How to Make a Bad Impression

This post originally ran in Ad Age. The image above is the rate card from QQTube.

How to Make a Bad Impression

When it comes to digital marketing, you always get a second chance to make an impression. Impressions are cheap, and way too many buyers and sellers still rely on them as currency.

While this has been an issue for decades, marketers can’t stop their impression addiction. This is partly because impressions (and equivalent metrics, such as video views) are a unifying metric across disparate tactics such as display advertising, video advertising, content marketing, influencer marketing, and public relations. Meanwhile, branded content tends to get little distribution without paid media amplification. The easiest way to promote content is to buy impressions, so reach becomes a goal in and of itself.

This racket needs to stop. Impressions mean absolutely nothing. Here are a few of the more salient reasons why:

  • There’s a reason why reach and frequency are the chocolate and peanut butter of traditional media: frequency provides at least a hint of context as to whether an audience is overexposed or underexposed to a message. And yet it’s all too common to see pitches and reports stating something like, “Kevin has a million Twitter followers,” “Josie got 10,000 views on her Instagram post,” or “This native ad was viewed 10 million times.” Often, there is no further context beyond some version of the impression count. Frequency isn’t perfect, but it’s a start, and online, there is no standard as to what that context should be.
  • Fraud is rampant. Imperva Incapsula reported that 29% of all online traffic in 2015 came from malicious bots, while Distil Networks puts the figure at 19%.
  • Bad bots are getting smarter. Wired’s May 2015 issue describes how Trump supporters (if not the campaign itself) are using bots to influence the election. Wired writes, “We’ve caught bots disseminating lies, attacking people, and poisoning conversations.” Impersonating real-looking voters is harder than impersonating anonymous web visitors.
  • Fraud can take subtle forms. For instance, consider a Twitter user who is a real person and wants to come off as a more popular influencer, perhaps to attract sponsorships. Such a user can run bots that constantly follow others, and then unfollow anyone who doesn’t reciprocate. All this user cares about is boosting impressions. Here, the action may be banal, but the intent isn’t, and marketers who aren’t careful can wind up paying a lot of money to reach people who only influence Twitter bots but not human beings.
  • All kinds of impressions can be bought. For evidence, below is a case study on how cheap and easy it is to inflate an impression report.

Consider the YouTube video “[Zombie Highway 2] A pretty good ride.” It was published in November 2014, and as of May 13, it had tallied all of four views. All four were probably from the content creator checking out his own YouTube account (I can assume this because it’s my account). Several days later, it had more than 10,000 views, and within a couple weeks, the view count will likely top 100,000. The total cost of all this activity was $60, purchased from QQTube which calls itself “the largest supplier of YouTube views in the industry.” They add, “We make it easy to simply go viral.”

It’s easy to “simply go viral” when all you care about are impressions. Anything can be made to look like it went viral. What’s telling is that having tested QQTube on other pointless videos from personal accounts (I’ve never used this for any client work), these views never deliver any meaningful results. They don’t even generate likes or shares, so those need to be bought separately. Want to cover your tracks? Buy some dislikes, which cost the same as likes at $9.60 per 1,000.

On average, each viewer watched less than five percent of my video. Russia, Argentina, Vietnam, and Indonesia are top viewer sources. I can’t imagine many marketers are targeting Russians, Argentinians, and Indonesians at the same exact time with the same piece of content, so most marketers would be displeased with such a result. Some others may prefer blissful ignorance and enjoy their inflated report.

There is one simple way to pop this impression bubble: demand better. Demand context. Demand accountability. Demand to know how some impression number could possibly address any of your goals. Demand that greater efforts are made at delivering successful results than the efforts fraudulent bot operators wage at undermining your efforts. Demand more by asking “who” and “how” and “why.” Demand another option. Impressions alone aren’t worth anything,