Newspapers, we all know, failed to exploit the great original promise of digital advertising. So when the 2015 numbers are finally in, you can expect last year to have been the worst ever in their great decline – by all accounts advertising is down sharply across the board. But they’re not the only ones. Ironically, digital media companies have failed to exploit the promise of digital advertising too. That’s why a long overdue shakeout is coming.
It’s coming not because of a glut of content, as several journalists in the commentariat have suggested lately, inadvertently offering further evidence that if you flout your own innumeracy and have never sold anything to anybody, you’re probably not worth taking seriously. No, it’s not a glut of content that’s driving the shakeout, it’s a glut of undifferentiated content – and even more important, a glut of undifferentiated advertising.
I mean Buzzfeed and Gawker may approach the tactical task of building distribution differently, but their content and positioning is pretty much the same. How many undifferentiated products targeting a young demo with a snarky voice does the world need? And the top-down, bot-driven, programmatic placement of commodity banners might be an inexpensive way to place dead inventory across the Internet and make a dime or two. But it has reduced digital impressions to a low-price commodity and it further undermines the perceived marketing and advertising value of digital.
Over the years, every research study I’ve seen that examines consumer attitudes towards advertising concludes that consumers hate it – but do they really? If you’re not in the market for a car and see, hear or read an auto ad, it’s an irritating intrusion. But if you happen to be looking for a car, it’s exactly what you want.
That’s why we were so excited, back in 1995, when all this began. We thought the age of mass media was ending, the age of personal media beginning. My news, my entertainment, my advertising, delivered in the style I want, sufficiently relevant and useful for me to pay for or subsidized by advertising especially relevant to…me. Yes, me. Right now. Right here. Just as the great Peter Drucker had predicted in the 1970s, when he described the perfect advertisement as one of which the consumer could say, “this is for me, and me alone.”
Direct marketing thought leader Lester Wunderman had seen it coming too. Because the baby boom generation “demands attention to their unique differences rather than their similarities,” Wunderman had written, “we would one day end up developing technologies that would allow us to achieve the marketing miracle of selling the right thing to the right person at the right time.”
Until 1995, when it all began, media sales people had made a really good living by exploiting a highly inefficient advertising marketplace. The mass media business was based on the illusion that audience was somehow quantifiable, but the truth was that no advertiser had any way of knowing how well his message had worked or even, in many cases, if it had worked at all. An empirical artifice of cost-per-thousand and gross ratings point was conjured up simply to give the advertiser consolation, and he swallowed it because he had few alternatives. Newspaper sections, tv programs and channels, and radio formats each offer a crude form of segmentation, but that’s all, they are blunt advertising instruments. Magazines also offer a degree of audience selectivity, but to reach those Sports Illustrated readers who happen to be interested in buying a car this month, the auto advertiser still has to buy an ad that 95% of readers will flip right by.
“Half the money I spend on advertising is wasted; the trouble is I don’t know which half,” department store visionary John Wanamaker is supposed to have said in the early 1900s.
Then along came direct mail. The business of advertising evolved from blasting out a hopeful advertising message to a large, amorphous, unqualified mass market a thousand eyeballs at a time to targeting a specific appeal to a pre-qualified audience segment whose behavioral and purchase response was predictable and measurable. RFM – recency, frequency and monetary value – supplanted CPM. The growth of direct mail was steady but it wasn’t spectacular because of the cost of developing, producing and delivering high-end print creative – the bulk of direct mail consisted of simple flyers and coupons. But the rational argument of targeted, measurable reach was sufficient to drive consistent growth over the long term despite these creative shortcomings. Performance and results seemed to matter more to advertisers than poetry, especially to local advertisers, who in the aggregate spend most of the money. In the right environment, they were even willing to pay a premium for it.
As the Internet arrived, direct mail was steadily closing in on newspaper advertising share. The exponential growth in computer processing power was accelerating its impact. With the marketing database it was like looking at a market through a microscope with continually improving magnification. The market fragments had been there all the time, there was just no way to see them before and no way to access to access them efficiently.
To those of us there at the outset, the Internet looked like direct marketing on steroids. It promised all advertisers, not just classic direct mail advertisers, the two golden keys to advertising efficiency: Targeting, and measurement. At last, the advertiser had the tools to hold media accountable for advertising performance. Commercial search was the first manifestation of accountability. Advertisers now knew who clicked on an ad and what they did when they got there. In search, cost per click and click-through rate quickly supplanted CPM and GRP. And that scared the shit out of media sales people.
Mel Karmazin, the former head of CBS, put it best when in 2003 he said:
“You buy a commercial in the Super Bowl and you had no idea it worked. You had no idea if you sold product…I loved that model. But Google? Google is fucking with the magic.”
