With insights from Arnaud Lauga, Data, Technology and Innovation Director, Publicis Media France
View our latest whitepaper on attribution modeling: Multichannel Attribution, Contribution & Optimisation
In partnership with the Electronic Business Group (EBG) Task Force, Performics has developed a whitepaper on Multichannel Attribution, Contribution & Optimisation. In the first chapter, we uncover customized models past the first click attribution model.
Last click attribution is a model that gives credit of a sale to the “last click.” Despite the fact that this model is no longer the most relevant way of determining how consumers purchase, advertisers are still making the mistake of attributing conversions to the final click before the sale. This model is problematic because advertisers aren’t able to understand the motivation behind the consumer’s buying decision. To demonstrate, using the last click attribution model, an advertiser can see if the consumer last clicked an ad or if they clicked through a post on Facebook that led them to the brand’s website. But, this doesn’t allow advertisers to understand the consumer’s intent. Advertisers should be looking for further analysis behind the consumer’s purchase: How many times did the consumer previously visit the website before purchasing? Did the consumer conduct preliminary research among competitors before purchasing? Etc.
From several marketing analysis carried out on H12015, Performics discovered that on average, “it takes 14 marketing actions to create a digital conversion, and more than half of digital conversions are accomplished through a single action.” And while the last click attribution model isn’t always accurate, it shouldn’t be discounted as it still provides value to consumers.
Additionally, brands measure “performance” differently, depending on their end goal. And because this measurement is never black and white, brands should understand that there are several types of attribution models. They can each be leveraged depending on what’s being measured.