategy at Contently (and my previous boss/friend!) discussed dangers of Facebook dependency in Fast Company last May.

Ben Thompson of Stratechery, a trends and investing guru, warns advertisers and publishers of “aggregators” like Facebook. He explains that vis-a-vis their position as an ad platform, Facebook’s and Google’s goal is to solely to grow their network to as many users as possible, not necessarily keep advertisers happy, because “the more dominant an aggregator the more powerless the supply.” The aggregator brings buyers (advertisers and publishers) in on their own terms, because no one else has what they have — user attention. Hence duopoly of attention. That’s why Facebook happily made those changes to their News Feed algorithm, because their goal is to grow and keep their network of users, and they know advertisers have no other network at that scale to go to.

Thompson draws this analysis in the context of BuzzFeed’s slow growth and recent layoffs. BuzzFeed grew using Facebook Instant Articles, or sponsored Page posts. Except then this happened, as Thompson so eloquently explains:

Instant Articles relied on the Facebook Audience Network, not Facebook’s core News Feed ad product, and nearly all of Facebook’s energy went into the latter. Companies that embraced Instant Articles — and, in the case of BuzzFeed, built their business models around them — were left earning pennies, mostly on programmatic advertising.

What would have happened to Buzzfeed if there was no Ad Tech (programmatic advertising)? Notice that Buzzfeed’s downfall relied on the Audience Network, which I explained in Case 1 was deeply flawed. But again, even if they had advertised their articles as ads in the News Feed ad product, not sponsored posts, Facebook is still in full control of who sees their content. With dark ads (News Feed ad product), viewership is gained and lost strictly according to how many dollars paid (much more expensive per impression than any ad placement tech company due to lack of competition). With the algorithm changes, a brand can’t build its audience organically with time and sweat and hard effort, accruing like by like day after day, because they’re getting cut out of the NewsFeed as a Page. This is what happened to BuzzFeed.

That creates a philosophical argument: are we in the Gilded Age of attention? If no one can get attention if they’re not paying for Facebook dark ads, an SEO specialist, or Google Ads, then yea, maybe we are. So then what is the only kind of tech that can break that monopoly of attention up and offer a competitive attention market for advertisers? AdTech.

The Gilded Age of Attention Means Costs Will Skyrocket, Too.

In the “Gilded Age” of advertising, when an aggregator grows ever-stronger, not only will it control attention, they will broker that attention at an increasing cost (Facebook is far more aggressive about this). Gary Vaynerchuk talks about this on LinkedIn a lot.

By eliminating opportunity for organic growth, Facebook will leave brands with no option but to purchase ads at an increasing cost, tightening its grip on any brand’s access to its own audience.

The Gilded Age of Attention Growth Chart. Traditionally, organic growth marketing means that as attention increases, cost for more of that attention decreases (sort of like wholesale purchasing at Costco). New Gilded Age of Attention Marketing is when Facebook becomes the only broker between brands and their audiences, it therefore will continually increase the cost to purchase more and more attention. As cost over time increases, one dollar will buy less and less attention (sort of like Carnegie Steel Co., but even worse).

In organic-heavy growth models, brands pay an initial investment to get attention (say, Buzzfeed’s Sponsored Posts) and then build up their network, making their reach less costly over time because people share “organically”. In other words, there is a cost deduction with growth over time (green line on the chart).

In the new model where a brand can only reach an audience through paid advertisement on Facebook’s digital ad network, not only will impressions be strictly controlled through these ads (which would yield a linear correlation between cost and impressions), but also the cost is increasing, so it will cost more and more over time for less and less attention (red line on the chart).

This is the most grave danger and current risk for advertisers. What will happen is that attention and brand growth will be negatively correlated to dollars paid. Organic growth will be obsolete (or at least much, much harder to achieve).

Mark Zuckerberg talked about changing the News Feed algorithm in 2017/18 as a way to “bring people closer together” As if. By implementing the change (and continuing throughout the year), Facebook is prioritizing its ad network over organic growth, shifting brand growth on its network to look more and more like the red line on the chart. For this reason,

AdTech engineers, we need you now more than ever. Only competition in the ad tech market will allow every single brand the opportunity to advertise with the chances of growth along the green line, not the red.

TLDR; viewership will be gained and lost strictly according to how many dollars are paid at a given time. Growth will slow with increasing costs for ad results. A brand will no longer “own” its audience as aggregators will be the only brokers with access to that audience.

Zuckerberg Is 2019’s Tywin Lanister

I want to throw in this GOT analogy because I think it will be helpful to understanding what Facebook is doing in the advertising world. If you already get the picture, head on to the next subsection.

Imagine you’re a traveler in medieval times. You gallop on your trusty stead towards the gates of a new city called King’s Landing. That city has food, water, shelter, and perhaps opportunity for work. However, in order to get into the gates of the city, you need to pay. The entrance fee costs 15 shillings, which you calculate will be more than what you’ll earn. You continue on the King’s Road to another city.

Over yonder is a city called Winterfell. You trot towards Winterfell, and the entrance fee is 10 shillings. Still too much; you’d be breaking even.

Eventually, you cross a bridge and get to Dorne, a lovely sunny city that charges only 5 shillings to enter because Dorne needs laborers. You gladly pay the entrance fee into the gates. That day, you’ve earned 10 shillings.

Every day you come back to Dorne, and pay five shillings each time. As you meet more people in Dorne, your reputation and your business grows. The guards at the gate know how much your services improve Dorne, and they offer to lower the cost of entrance to three shillings. As you’re making more money by working for more people, you might even become a citizen of Dorne and not have to pay an entrance fee at all; everyone knows you and wants your services. This is called organic growth. In The Gilded Age of Attention Growth Chart, this would be growth along the green line.

