Forget Impressions: Digital Advertising’s New Math For The New Year

derekhuyster"Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Derek Huyser, head of business development at AdYapper.

What is it about money that makes people lose all connection to reality? We’ve all heard the stories of lottery winners going bankrupt. Many perfectly normal people suddenly forget how to do simple math and think that their money is infinite.

It turns out that digital media professionals are not immune to this affliction. The cause of the problem is an obsession with impression volume.

There is a particular equation that governs digital media: budget/CPM = impressions.

This equation encourages a focus on increasing impression volume. It tricks people into thinking that if a budget is high enough and CPM, or cost per thousand impressions, can be brought down low enough, impressions can go up seemingly forever. We have forgotten the meaning of the word “impression”: It should be an ad that was actually seen by a real person.

With the focus on creating more and more impressions, everyone can justify spending more and more of the budget. Here’s a typical example: Imagine you have a major retail chain looking to spend $5 million to reach women in the 20 states where they have stores. When the media planner sends out the RFP, the goal is to spend the $5 million with the lowest average CPM possible to maximize impressions. Publishers all clamor for the budget, offering a seemingly infinite supply of impressions at low prices.

The problem? It ignores the fact that there are not actually an infinite number of people that will be exposed to the ad. By focusing on increasing impression volume, advertisers and vendors often lose sight of the natural constraints reach and frequency set on volume. How many times have marketers been part of an advertising campaign that targeted a certain audience only to hear that the audience was “too small” or that the target frequency was “too tight” and that the budget had to be spent no matter what?

What’s worse, the obsession with forever increasing impressions causes the market to allow in fraud, bots and nonviewable advertising, not to mention low-quality placements.

In advertising, most channels deal with reach (how many people) and frequency (how often they’re contacted). Yet in digital, these metrics have been abandoned for impressions. Over the past few years, technology has improved to the point that advertisers can actually target the same individual across publishers and channels, and that reach and frequency metrics are reasonably accurate. Companies, ranging from demand-side platforms to ad servers and viewability vendors, can all help manage reach and frequency.

I think it’s time that we bring reach and frequency back to the forefront of digital advertising with a new equation: Viewable reach x viewable frequency = impressions.

With this new equation, planners would work out the correct reach and frequency and then build a plan that satisfied the right number of impressions at whatever cost that campaign might be. The impression number would be governed by real-world limitations. This would allow the planner to accept a very obvious but somehow forgotten reality that there are only so many people out there in the world, there are only so many good impressions and they should really only see an ad a certain number of optimal times.

Imagine the new improved scenario: The planner’s primary goal now is to calculate the total number of women in those 20 states (let’s say it’s 20 million), learn the optimal frequency (let’s say it’s two per day), and build a plan around showing those 20 million women two effective messages per day. That’s about 40 million impressions per day. If it costs $5 million, fine. If it costs something else, well that’s fine, too. But note that there’s never a reason to show more than 40 million impressions per day.

With the old equation, in comparison, that $5 million budget may be divided to find the number of impressions needed for a $4 CPM, for example. This means you have 1.25 billion impressions to work with. With these three numbers already decided, reach and frequency become an afterthought, and the focus shifts to negotiation for more impressions by decreasing CPM, which creates the risk of lower-quality impressions.

Advertisers should think twice about high-volume but low-quality traffic. Instead, they should care about quality impressions and conversion rate – a combination of the new impression formula plus conversion or performance metrics that can be validated internally – to squeeze more performance out of the impressions they know they can buy. Rather than spending more budget than is necessary, they could have leftover budget that may be reinvested in other ways.

At a minimum, this kind of thinking would eliminate vast amounts of waste from volume-driven ad budgets. Some advertisers have already altered their practices to focus on people instead of impressions. Proctor & Gamble’s well known “Hawkeye” project focuses around a database of unique people to target.

By focusing on true, quality impressions – ads seen by actual people the right number of times, tied to performance – advertisers can increase quality instead of quantity and boost return on ad spend, which should be the goal. This is a good thing.

By knowing their audience and spending only what they need to in order to maximize their engagement with that audience, advertisers increase performance and eliminate waste.

Follow AdYapper (@AdYapper.com) and AdExchanger (@adexchanger) on Twitter.

3 Comments

  1. Totally agree. We will continue to see a shift to a more performance oriented mindset, even amongst "branding" and "awareness" budgets, as the industry evolves in 2016 and beyond. Impression delivery should never be a campaign goal.

    Reply
  2. Good explanation of how the "tail has wagged the dog" in widespread thinking that volume of impressions = impact. Backing out metrics from campaign performance and ROAS is far more valuable in meeting client goals.

    Reply

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>