Digital Planners: Practice 'Darwinian' Optimization

"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 Brian Decker, a managing director at Mindshare, a GroupM agency.

Most people in the digital marketing world wouldn't credit Albert Einstein as a founding father of digital marketing, but I beg to differ.

The famous quote attributed to Einstein, "Insanity is doing the same thing over and over again and expecting different results," has been tattooed on my brain as a constant reminder of the critical importance of testing and dynamic optimizations.

For example, I worked on a very effective local campaign last year for a client in the QSR industry. Several months later we replicated the original media plan partners and tactics based on historical performance, but noticed several significant differences. Some of those differences related to the following:

  • Geography matters. Media partners performed differently market by market.
  • Testing different types of media buys (i.e., CPM versus dynamic CPM versus CPA versus CPC) drove different results.
  • National digital media weight had a direct impact on local performance.
  • Geotargeting will typically drive stronger DR performance than local online media, including local digital newspapers.

Just because a strategy, tactic or media partner worked last year or even last month doesn't guarantee the same results.  Bottom line is that "complacency just doesn't work" in dynamic media ecosystems.

Digital media is dynamic and we need to be constantly focused, proactive and fluid as we build effective media investment portfolios that reflect the rapid changes that are happening in social, hyper local, mobile, video, display, multi-cultural and cloud-based platforms. In fact this necessitates a more dynamic approach to targeting, leveraging data, measuring attribution and adaptive or "Darwinian" optimizations.

Darwinian Optimizations

Patience may be considered a virtue in the analog world. In digital, the ability to measure, track and optimize can be the difference between a campaign's success and failure.  For more direct response focused accounts, we typically use a seven-day performance period to evaluate future success.

Within the parameters of testing, failure is always viewed as an opportunity to learn. Therefore it is imperative that "we fail smart and fail quickly." Unfortunately, media partners who cannot adapt or meet our client performance goals are removed from the media plan and typically from future planning consideration.

As a result we are always testing two to six new partners against a wide array of tactics every quarter. From a Darwinian perspective, we have taken a "survival of the fittest" approach when testing and only the strongest survive.

In a recent article, AdExchanger referred to this approach of ours as "tough love." However, this is ultimately not about media.  It is only about the client and meeting their business and marketing objectives.  While maintaining great relationships with our media partners is exceptionally important, it is ultimately about moving rocks and measuring distance for our clients.

With this approach toward active optimization, there is no room for complacency. Once again the choices we make today may be totally irrelevant tomorrow or the day after. This is based on the rapid consumer adoption of individual and connected devices and emerging touch points. But does this invalidate Einstein's principles on repetition and really expecting different results? Or is this just the wrong question (relatively speaking)?

Follow Brian Decker (@briandecker6) and AdExchanger (@adexchanger) on Twitter.

3 Comments

  1. Ironically, the whole reason we need statistics is that often people do the same thing over and over again and they get different results.

    So does this method leave room for re-experimentation? A partner may underperform in one context but then excel under a different set of circumstances (ie, season, product category of brand). I suppose trying the same partner again isn't "insanity" provided one reasons that the conditions have changed and prior performance doesn't guarantee future performance.

    Reply
    • Thanks for the insight, Brian. I am sure, like me, you struggle with the question of ad experience on mobile. All other technical challenges aside, and the rise of programmatic buying and increased targeting in mobile aside, we're mostly still dealing with a pretty sub par ad experience. That needs to change. Curious what you guys are doing at the agency level to help drive that change? (I am a believer that a rising tide lifts all boats :))

      Reply
      • Joanna - I keep hearing about the need for a "better ad experience" and would love an explanation of what that means. Is it the creative? Is it the placement or relevancy? Is it the flow or call to action? Or is it the static "banner ad" medium? As an agency, we want to offer creative solutions and a better experience for clients and users, but struggle to see beyond what is offered by the platforms. I do agree with Brian that these optimization methods need to be readdressed as accounts evolve - brand equity that is building over time and different messaging have all increased success in areas that were previously ineffective. Great article and insights all around - thanks!

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>