Why You Can’t Just Buy a Marketing Attribution Model

Matt Hertig Alight Insights, Analytics for Marketers, Marketing Measurement Leave a Comment

Multi-Channel Attribution has become a very hot phrase over the past few years. Like many buzz terms, marketers understand the concept behind multi-channel attribution but don’t understand how to effectively implement this strategy.

If you aren’t quite familiar with this term yet, the idea behind multi-channel attribution has become very popular in the digital marketing realm by looking beyond “last-click” metrics that are the standard in virtually all web analytics tools – including our preferred platform Google Analytics.  In order to understand the total value of a marketing channel, you must understand how the channel falls into the following conversion categories:

  • “Introducer” (First Interaction) – this is the marketing channel utilized by the visitor to first interact with your product, brand or service.
  • “Assistor” (Second to Second-to-last Interaction) – these are the marketing channels used by that same visitor, usually in a 30 – 90 day window, but aren’t associated to final conversion event. 
  • “Closer” (Last Interaction) – this channel is used by the visitor when taking the conversion action established on your website such as a transaction, appointment, lead, etc.

For example, digital banner ads have historically not performed well as a “closer.” While some banner ads do create a last-click conversion event, banner ads are typically top performing as “introducers.” By their nature and purpose the value of banner ads is to build the top of your funnel whereas other channels such as paid search are more likely to be “closers”.

Unfortunately, a multi-channel attribution model isn’t as easy as ordering your favorite latte. In order to construct a multi-channel attribution model, a data scientist combines the behavior of all marketing channels within their respective conversion categories into a model. This complex model building process is designed to articulate the influence and direct impact each marketing channel is delivering. This new insight moves media buying decisions beyond just a simple last interaction to an investment strategy based on how marketing channels are working together to drive conversion results.

While this sounds fairly straight forward, the underpinning of an actionable multi-channel attribution strategy is dependent on an accurate and robust web analytics implementation. Without this, you aren’t able to clearly answer basic performance questions let alone build a multi-channel attribution model.

Multi-Channel Attribution is the future standard of how performance will be measured within marketing across all channels – both online and offline. A limited number of organizations are ready for this step as they have the fundamental data building blocks in place to support this strategy – most however aren’t there yet.

Be sure you get your tracking and measuring strategies right before jumping into modeling and remember there is more to attribution than last click.

Leave a Reply