What is an attribution model?
Attribution is the process of assigning credit to steps along a customer journey that has led to a conversion. There are two attribution models which determine what should be credited based on following a set of rules or by observing data; rule-based models and a data-driven model.
The rule-based attribution models disregard the conversion type or user behaviour and focus on fixed rules. There are five models available in Attribution in Google Analytics:
- Last click – The last click model provides all the credit of the conversion to the event which was last clicked.
- First click – The first click model is the opposite to the last click and provides all the credit of the conversion to the event which was first clicked.
- Linear – The linear model shares out the credit of the conversion equally to all of the clicks along the path.
- Time decay – The time decay model provides credit of the conversion based on when the click took place. The closer in time to the conversion, the more credit the click will receive. Using a 7-day half-life, the credit is then distributed. This means a click which happened 8 days before the conversion will receive half the amount of credit of the click that happened a day before the conversion.
- Position-based – The position-based model provides 40% of the credit to the first click and the last click and then shares the remaining 20% across the other clicks along the path.
The data-driven attribution model is different to the rule-based attribution models. It requires a certain amount of data and uses machine learning to observe your account’s data and calculates the actual contribution each click provided to the conversion. Because of this, data-driven attribution models are specific to each advertiser and conversion type.
Attribution and Multi-Channel funnels
The new Attribution feature is separate to the Multi-Channel Funnels which is already on Google Analytics. Depending on your attribution needs, one of the features may be a better fit and Google have compared the two for you to decide which is best for you.
By taking advantage of the Attribution feature, you will be able to:
- Accurately report conversion totals which have been de-duplicated across all digital channels
- Have a consolidated, consistent view of all digital performance
- Build an understanding of your customer journey
Creating an Attribution project in Google Analytics
It’s simple to create an Attribution project in Google Analytics and Google also have a step-by-step guide which you can follow. Once set up, it takes 72 hours for Google to collect data to create the first model. Once it has done this, it will then take 30 days to create a more robust model. During this time, your reports will show results based on the data that has been collected. You can then display the results on a complete training window once all the data has been collected.
At bolt, we will be keeping a close eye on the new beta feature to see how it can provide us with more insightful data which can be used to improve both ours and our clients’ campaigns.