“Half the money I spend on advertising is wasted,” John Wanamaker once famously complained. “The trouble is, I don’t know which half.” A self-made millionaire, Wanamaker helped pioneer modern commerce when he launched his flagship department store in the late 19th century. He was among the earliest generation of marketers to understand the value of advertising, but he remained bedeviled by challenges that seem all too familiar more than 150 years later.
How do you know that your ads are reaching their intended audience? Or that your audience is actually seeing those ads?
Today, many advertisers continue to share Wanamaker’s concern that a large portion of their budget is wasted. They just can’t pinpoint which part. The problem is all the greater in programmatic advertising, where an outsized portion of budgets flow to middlemen before they’re even put to mysterious work.
In Wanamaker’s day, marketers had to guess which consumers might be receptive to their message and blanket newspapers and magazines with creative in the vain hope that it would reach the right audience. Later, in the 1930s, cognitive psychologists like George Gallup, a professor at Drake University who later started Young & Rubicam’s first research department, and Paul Cherington, professor of marketing at the Harvard Business School who held the same research position at J. Walter Thompson, developed the first models to identify specific consumer demographics.
But the old dilemma remained. You still had to buy a lot of print—and later broadcast—media to reach your audience. You could reach a segment, say, middle-class men over 35 who owned their own homes and held managerial positions, but not a specific person. You couldn’t guarantee that people in that segment would actually view your ad on page 58 in Time magazine or the 30-second spot you purchased on Wide World of Sports. Like Wanamaker, you knew you were wasting a lot of money, but you couldn’t nail down the specifics.
Digital is still wasteful
The digital age was supposed to solve that dilemma. The power of data-driven decisioning and machine learning was supposed to make advertising a pinpoint-precision exercise. But for the most part, it hasn’t.
Marketers are rightly concerned that programmatic advertising is a crapshoot; they can’t easily use their CRM data to advertise directly to their target audience. Roughly half of all ad units are un-viewable (which is to say, they fall outside the screen frame). And a surfeit of technology middlemen offering services of questionable value takes a massive bite out of the ad dollar.
Little wonder, then, that approximately 90 percent of incremental digital advertising dollars flow to two platforms: Facebook and Google. Marketers buying ads on the Facebook feed, Google Search and YouTube can leverage their CRM data to engage specific people in a targeted conversation. They pay only for ads that are viewed. And unless they choose to use intermediary technology providers, they know where their ad dollars land (that is, directly with Facebook or Google).
But there are problems. Facebook and YouTube rely largely on user-generated content (UGC). And that content is often of very low quality or worse. It’s a daunting task to screen out hate speech, graphic violence or fake news in real time, which means that marketers incur massive brand safety risks. Likewise, concentrating ad dollars on just two platforms is constricting. There is only so much real estate on the Facebook news feed. There are only so many cat videos worth advertising on.
Solving the targeting problem
What if you could advertise with the same precision and accountability—engage specific people and pay only when they viewed your ads—on several thousand quality domains and apps that offer superior, brand-safe content?
What if there was a technology platform that could achieve these same outcomes on the highest-quality media properties, without the risks associated with UGC?
Finally, what if you could see how your ad dollar got spent? That is, what if that platform deployed block chain and other emergent technologies to document both its own transaction fees and that of every other technology provider that delivered the ad?
AppNexus has been building for that future. Three years ago, we dedicated a team of over 200 engineers, product managers and data scientists to the task of building a technology platform that would enable marketers to advertise to specific people—not types of people (which is to say, cookies)—and to pay only when those ads were viewable.
We want to make it possible to achieve the same precision and accountability previously only available on Facebook or on Google Search and YouTube, but on high-quality sites and domains where consumers spend the bulk of their time.
We want to harness the power of machine learning to solve problems that marketers and their technology partners have been grappling with since the dawn of modern advertising.
We’re almost ready for the big reveal. On Nov. 8, at our New York Summit, we plan to introduce the next generation of buy-side technology. Details to come.
Don’t waste half of what you spend on advertising. Make every dollar count.