Customers go through an entire journey before the conversion occurs on a web shop. Within this journey, we especially want to know: which moment was responsible for the conversion? This is where we want to spend the most marketing budget. To find this out, we need an attribution model.
The importance of an attribution model
H1, a partner of ours that we work with more often, asked us to explain more about attribution models. Monika and Angelique Nanning of H1 discuss how an omnichannel attribution model can determine which channels work best. And with that: how the marketing budget should be distributed.
The attribution model explained simply
The definition of attribution model is "a system that determines the value of a given action." We explain this using the following example:
Suppose a marriage is the ultimate conversion is in a customer journey. Several moments led to the marriage, these are touch points. The couple met through a dating app. The first date took place in a restaurant. And with an engagement ring, a marriage proposal took place. The question is, who is responsible for the marriage?
The app developer will say it was him, because thanks to the app they got to know each other. The restaurant owner says the successful first date at his restaurant led to a relationship. But without an engagement ring, no engagement according to the jeweler. Conclusion? They are all right.
The different attribution models
Within e-commerce, there are omnichannel attribution models to answer the question of which online touchpoints were responsible for conversion. We briefly explain them.
- Last interaction: in this basic model, the last touchpoint that comes before the conversion gets all the credit for the conversion
- First interaction: in this model, the first touchpoint gets all the credit.
- Linear: here the credit is shared equally across all touchpoints.
- Time decay: the touchpoint closest to conversion gets the most credit. Below is a possible distribution.
- Position-based: 40% of the credit goes to the first and last touchpoint and 20% goes to the touchpoints in between.
- Datadriven: our favorite! This model uses data from a web shop. The different customer journeys are compared to identify the touchpoints that are most likely to cause conversion. Those touchpoints get more credits, which in turn prove that it is important to invest in them.
Make a choice and analyze
All online activities are ultimately aimed atto achieve conversion. From emails and social media messages to display banners and shopping ads. A banner campaign, according to the final interaction model may have meant nothing to theconversion, at the same time a different attribution model gives a different outcome.
It is important to determine which attribution model(s) are most important for each webshop. Discover the best performing touchpoints and allocate the marketing budget with this knowledge.