Media sales people thought CPC was just fine for direct response advertisers and the price-and-item advertisers that the newspaper retail franchise was built on. But they were terrified of the implications it held for their core business. If they transparently eliminated all the waste in their impressions-based brand advertising products, the bottom would fall out of their business. Broad gauge CPM will still work just fine, they told their clients. It will have to.
And amazingly, advertisers bought it, proving once again that there’s nobody more conservative than an advertiser, particularly a CPG brand marketer responsible for lines of business generating millions in profit on razor-thin margins and unwilling to risk anything new.
Lots of people and many companies – Google, Facebook, Twitter, Vox Media, AOL Sales – keep trying to establish different ad formats and different ways to measure success. As a result, any buy a digital brand manager makes is structured on slightly different terms and audience definition. But fundamentally, the CPM and GRP of broadcast television are still, together, the de facto standard for measuring ad reach in digital, the most measurable medium yet. Go figure.
Digital tonnage prevails. The unique opportunity to deliver highly customized and quantified audiences has been squandered. Only gigantic platforms like Google can hope to win – they have all the leverage because they own all the eyeballs. I know how that works, once upon a time newspapers had that kind of pricing dominance, down in their local markets. At Google, just 30 big companies – Apple, Microsoft, Nike and the like – contributed $3.5 billion in advertising revenue last year – and that line is growing at 30% a year.
In a media age where the laser-guided missile has supplanted the carpet bomb, when audiences can be segmented into finer and finer slivers, when propensity to buy can be narrowed in time to less than a week, when messages can be crafted based on mines of data unimaginable just a decade ago, and when an endless data feedback loop can guide real-time refinement of advertising messages, we find ourselves with digital media products that still rely for direction on little more than an editor sticking his finger in the air and on primitive advertising messages still anchored on how much it will cost an advertiser to reach a thousand complete strangers at a time.
What are the lessons here? First, the digital media products that are doing well are highly differentiated from their competition and developing segmented audiences in a disciplined way. I’ve written about this before. Click: The digital differentiation challenge The tighter the segmentation, the better the demographic qualification, the higher the value – the more an advertiser has to pay for access. Vox Media, with blogs like Vox News, SB Nation (sports), The Verge (culture), Eater (food), Racked (fashion), Curbed (home), Re/code (technology) and others, has become the archetype. Click: Smart media brands for a new generation Vox has re-imagined the classic broad newspaper product mix, rebundling it into a network of independent but mutually-supportive vertical blogs for segmented audiences. The network now attracts 170 million unique users a month. And it’s making money. Not a lot. But it’s profitable, and that’s saying something.
And it’s proof that when selling CPM in digital, it’s important to ask cost-per-thousand-of-what, exactly? Of passers-by? Of prospects? Of past customers? Of the competitor’s customers? Of users identified and targeted against particular criteria? Of what?
Second, the marketing database which captures and cross-hatches user data is at least as important as the front-end development of the product itself. At AutoTrader.com, the marketing database is so well developed that ATC sales people know who in town is in the market for a car this week. That’s some sales tool. Data like that drove revenues in excess of $2.3 billion in 2015. It is astonishing that so many digital media companies know so little about their users. For some, there’s too much data and not enough insight. For others, they have marketing databases that are no more revealing or insightful than the kludgy print circulation databases still used by newspapers.
Third, and this I know is a crude generalization, Internet sales people are, as a rule, poorly trained and poorly led. That’s the main reason why the per view price of a digital ad continues to drop and ever-more ad dollars are concentrated with Google and Facebook. Digital sales people are doing a dreadful job of pitching – and holding – value.
Part of that is due to what they have to sell. Digital has failed to creatively deliver the allure and pull that brand advertisers need. This is not to say that banners are not effective – it’s important to remember that newspapers until 2005 were the largest ad medium in the world and they got there by selling lines of agate type and primitive rectangular space. It is to say that banners are merely one step in a larger presentation sequence that over time leads a prospect through the channel all to way to affinity and purchase.
Video is becoming increasingly important in that process. Television advertising was rather like breaking and entering, where the advertiser suddenly jumped in front of the viewer and yelled to attract attention. That worked when there were only three TV networks and one bathroom. Today there are tens of thousands of competing channels, and most of my friends live in houses with two or even three johns. Nobody can be compelled to pay attention online.They need to be enticed. But how do you measure enticement? Google is doing some interesting work on that. Click: You Tube brand lift As you can see, we still have a very long way to go…
And even with the growth of digital video advertising, the point still stands. Many digital sales people I have worked with do not speak with any fluency the language of bound vertical markets, of tightly segmented and highly qualified audiences, of lifetime value. They’re still out there selling unqualified audiences and broad GRP campaigns, still leery of promising empirical, targeted performance.
They may as well be out selling cable television. What they’re doing is really not that much more refined.