One day, the evil King of King’s Landing named Marke declares war to control the entire realm. He starts by conquering each city, including your beloved Dorne. Now, there is no more free entrance to your customers as tribute for the loyalty you’ve built; you have to pay the initial fee of 15 shillings every day no matter how many people love and need your work. With Marke’s total control over the realm, there is no other city in which you can get to customers for a lower entrance fee.

What’s more, the leadership at King’s Landing knows that your small business will increasingly rely on the customers inside the city gates that they control as they conquer more cities. So, they raise the price of the entrance fee, charging a fee per customer every single day just because they can; there’s no other city you can go to. So, you end up with less customers and higher fees to entrance. Your business would be growing along the red line in our Gilded Age of Attention Growth Chart.

In case it wasn’t clear, the advertiser is the knight. Facebook is the royal government that controls the gates, and the customers inside are Facebook users.

Bad Business Models And Good Ideas Are Not The Same Thing (Thank God)

Now, let’s shift focus from the aggregators’ grasp on attention to the only other companies that can break up that grasp: AdTech companies.

Eisenberg calls on engineers to leave AdTech because, from a business model standpoint, revenue is too volatile. But, I argue, that doesn’t mean AdTech necessarily can’t have a scaleable, foolproof model if they don’t make some changes to those business models. Let’s explain:

Ad placement technologies operate with an bidding and optimization system. The advertiser competes for impressions and clicks, where the cost for a click varies according to the time of day depending on the volume of traffic (some ad placement companies charge by impression). In Case 1 of this series, I spoke about a hypothetical client and advertiser, a plant-based facial serum. Let’s say my serum client has a $1 budget that day and wants to publish only through Outbrain, an ad placement software. There’s a spot at the side of an article on Women’s Health that costs 80 cents per click, on that day, at that time. With the serum’s $1 budget, Outbrain will take the 20 cent delta as profit.

My rough infographic of the current ad placement tech business model.

Eisenberg explains that because Outbrain can’t control Women’s Health Magazine’s traffic (what if Women’s Heath starts publishing bad content and people stop visiting?), their revenue is subject to that volatility, too.

This creates risk, which could drive down Outbrain’s value. If we’re defining value by stock price, we can take a look at Criteo Labs, the largest ad placement and retargeting firm, whose stock has decreased 22% since they went public. This and other ad tech stocks have led to speculation of the entire AdTech industry. Wall Street Journal predicts that investment in AdTech will fall, and according to Adweek, GDPR has posed some new obstacles to advertisers which could also lead to a fall in stock price.

Is this because of this slim margin business model, though? Maybe.

I won’t argue it’s not. But that doesn’t mean that AdTech as a tool is bad. It means the business model needs to be reconstructed. This provides all the more reason for engineers to keep working in AdTech, ensuring the tech itself survives and innovates while business models iterate with time.

I acknowledge this is diagnostic, not prescriptive. Eisenberg himself once suggested, though, that perhaps the placement providers should charge advertisers a flat rate, or create exclusivity deals with publications to increase margins (i.e., Taboola gets all the ad space on Women’s Health and can charge advertisers accordingly for guaranteed space, and won’t have to compete with other ad placement services).

Where’s the real risk for advertisers?

True, the current business model that AdTech operates on is risky. However, we spoke about the tightening grip Facebook and Google are holding on a brands’ ability to communicate with its customers.

If these companies’ ability to entirely eliminate earned attention is not perceived as a risk for advertisers, I don’t know what is.

Facebook and Google’s monopolies of attention pose a far greater of a risk of volatility to every single advertiser than unpredictable publisher traffic in a decentralized placement model.

I think this duopoly risk will stabilize advertiser purchasing for ad placement tech companies, despite potentially unpredictable traffic among individual publications. Increasing rates on Facebook and Google will lead to stable, if not increasing bidding rates in other ad solutions, which will keep the revenue and margins afloat for the ad placement tech companies operating with the traditional bidding and optimization model.

AdTech in the Next Year

If we’re not seeing the results we imagined from these companies just yet (sustainable growth in stock price) it’s not because they’re past their prime or the idea is bad, it’s because the business models pose too much risk on margin (maybe).

But here’s the thing: that duopoly we spoke about earlier will have to be broken up. Facebook and Google are alienating its advertisers as it grows its network (red line in the chart), all at the same time that countless brands are just figuring out how to understand and measure the impact of ads from all of these networks on their bottom line (decentralized ad placement, Google, and Facebook). Brands will realize that cost of attention will be negatively tied to growth of attention and will have to start pouring their money into AdTech solutions, which will offer a more affordable way to communicate with their consumers.

As Thompson concluded in his Stratechery post on Buzzfeed:

What is clear, though, is that the only way to build a thriving business in a space dominated by an Aggregator is to go around them, not to work with them.

Together, this suggests “AdTech” companies are hardly old fruit; in truth, perhaps they’re just fresh off the tree, barely even ripe. The fall of Buzzfeed, as Thompson pointed out, is one indication of how necessary they will be.

So, to the engineers out there, Israeli and otherwise, there’s actually a lot of promise for continued growth in AdTech; for the sake of everyone, don’t give up yet.

This piece was published by Aya Abitbul at Studio 96, a firm that produces content for meaningful brands at scale. Check out our company here.

CEO of Studio 96, a publisher of interactive luxury coffee table books.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store