Digital advertising’s perverse incentives
Contributor Peter Minnium explains how selecting the wrong metrics in digital advertising can lead to ‘perverse incentives.’ To succeed, marketers need to know the metrics that matter.
In digital advertising, as in life, it’s important to start with the desired end in mind.
There’s a well-worn tale about a city slicker who gets lost while driving the back-country roads of Maine en route to the capital city. After driving for what seems like an eternity on unmarked roads, he comes upon a farmer on a tractor in his field. Relieved, he hurries to the local and asks, “How do I get to Augusta?” The farmer, after sizing the man up, responds laconically, “Well, I wouldn’t start from here.”
It is imperative for success in digital advertising to start off from the right place — with a clear focus on the metrics that matter.
Yes, technology incited an explosion of content, devices, formats and opportunities to reach people, together with a cavernous quantity of data and a mind-boggling suite of metrics. Marketers can easily feel lost in unfamiliar territory like our Google Maps-less city slicker, struggling to adapt to a constantly changing environment and forced to choose — often inopportunely — from a plethora of potential navigation strategies.
Although studies repeatedly negate the link between sales and click-through-rates, the top five ways that native ads are evaluated are engagement rate, conversion rate, click-through rate, cost per conversion and quantity of interactions, says eMarketer.
While otherwise informative, each of these short-term behavioral metrics reveals little about how communications have affected brand perceptions, equity or any other measure of long-term brand health. This isn’t only true of mobile; we see this morass of metrics across many digital campaigns.
Perverse incentives
Youngme Moon, in her book, “Different: Escaping the Competitive Herd,” postulates that “the minute we choose to measure something, we are essentially choosing to aspire to it. A metric, in other words, creates a pointer in a direction. … In track and field, we happen to measure speed, and so we cultivate a nation of speedsters. If we happened to measure running style, we would cultivate a nation of gazelles.”
The metrics for success that we set are important because they determine how we go about achieving it. Choosing the wrong metrics can lead to the opposite of the intended outcome — a phenomenon known as “perverse incentives.”
For example, doctors are financially incentivized to order more procedures, whether or not they help the patient; investment bankers take huge risks to reap bonuses, then find new jobs before their clients feel the pain; and politicians borrow money for policies that will make them look good in the short term for re-election, even though it will put their constituents on bad financial footing later.
The potential for perverse incentives in digital marketing is strong. Marketers have 197 digital metrics with which to measure performance (according to the Advertising Research Foundation’s “The Digital Metrics Field Guide”) and over 30 different metrics that can be used to measure engagement, according to the Interactive Advertising Bureau.
We see the outcomes of digital marketing’s perverse incentives everywhere: click-bait headlines; scads of inventory “below the fold” that count for impression delivery; autoplay video in the margins of web pages that yield “completes;” ads that look like content in a stream for attention; and interruptive interstitials that require interaction to close.
While the sell side of the industry has woken up to the perils of perverse incentives, with efforts like the Coalition for Better Ads, the real driver of change will come from marketers who clearly articulate objectives that drive the right behaviors from the start.
Setting the course
It is crucial to first distinguish between diagnostics and outcomes. Most short-term behavioral metrics are diagnostics that reveal how viewers engage with an ad. Outcomes reveal the extent to which the ad affected the brand. Relying too heavily on behavioral measures can risk optimizing to the wrong metrics.
Ipsos Connect (my employer) recently studied two ads for a client where ad #1 had a much longer average view time, but ad #2 delivered 150 percent higher brand consideration. Previously, the marketer had optimized spend to support ads with the greatest view time, assuming that people preferred them. When research proved the ads had less impact than presumed, moving resources to ad #2 brought more benefit to their brand objectives.
With this in mind, marketers should clearly set out where they want their brand to get to and the role of communications in the journey. This means determining answers these questions:
- What attitudes or behaviors do you wish to influence?
- Over what time frame?
- Which metrics demonstrate their success?
If the goal is to increase brand awareness, behavioral measures like impressions, reach and views are important. Marketers also need to understand how the ads impact spontaneous awareness via brand-linked memories and recognition of the campaign. If the objective is to drive short-term sales, the focus would be on conversions, site or store visits and short-term purchase intent.
Metrics that matter
The biggest challenge brands face today is getting noticed amid the clutter. Consumers are exposed to a seemingly infinite amount of content; how can brands improve the chances that their communications will gain attention? Three principles can help:
1. Mandate viewability
Ads can’t have an impact if they aren’t seen. As obvious as that sounds, viewability is still a big issue. Just 52.9 percent of desktop ads meet the IAB’s viewability standards, Moat’s CEO told Advertising Week attendees last year, according to MediaPost, and only 41.4 percent of mobile ads achieve it.
It’s generally accepted that ads get more attention on less cluttered sites, viewability is likely to be better with premium publishers vs. ad networks and exchanges, and time-in-view is more important in driving brand impact than the percent of ad in view.
2. Prioritize brand impact
Digital is no longer a direct response-only medium. With brand-building advertising, it is important to measure more than short-term responses: Brand impact is key. This means being sure that viewers not only pay attention and spend time with the ads but that they also give credit to the brand.
According to Google, viewers are more likely to skip a YouTube ad if the brand is present within the first five seconds — but ads with branding in the first five seconds are more likely to drive positive brand impact. This is important for advertisers who judge ads based only on view-through rates and completes.
The ideal is to make gripping ads with the branding inherent — well integrated into the ad — so that viewers will be compelled to watch and give credit to the brand.
Measurement should be holistic, with both behaviors and attitudes measured and priority given to metrics that evaluate against the brand objectives.
3. Optimize to context
Different journeys require different tactics. Ads that are effective on TV don’t necessarily perform well online. Ads that work in mobile may be so-so performers on desktop. Mobile ads that are winners on Facebook may lose on Twitter. Digital environments are unique and should be treated as such.
‘Well begun is half done’
These words, spoken by Aristotle over 2,000 years ago, are still true today, and particularly so in the digital advertising world. Start your digital journey with a clearly drawn map of the outcomes you wish to achieve and the measures to gauge your progress, and you will be well on your way out of the woods and en route to success.
Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.
Marketing Land – Internet Marketing News, Strategies & Tips
(52)
