Magic attribution is based on the MagicScript algorithm with an enhanced Magic Linear attribution model. Magic Linear moves away from the standard approach to PPC campaign management and shows the conversion path in a completely different way.
Magic attribution is based on the MagicScript algorithm with an enhanced Magic Linear attribution model. Magic Linear changes the classic view on the management of PPC campaigns and shows the conversion path in a completely different way.
Previous
Next
How does the linear
attribution model work?
Google Ads
Advertisements
20%
Google Search
Organic
20%
Facebook Ads
Advertisements
20%
Webinar
Outbound
20%
Web visit
Direct
20%
The linear attribution model ascribes conversions and their values equally to all the sources of the conversion path.
The attribution model is a rule or an array of rules, based on which conversions and their value are ascribed to individual touchpoints (Touchpoints are the individual interactions with the sources the customer came in contact with during the conversion path, this can be, for example, paid clicks, unpaid clicks, visits, impressions and so on.) in the customer’s conversion path.
We chose the linear model because we think its way of adding conversions and their values is the most appropriate and fair. The linear attribution model assigns conversions and their value to all resources that participated in the conversion path. By testing, we found that the Linear attribution model is much cheaper to calculate than the “Markov model” and at the same time equally suitable for 95% of projects.
- It’s core is the Linear attribution model
- We use a 30 day look-back window
- We exclude Direct and “Brand” visits
It works as a Cross channel attribution model. That means that unlike the attribution models available in Google Ads, it uses data not only from Google cpc, but from all the sources from which data flows into Google Analytics.
Attribution models
comparison
Attribution models comparison
Inner channel attribution models (works only with clicks from G ads)
Several attribution models
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods
Several attribution models
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods
Inner channel attribution models (works only with clicks from G ads)
Several attribution models
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods
Several attribution models
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods
Cross channel attribution models (works with data from multiple sources)
One attribution model, Last Non-Direct
Click, other only in Comparison tool
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods
One attribution model, Last Non-Direct
Click, other only in Comparison tool
Needs to collect historical data to be able to work with margin
Needs margin data at the time of the order
Can work with cancellations and returned goods