Why Programmatic Should Be Kicked to the Curb

By Dr. Augustine Fou – Independent Ad Fraud Researcher

Programmatic advertising is a great way to scale your ad campaigns quickly and easily.

Yes, that is if you don’t care that the scale is not “human” scale. Real human audiences are finite; the amount of time they spent online consuming content is finite. The “scale” of ads in programmatic have been completely out of whack from reality for the last decade.

It’s all fake scale. Note the “loudest alarm clock app” that no one has heard of before, is selling 3.2 billion impressions per month (how many minutes do humans use alarm clock apps per day?).

Those quantities are larger than ESPN, and nearly as large as Hulu and Spotify.

Programmatic ad buying is one of the most efficient ways to reach your target market.

Adtech vendors and agencies have deliberately or ignorantly foisted the term “cost efficiency” on unsuspecting advertisers, causing them to think that “we got such a great deal in programmatic because the CPM prices are lower.” That is flawed thinking. Yes, you used to pay US$30 CPMs to real publishers, now you pay $3 CPMs in programmatic.

But you are buying 10X more quantity now, so you are still spending $30 total ($3 CPM x 10X quantity). You did NOT save any money and you did not get a good deal. In fact you got a crappy deal because the extra 90% of impressions you are buying now are all faked. The large quantities of ads, the low CPM prices, and the high clicks are all driven by bot activity.

Fake sites and fake mobile apps can sell enormous quantities of ads at really low CPM prices because, unlike real publishers, fake sites have no costs of content.
And no scruples will hold them back from loading 6,562 ads and trackers on a single page to maximize profits.

In the world of online advertising, data is king.

Yes, but more data doesn’t mean people understand it or use it, or use it correctly. And it doesn’t mean the data is accurate. Furthermore, bad guys are actively falsifying the data, so you must have a way to determine what data is real vs faked in the first place.

Remember the Uber lawsuit where they are suing 100 mobile exchanges for falsifying the data and claiming credit for app installs that had already occurred? And see the video below where false data is written directly into Google Analytics. The traffic didn’t occur; but over 10,000 simultaneous visitors appear to be on the site at once.