Attribution may be one of the main challenges for ecommerce businesses since it’s directly related to budget allocation. Let’s explore the main challenges, the origins and some available solutions.
Measurement attribution, marketing attribution all refer to the concepnt in which digital advertisers have to decide which channel deserves credit for one purhcase or sales.
In our modern digital environment, marketing platforms essentially do choose that for us. Which means that if you are using any digital analytics solution or you get a report of performance in a media platform (Google, Facebook), these platform have by default a attribution methodology chosen for you.
What is an attribution methodology?
In short, it is the logic by which you assign the credit (or part of it) to a marketing channel for a single conversion. Easy, right?
So if you think that all the credit for a purchase has to go to the end of the funnel (the last interaction a customer had before purchasing), then your attribution methodology is last-click.
There are two types of attribution methodology: the “pre-defined” ones (also called rule-based or heuristic attribution) and the data-driven ones (also called ensemble).
Now, what’s the difference? In rule-based methods, you pre-define how you want to give the credit for all the purchases. So if you choose first-click attribution, that means all of your “Views” in the digital analytics solution are going to represent this decision.
In data-driven attribution, it’s different. The system takes the given data into consideration and generates a result based on the logic of your model. Fascinating, right?
What ecommerce usually choose for their attribution method?
The short answer is: last-click attribution. The vast majority of e-commerce businesses choose last-click attribution.
Is Last-click attribution bad?
No, it’s not. It’s definitely confusing, but it’s not bad. It’s just a method. For example, if you think that users shop in a linear way, then you should trust last-click. If you think that the consumer journey is more complex, then you should keep reading.
Consumer journey is not linear.
How we think users shop
How consumers actually shop
Traditionally, the industry — and of course marketing practitioners — thought that consumers transitioned from one stage of purchase to another in a linear way. So someone who was not aware of a product or service would watch an ad or read an article. A proportion of this cohort would get convinced, visit the website, and request a price quote. Similarly, a proportion of those would be convinced to proceed with their purchases immediately (the same purchase funnel).
That may have been the case 20 years ago.
Today we know that’s not the case. Users are constantly reviewing products, creating needs for themselves, and indeed being influenced by ads — but not in a linear way. It’s much more complex. Google published a very long and comprehensive piece of research on how we shop (Rennie & Protheroe, 2020). According to them, the way people make decisions is messy.
What e-commerce businesses need
Well, the short answer is that e-commerce businesses in the modern world require data-driven attribution. They have to account for the complexity of the consumer journey, but they also have to account for the increasing influence of view-first channels. It is evident that platforms like TikTok, YouTube, Snapchat, and Pinterest have won consumers’ hearts. A common challenge in interpreting the impact of those channels is the lack of clicks and, as a result, the lack of website or app visits.
Therefore, we have developed a method that not only takes into consideration the complexity of modern consumer journeys but also, very importantly, accounts for impressions. Essentially, we consider an impression or a video view as a touchpoint and, with the support of statistical modeling, we are able to identify the true impact of each impression or video view.
Are you interested in having a look at our platform and gaining access? Well, you just have to email us at info@mar-sci.com.
Get access to our open source marketing analytics platform.
References:
- How people decide what to buy lies in the ‘messy middle’ of the purchase journey, Alistair Rennie, Jonny Protheroe, July 2020
- Berman, B., 2020. Paths to purchase: the seven steps of customer purchase journey mapping. Rutgers Business Review, 5(1), pp.84-100.
