The Promise of Digital Has Not Yet Been Fulfilled

Another week brings another publicly documented example of advertisers taking matters in their own hands and actually checking if their digital ad spend is driving real business results. So far 2 for 2 say no. And many other experiments are going on in private. This is good for the whole marketing industry - digital or otherwise.

Without even mentioning digital ad fraud (but I just did), the way many brand marketers have thus far been using digital is sub-optimal. I've said over at least the last decade, and the data bears this out, that digital is mainly a performance channel (as opposed to a branding/awareness channel). And it should be used as such, for optimal outcomes. However, most consumer packaged goods and brand marketers have used it as another branding medium, like TV and print, from which their ad dollars are being shifted.

When the objective of the digital campaign is only "reach and frequency" then the actions tend to be mainly "buy as many impressions as possible, and lower the average costs (as a measure of improving the performance of the ad spend)" when measuring real business impact (although hard) should actually be the real goal. By trying to buy as many impressions as possible, the marketers expose themselves to more ad fraud (there, I mentioned it again) because there are just not enough humans to generate all those trillions of ad impressions. But there are plenty of bots that can generate virtually unlimited impression "inventory" to sell through ad exchanges.

Note that this does not mean that all digital reach campaigns are not "clean." When you need to drive awareness in digital, you should use awareness tactics like display ads and video ads. But always make sure there is fraud detection in place, verification technologies that show proper targeting was achieved, and other "active measures" to reduce fraud and maximize ads shown to humans. And be sure to focus on outcomes that humans do, and bots don't or can't. That is how you ensure that even digital reach campaigns can be clean - human, and viewable. Otherwise these digital campaigns will be useless from a business outcomes perspective.

Revisiting a series of slides I published in 2015 (above), depending on the nature of your product, the industry you are in, and where your customers' missing links are, there is a simple way to "mix optimize" your marketing mix -- by selecting the right channels and tactics. In brief, if you are a "low cost, low consideration" product like soup and soda, or toothpaste, CPG, or toilet paper, then use branding channels and tactics. If you are a mature product whose brand is already well know, you don't need (as much) branding; so you can allocate more dollars to "decision support" tactics like paid search and organic search (content marketing a.k.a. SEO).

If however, your product is more complex or bigger ticket, where humans tend to research it more, then your marketing mix should be more heavily allocated to decision support -- "what are the bits of information they need to move themselves along their own customer journey?" I call these Missing Links(tm). Missing Links are observable (what did they ask about and what did they search for) and addressable (we can take action to help prospective customers to answer missing links, via content or marketing messages). See slides 5 - 10.

So the following 2 public examples of big spending marketers cutting back on digital is a good sign. They will not stop spending in digital, but they may start to spend better. For example, Pampers is already synonymous with diaper (you buy "Pampers" for your baby not "diapers"). So instead of just reach and frequency, the Pampers marketing team can spend more wisely and efficiently in content, search marketing, and other digital tactics that provide efficient decision support for new customers - so they choose Pampers over other competing products. The branding/awareness stuff is still done, but by television, which is its most efficient use.

One more useful real-life example is the following chart. When measuring both the display ads and website (landing pages) for a marketer, we can see the phenomenon known as "end of month traffic/impressions fulfillment" (top part of the chart). Notice the increase in volume (green volume bars) into June 30th. And notice the bottom part of the slide shows NO DIFFERENCE in website traffic; in fact it is a bit lower heading into July 1. So for this marketer, whether the ads were fraudulent or not, it was not productive digital spend. (Yes, for those nit-pickers out there, whether you get any clicks to a website depends on whether there is a call to action or whether it was for branding or performance, etc.) But this is still a good visual for end of the month impressions fulfillment.

July 28, 2017 - AdAge: P&G Slashes Digital Ads by $140M Over Brand Safety. Sales Rise Anyway

July 28, 2017 - WSJ: P&G Cuts More Than $100 Million in ‘Largely Ineffective’ Digital Ads

March 29, 2017 - NYTimes: Chase Had Ads on 400,000 Sites. Then on Just 5,000. Same Results.

About the Author:  “I advise advertisers and publishers on the technical aspects of fighting digital ad fraud and improving the effectiveness and transparency of digital advertising. Using forensic technologies and techniques I help to assess the threat and recommend countermeasures to combat fraud and improve real business outcomes.” 

Follow me here on LinkedIn (click) and on Twitter @acfou (click)

Further reading: https://www.slideshare.net/augustinefou/presentations