The Kingdom Of The Last Click Has Fallen -- It's Time To Calm The Chaos

by , Jun 19, 2012, 10:00 AM
  • Comment (5)
  • Recommend (11)
Subscribe to Metrics Insider

Last click, as a measure for the performance of digital media, was king for most of the Internet era. Now that its reign is over, the industry must work together to ensure a streamlined attribution ecosystem.  The accurate valuation of media is essential to the health of the digital advertising marketplace, as is a more efficient media planning and buying process.  Our industry has the chance to seize the great opportunity of this new era by facing its immense challenges now, in the early stages of its development.

For at least a decade, the last click stood as the default unit of measure for tracking how well various elements of a digital marketing campaign contributed toward campaign goals. Whatever a user clicked on to go the advertiser’s site -- a link, banner, search result, etc. -- got all the credit. A boon for some, an albatross for others, without systems to gain more valuable insights, this overly simplistic practice was the standard.

Until now. Recent advances in data processing and analytic capabilities have provided marketers with powerful tools for improving the attribution of value in a digital marketing campaign.  No longer are they forced to use the last click. It's now possible to track and analyze the wide range of marketing influences and user actions that contribute toward campaign goals, whether they are direct-response or brand-oriented. This evolution represents important progress for the digital advertising marketplace.

However, change is messy. The rule of the last click has been replaced by pandemonium, by competing forces that are no less confusing and intricate than the fight for the iron throne in HBO’s "Game of Thrones." New vendors, models, and techniques for calculating attribution mixed with today's data-driven, hyper-optimized advertising ecosystem have led to confusion and tension in the marketplace. Factors contributing to the chaos include:

Competing and conflicting discrete data sets. Marketers, publishers, exchanges, and vendors each perform attribution calculations using their own proprietary data of user events -- and everyone, naturally, is inclined to defend the validity of their own work. This turbulence is similar to the impression-discrepancy issue, only augmented by a lack of universal terminology, as well as the complexity and variety of calculations.

An absence of a Pythagorean Theorem equivalent. When it comes to attribution, it's unlikely there will ever be, or should be, standardized relationships between the values of channels. For one campaign and its particular goals, it may be reasonable to weigh display ad clicks higher than Twitter interactions; for another campaign, it might be appropriate to give more weight to specific user actions that signal engagement. A vendor that specializes in attribution modeling must tailor its solution to individual campaigns -- not every marketer, but every campaign -- as each has its own unique media mix and objectives. This level of customization complicates the attribution process, but must be preserved. Efficiencies should stem from better enabling this level of sophistication.

How hard it is to talk about attribution.  The growing attribution ecosystem is lacking a universal lexicon. This lack of definitions makes it terribly hard for marketers to compare and contrast various models of attribution calculation, and to confidently choose or devise the one that’s right for their specific campaign. When buyers cannot quickly understand what an attribution model provides, their decision-making slows, their certainty in transactions is compromised, and ultimately their dollars go to media that provides an easier-to-understand return on investment.

The development of a common language -- and, even more simply, defining what the key common characteristics of attribution calculations are across the ecosystem --  is the most critical step to enabling attribution capabilities that are both efficient and amply sophisticated. We must have the means to compare and contrast methods of valuation in order to truly take advantage of this wealth of models and insight into user behaviors.

At the IAB Advertising Technology Marketplace on June 21 in New York City, we will dive into this issue and reveal information that we hope will lay the groundwork for a more transparent and systemized attribution landscape. This is your opportunity to be at the forefront of progress that will lead us, as an industry, toward the more effective use of advertising dollars, which inevitably brings with it higher CPMs and increased ad revenue. When a marketer can identify what’s working and optimize their campaigns accordingly, they’re going to spend more.

5 comments on "The Kingdom Of The Last Click Has Fallen -- It's Time To Calm The Chaos".

  1. Media Maven from NA
    commented on: June 19, 2012 at 10:33 a.m.
    Patrick, I think you definitely need to explore the effect of Do Not Track and plugins such as ghostery to attribution modeling. What happens when we can't follow a users behavior? Could this be force that pushes the industry from attribution modeling and performance based display campaigns towards more brand-centric campaigns?
  2. Bob Fetter from Pluris Marketing
    commented on: June 19, 2012 at 11:04 a.m.
    Good stuff Patrick. I find that any allocation scheme is in effect a self-fulfilling prophesy unless backed by intensive modeling and analytics. I wrote this back in 2007 when the situation was a bit less complex than today. http://multichannelmerchant.com/lists/matchbacks_04092007/
  3. Armen Avedissian from Convertro
    commented on: June 21, 2012 at 10:07 a.m.
    We could not agree more with this article, at Convertro(www.convertro.com) we have created a solution suite to exactly cater to these online last-click issues with multi-attribution and the offline cross-platform needs for clients and agencies with consistent metrics for Online and Television; the following articles may be of interest - VentureBeat: http://venturebeat.com/2012/04/09/pinterest-drives-more-revenue-per-click-than-twitter-or-facebook/ MediaPost: http://www.mediapost.com/publications/article/176315/the-cookie-stuffing-controversy.html Yahoo Finance: http://finance.yahoo.com/news/according-convertro-television-spending-boosts-210000042.html
  4. Steve Latham from Encore Media Metrics
    commented on: June 21, 2012 at 12:41 p.m.
    Good note Patrick. Compounding the problem is confusion about who should be leading the charge - the agency or the advertiser? Media measurement has historically been the responsibility of the agency, but most shops are overworked and understaffed with limited capacity to figure this out. The Advertisers should be more involved in defining requirements and how the data will be captured and used; after all it is their data. Hopefully this will be the year both sides become more engaged and work together to evolve the way they measure media. Here's a post on same topic on iMedia http://bit.ly/tyjrWk.
  5. Domenico Tassone from Viewthrough Measurement Consortium
    commented on: June 22, 2012 at 1:48 p.m.
    I am certainly no fan of clicks, but we keep hearing this "click is dead" over and over again. It is great that IAB is addressing this issue but attribution is not the low-hanging fruit here - in fact, it is very complex and controversial. When will the IAB do something as simple as put forth a definition for viewthrough?

Leave a Comment

Sign in to leave a comment. Don't have an account? Join Now

Recent Metrics Insider Articles

» Metrics Insider Archives